ceph/doc/changelog/v10.2.8.txt

   1 commit f5b1f1fd7c0be0506ba73502a675de9d048b744e
   2 Author: Jenkins Build Slave User <ceph-release-team@redhat.com>
   3 Date:   Thu Jul 6 14:56:18 2017 +0000
   4
   5     10.2.8
   6
   7 commit 66dbf9beef04988dbd3653591e51afa6d84e3990
   8 Merge: 2f491b2e5e f46ccf2cb4
   9 Author: Nathan Cutler <presnypreklad@gmail.com>
  10 Date:   Tue Jul 4 17:43:57 2017 +0200
  11
  12     Merge pull request #14710 from smithfarm/wip-start-race
  13
  14     tests: rados: sleep before ceph tell osd.0 flush_pg_stats after restart
  15
  16     Reviewed-by: Sage Weil <sage@redhat.com>
  17     Reviewed-by: Kefu Chai <kchai@redhat.com>
  18     Reviewed-by: David Zafman <dzafman@redhat.com>
  19
  20 commit 2f491b2e5e8b1f340b28415c3bd3d9628603c377
  21 Merge: 552a573f84 a372b4eca1
  22 Author: Yuri Weinstein <yuri.weinstein@gmail.com>
  23 Date:   Tue Jul 4 07:31:50 2017 -0700
  24
  25     Merge pull request #16089 from ceph/wip_fix_point_jewel
  26
  27     qa/Fixed upgrade sequence to 10.2.0 -> 10.2.7 -> latest -x (10.2.8)
  28
  29     Reviewed-by: Nathan Cutler <ncutler@suse.com>
  30
  31 commit 552a573f8426ecfec1a0df21a6c3941afd4e460c
  32 Merge: 53a3be7261 55eeaadfc4
  33 Author: Yuri Weinstein <yuri.weinstein@gmail.com>
  34 Date:   Mon Jul 3 17:23:43 2017 -0700
  35
  36     Merge pull request #16088 from smithfarm/wip-fix-client-upgrade-centos
  37
  38     tests: run upgrade/client-upgrade on latest CentOS 7.3
  39
  40     Reviewed-by: Yuri Weinstein <yweins@redhat.com>
  41
  42 commit a372b4eca1f25647541943918ae737f20783db11
  43 Author: Yuri Weinstein <yweinste@redhat.com>
  44 Date:   Mon Jul 3 14:18:14 2017 -0700
  45
  46     Fixed upgrade sequence to 10.2.0 -> 10.2.7 -> latest -x (10.2.8)
  47
  48     Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
  49
  50 commit 55eeaadfc4025c83cb63c951265710868df0325f
  51 Author: Nathan Cutler <ncutler@suse.com>
  52 Date:   Mon Jul 3 22:55:21 2017 +0200
  53
  54     tests: run upgrade/client-upgrade on latest CentOS 7.3
  55
  56     Before this patch, all centos jobs were failing because there are no longer any
  57     CentOS 7.2 machines in Sepia.
  58
  59     Signed-off-by: Nathan Cutler <ncutler@suse.com>
  60
  61 commit 53a3be7261cfeb12445fbdba8238eefa40ed09f5
  62 Merge: 84bd162978 d33b30cdb0
  63 Author: Nathan Cutler <presnypreklad@gmail.com>
  64 Date:   Fri Jun 30 16:43:38 2017 +0200
  65
  66     Merge pull request #15504 from Vicente-Cheng/wip-20151-jewel
  67
  68     jewel: ceph-disk: do not setup_statedir on trigger
  69
  70     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
  71
  72 commit 84bd162978e48eead40335bcbd92e4ab18e9c590
  73 Merge: c710689109 8e0e4a0ce7
  74 Author: Nathan Cutler <presnypreklad@gmail.com>
  75 Date:   Wed Jun 28 10:20:23 2017 +0200
  76
  77     Merge pull request #15904 from smithfarm/wip-20413-jewel
  78
  79     jewel: tests: upgrade:hammer-x/stress-split-erasure-code-x86_64 fails in 10.2.8 integration testing
  80
  81     Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
  82
  83 commit c7106891096c895f0cc5c2cef438078ea48de95d
  84 Merge: 5c6cb14806 9d3110c276
  85 Author: Nathan Cutler <presnypreklad@gmail.com>
  86 Date:   Wed Jun 28 08:16:33 2017 +0200
  87
  88     Merge pull request #14930 from smithfarm/wip-19829-jewel
  89
  90     jewel: tests: New upgrade test for #14930
  91
  92     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
  93
  94 commit 5c6cb1480699f5ce464e25b9cacdda770ce3660d
  95 Merge: bdc085d02a d43e19d886
  96 Author: Nathan Cutler <presnypreklad@gmail.com>
  97 Date:   Wed Jun 28 08:16:04 2017 +0200
  98
  99     Merge pull request #14392 from asheplyakov/19508-jewel
 100
 101     jewel: osd: pg_pool_t::encode(): be compatible with Hammer <= 0.94.6
 102
 103     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 104
 105 commit 9d3110c276917055b078cd14c181b2bda2625821
 106 Author: Nathan Cutler <ncutler@suse.com>
 107 Date:   Sun Jun 25 10:32:16 2017 +0200
 108
 109     tests: upgrade/hammer-x/v0-94-6-mon-overload: tweak packages list
 110
 111     Include some hammer dependencies that aren't in the jewel default packages
 112     list, and exclude some java packages that may not be in the hammer repo and are
 113     not needed for the upgrade test in any case.
 114
 115     N.B.: This cannot be cherry-picked from master because upgrade/hammer-x was
 116     dropped in master.
 117
 118     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 119
 120 commit 6a64f8901bb3b218a8dc58b11d6c13033d45f067
 121 Author: Nathan Cutler <ncutler@suse.com>
 122 Date:   Wed May 3 11:39:27 2017 +0200
 123
 124     tests: upgrade/hammer-x: new v0-94-6-mon-overload subsuite
 125
 126     This is not a cherry-pick from master because direct upgrades
 127     from hammer to kraken+ are not supported.
 128
 129     Fixes: http://tracker.ceph.com/issues/19829
 130     References: http://tracker.ceph.com/issues/19508
 131     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 132
 133 commit bdc085d02ab9723f6b90b6a3047bc51cf224b930
 134 Merge: e41ae4a7b1 d2d4b7202d
 135 Author: Nathan Cutler <presnypreklad@gmail.com>
 136 Date:   Tue Jun 27 12:41:59 2017 +0200
 137
 138     Merge pull request #15936 from batrick/i20412
 139
 140     qa: enable quotas for pre-luminous quota tests
 141
 142     Reviewed-by: John Spray <john.spray@redhat.com>
 143     Reviewed-by: Nathan Cutler <ncutler@suse.com>
 144
 145 commit d2d4b7202d77e5696eb18c4da4f7d614116ced36
 146 Author: Patrick Donnelly <pdonnell@redhat.com>
 147 Date:   Mon Jun 26 19:04:48 2017 -0700
 148
 149     qa: enable quotas for pre-luminous quota tests
 150
 151     This cannot be cherry-picked from master because the config option is removed
 152     since 0f250a889dba2100d3afcea0a18e4f6a8d086b86.
 153
 154     Fixes: http://tracker.ceph.com/issues/20412
 155
 156     Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
 157
 158 commit e41ae4a7b1b9cc4394473e21f6e6d6ef9cab1d59
 159 Merge: a21af3b7a1 682b4d717c
 160 Author: Sage Weil <sage@newdream.net>
 161 Date:   Mon Jun 26 21:24:11 2017 -0500
 162
 163     Merge pull request #15933 from smithfarm/wip-hammer-jewel-x
 164
 165     jewel: tests: drop upgrade/hammer-jewel-x
 166
 167 commit 682b4d717c96b516c315a01b1174af3503dedba6
 168 Author: Nathan Cutler <ncutler@suse.com>
 169 Date:   Tue Jun 27 02:27:22 2017 +0200
 170
 171     tests: drop upgrade/hammer-jewel-x
 172
 173     This suite doesn't have any test logic in it. Its existence in the jewel branch
 174     appears to be an oversight.
 175
 176     This cannot be cherry-picked from master because the upgrade/hammer-jewel-x
 177     suite is present (and justified) in master and is not currently being dropped
 178     there.
 179
 180     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 181
 182 commit a21af3b7a1ea5abfa4f344800e413d6249824204
 183 Merge: 615a6ab9b7 a744340790
 184 Author: John Spray <jspray@redhat.com>
 185 Date:   Mon Jun 26 16:25:12 2017 -0400
 186
 187     Merge pull request #15438 from Vicente-Cheng/wip-20027-jewel
 188
 189     jewel: mds: issue new caps when sending reply to client
 190
 191     Reviewed-by: John Spray <john.spray@redhat.com>
 192
 193 commit 615a6ab9b723d204c6de567750fe9450742fcedb
 194 Merge: 9b13b48b3e 8ac0e5c363
 195 Author: John Spray <jspray@redhat.com>
 196 Date:   Mon Jun 26 16:23:48 2017 -0400
 197
 198     Merge pull request #15000 from jan--f/wip-19846-jewel
 199
 200     jewel: cephfs: normalize file open flags internally used by cephfs
 201
 202     Reviewed-by: John Spray <john.spray@redhat.com>
 203
 204 commit 9b13b48b3ed919340789a41d065eb4a9a27110de
 205 Merge: d217da1742 b429fa1807
 206 Author: Nathan Cutler <presnypreklad@gmail.com>
 207 Date:   Mon Jun 26 18:25:03 2017 +0200
 208
 209     Merge pull request #15383 from asheplyakov/20014-bp-jewel
 210
 211     jewel: cls/rgw: list_plain_entries() stops before bi_log entries
 212
 213     Reviewed-by: Orit Wasserman <owasserm@redhat.com>
 214
 215 commit d217da174252f9126d530868aa472230ecba31ca
 216 Merge: e520040ece 4028774122
 217 Author: Zack Cerza <zack@cerza.org>
 218 Date:   Mon Jun 26 10:17:03 2017 -0600
 219
 220     Merge pull request #15870 from smithfarm/wip-swift-task-move-jewel
 221
 222     tests: move swift.py task from teuthology to ceph, phase one (jewel)
 223
 224 commit e520040ecec756ce181f716dad1c0bad41c77a7d
 225 Merge: dde8656e6b 1c0c9093ab
 226 Author: Nathan Cutler <presnypreklad@gmail.com>
 227 Date:   Mon Jun 26 09:24:04 2017 +0200
 228
 229     Merge pull request #15842 from smithfarm/wip-sortbitwise-jewel
 230
 231     qa/suites/upgrade/hammer-x: set "sortbitwise" for jewel clusters
 232
 233     Reviewed-by: Kefu Chai <kchai@redhat.com>
 234
 235 commit dde8656e6b5e7ffe66a6fd695cbc17dfb18fb43e
 236 Merge: 498c96e66a 06cf9f3650
 237 Author: John Spray <jspray@redhat.com>
 238 Date:   Sun Jun 25 19:59:40 2017 -0400
 239
 240     Merge pull request #15468 from smithfarm/wip-20140-jewel
 241
 242     jewel: cephfs: Journaler may execute on_safe contexts prematurely
 243
 244     Reviewed-by: John Spray <john.spray@redhat.com>
 245
 246 commit 4028774122954023265d7825fbf9e91dc526fdee
 247 Author: Nathan Cutler <ncutler@suse.com>
 248 Date:   Sun Jun 25 12:42:36 2017 +0200
 249
 250     tests: swift.py: tweak imports
 251
 252     The ".." form only works within the teuthology repo. With swift.py now in the
 253     Ceph repo, we have to be explicit.
 254
 255     Error message was: "ValueError: Attempted relative import beyond toplevel
 256     package
 257
 258     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 259
 260 commit 8e0e4a0ce7489542f47522e0a5161a5bf123c744
 261 Author: Nathan Cutler <ncutler@suse.com>
 262 Date:   Sun Jun 25 10:27:58 2017 +0200
 263
 264     tests: upgrade/hammer-x/stress-split: tweak packages list
 265
 266     Include some hammer dependencies that aren't in the jewel default packages
 267     list, and exclude some java packages that may not be in the hammer repo and are
 268     not needed for the upgrade test in any case.
 269
 270     N.B.: This cannot be cherry-picked from master because upgrade/hammer-x was
 271     dropped in master.
 272
 273     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 274
 275 commit a86ce728954a765797ce634025d43650d990e480
 276 Author: Nathan Cutler <ncutler@suse.com>
 277 Date:   Sun Jun 25 09:27:47 2017 +0200
 278
 279     tests: swift.py: clone the ceph-jewel branch
 280
 281     The master branch of ceph/swift.git contains tests that are incompatible with
 282     Jewel and Hammer. The ceph-jewel branch omits these tests.
 283
 284     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 285
 286 commit 498c96e66a91edc8bd614cfc8fc5a14b3d210a76
 287 Merge: 38af498f9f cda721bbbf
 288 Author: John Spray <jspray@redhat.com>
 289 Date:   Fri Jun 23 08:02:48 2017 -0400
 290
 291     Merge pull request #15472 from smithfarm/wip-20148-jewel
 292
 293     jewel: mds: Too many stat ops when trying to probe a large file
 294
 295     Reviewed-by: John Spray <john.spray@redhat.com>
 296
 297 commit 3d5b489369bb2cecccb1f36347654c0a37069d1c
 298 Merge: 38af498f9f 7b58ac97e9
 299 Author: Nathan Cutler <ncutler@suse.com>
 300 Date:   Fri Jun 23 08:35:27 2017 +0200
 301
 302     Merge branch 'master' of /home/smithfarm/src/ceph/upstream/teuthology into wip-swift-task-move-jewel
 303
 304 commit 7b58ac97e9dd195f4170e9e0ea00bae76d1f3ccd
 305 Author: Nathan Cutler <ncutler@suse.com>
 306 Date:   Fri Jun 23 08:27:42 2017 +0200
 307
 308     tests: move swift.py task to qa/tasks
 309
 310     In preparation for moving this task from ceph/teuthology.git into ceph/ceph.git
 311
 312     The move is necessary because jewel-specific changes are needed, yet teuthology
 313     does not maintain a separate branch for jewel. Also, swift.py is a
 314     Ceph-specific task so it makes more sense to have it in Ceph.
 315
 316     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 317
 318 commit 38af498f9f7b62f9f851364ae7f2691832423198
 319 Merge: d0ae1de51f aa0cd461df
 320 Author: Nathan Cutler <presnypreklad@gmail.com>
 321 Date:   Thu Jun 22 22:07:23 2017 +0200
 322
 323     Merge pull request #15529 from badone/wip-async-sleep-timer-fix-jewel
 324
 325     jewel:  osd: Implement asynchronous scrub sleep
 326
 327     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 328
 329 commit 1c0c9093ab913a82c1dc5656a54b4009bdc35c9c
 330 Author: Nathan Cutler <ncutler@suse.com>
 331 Date:   Thu Jun 22 11:32:42 2017 +0200
 332
 333     qa/suites/upgrade/hammer-x: set "sortbitwise" for jewel clusters
 334
 335     Inspired by 3734280522a913ca8340ebc98b80978f63bade6f
 336
 337     This cannot be cherry-picked from master because master does not have
 338     qa/suites/upgrade/hammer-x
 339
 340     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 341
 342 commit d0ae1de51f5faf26a2f4b0d5b7f494a4923f870d
 343 Merge: 64c011a8c4 de76fdbb9f
 344 Author: Kefu Chai <tchaikov@gmail.com>
 345 Date:   Thu Jun 22 11:59:51 2017 +0800
 346
 347     Merge pull request #15824 from tchaikov/jewel
 348
 349     qa/workunits/rados/test-upgrade-*: whitelist tests the right way
 350
 351     Reviewed-by: Sage Weil <sage@redhat.com>
 352
 353 commit de76fdbb9f435652e2c15326d00b01d26ab007a7
 354 Author: Kefu Chai <kchai@redhat.com>
 355 Date:   Thu Jun 22 08:06:43 2017 +0800
 356
 357     qa/workunits/rados/test-upgrade-*: whitelist tests the right way
 358
 359     --gtest_filter=POSTIVE_PATTERNS[-NEGATIVE_PATTERNS], so we cannot add
 360     multiple exclusive patterns using -pattern:-pattern, instead, we should
 361     use: -pattern:pattern
 362
 363     Signed-off-by: Kefu Chai <kchai@redhat.com>
 364     Conflicts:
 365             qa/workunits/rados/test-upgrade-v11.0.0.sh: this change is not
 366         cherry-picked from master, because the clone-range op was removed
 367         from master. and only supported in pre-luminous releases.
 368
 369 commit 64c011a8c4af27dc095b1a9190ccf1ca76d2cc8f
 370 Merge: e8da5e376f ab78cd040f
 371 Author: Nathan Cutler <presnypreklad@gmail.com>
 372 Date:   Tue Jun 20 22:54:00 2017 +0200
 373
 374     Merge pull request #14661 from smithfarm/wip-19575-jewel
 375
 376     jewel: rgw: unsafe access in RGWListBucket_ObjStore_SWIFT::send_response()
 377
 378     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 379
 380 commit e8da5e376fc426f85fcab84a5ae71b3c17ed0068
 381 Merge: 5a1e849ecf 1af6781d3c
 382 Author: Kefu Chai <tchaikov@gmail.com>
 383 Date:   Tue Jun 20 22:42:47 2017 +0800
 384
 385     Merge pull request #15778 from tchaikov/wip-upgrade-without-clone-range-jewel
 386
 387     qa/workunits/rados/test-upgrade-*: whitelist tests for master
 388
 389     Reviewed-by: Sage Weil <sage@redhat.com>
 390
 391 commit d43e19d88692bd318f0569559867df919c26d8db
 392 Author: Alexey Sheplyakov <asheplyakov@mirantis.com>
 393 Date:   Fri Apr 7 12:34:20 2017 +0400
 394
 395     jewel: osd: pg_pool_t::encode(): be compatible with Hammer <= 0.94.6
 396
 397     This patch is necessary for Jewel only since direct upgrades from Hammer
 398     to Kraken and newer are not supported.
 399
 400     Fixes: http://tracker.ceph.com/issues/19508
 401
 402     Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
 403
 404 commit 1af6781d3c60421930087d31124e62cae530ca24
 405 Author: Kefu Chai <kchai@redhat.com>
 406 Date:   Tue Jun 20 19:49:14 2017 +0800
 407
 408     qa/workunits/rados/test-upgrade-*: whitelist tests for master
 409
 410     The jewel-x upgrade test now runs this script against a mixed cluster on
 411     a machine with code from master installed.  That means we have to
 412     skip any new tests that will fail on a mixed cluster. CloneRange was
 413     removed in 0d7b0b7.
 414
 415     Signed-off-by: Kefu Chai <kchai@redhat.com>
 416     Conflicts:
 417         qa/workunits/rados/test-upgrade-v11.0.0.sh: this change is not
 418     cherry-picked from master, because the clone-range op was removed from
 419     master. and only supported in pre-luminous releases.
 420
 421 commit a74434079088129244b7aae6ccc6df7094282eba
 422 Author: Yan, Zheng <zyan@redhat.com>
 423 Date:   Sat Apr 22 12:27:12 2017 +0800
 424
 425     mds: issue new caps when sending reply to client
 426
 427     After Locker::issue_new_caps() adds new Capability data struct,
 428     do not issue caps immediately. Let CInode::encode_inodestate()
 429     do the job instead. This can avoid various races that early reply
 430     is not allowed, caps that haven't been sent to client gets revoked.
 431
 432     Fixes: http://tracker.ceph.com/issues/19635
 433     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 434     (cherry picked from commit 799703a4acb49db0b6cc99a23e4326767e694c3a)
 435
 436 commit d33b30cdb019937ff88f9724599f52f4e00d37cf
 437 Author: Loic Dachary <ldachary@redhat.com>
 438 Date:   Thu Jun 1 11:37:20 2017 +0200
 439
 440     ceph-disk: do not setup_statedir on trigger
 441
 442     trigger may run when statedir is unavailable and does not use it.
 443
 444     Fixes: http://tracker.ceph.com/issues/19941
 445
 446     Signed-off-by: Loic Dachary <loic@dachary.org>
 447     (cherry picked from commit 16bfbdd3d9988523bba31aace516c303057daa58)
 448
 449 commit f46ccf2cb4701cd93cd9b15a4e57b5b97798b947
 450 Author: Nathan Cutler <ncutler@suse.com>
 451 Date:   Fri Apr 21 11:05:05 2017 +0200
 452
 453     tests: rados: sleep before ceph tell osd.0 flush_pg_stats after restart
 454
 455     Even though we wait for HEALTH_OK after restarting the daemons, they are not
 456     ready to respond to flush_pg_stats.
 457
 458     The reason why the osd is not ready for "tell" command after "ceph health"
 459     shows that the cluster is "HEALTH_OK" is that the monitor fails to be notified
 460     that the osd in question is not up in "heatbeat_interval". Because infernalis
 461     does not have the osd_fast_fail_on_connection_refused support, the monitor
 462     needs longer to detect that an osd is down, and osd_heartbeat_grace is used to
 463     determine if an osd is down.
 464
 465     References: http://tracker.ceph.com/issues/16239
 466     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 467     Signed-off-by: Kefu Chai <kchai@redhat.com>
 468
 469 commit ab78cd040f6d3946ed40b6638ebcf52969a7cbb6
 470 Author: Yehuda Sadeh <yehuda@redhat.com>
 471 Date:   Wed Mar 8 14:52:34 2017 -0800
 472
 473     rgw: fix crash when listing objects via swift
 474
 475     Fixes: http://tracker.ceph.com/issues/19249
 476
 477     Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
 478     (cherry picked from commit a9ec5e8ce184e19c009863db4d3519f9d8af91bd)
 479
 480     Conflicts:
 481             src/rgw/rgw_rest_swift.cc ("key" element of RGWObjEnt struct
 482                is not a reference; fix)
 483
 484     (cherry picked from commit 92b35155ff7b7492f3c50bf4f2ff0ffef2ca1c55)
 485
 486 commit 5a1e849ecf215d82e31b9bdd0970cb04200de2c9
 487 Merge: 2469085d57 66c3db7aee
 488 Author: Nathan Cutler <presnypreklad@gmail.com>
 489 Date:   Mon Jun 19 22:46:47 2017 +0200
 490
 491     Merge pull request #14752 from cbodley/wip-19474
 492
 493     jewel: rgw: allow system users to read SLO parts
 494
 495     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 496
 497 commit cda721bbbfae00ec4244718ae20cbd9ae914c630
 498 Author: Yan, Zheng <zyan@redhat.com>
 499 Date:   Fri May 19 09:37:15 2017 +0800
 500
 501     client: update the 'approaching max_size' code
 502
 503     The old 'approaching max_size' code expects MDS set max_size to
 504     '2 x reported_size'. This is no longer true. The new code reports
 505     file size when half of previous max_size increment has been used.
 506
 507     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 508     (cherry picked from commit 9316b0442c6f828dcf8da952e4c7a63c4db1398d)
 509
 510     Conflicts:
 511             src/client/Client.cc - in jewel, second argument to check_caps() is
 512                 a bool (see 0df562a8e13 which is not in jewel)
 513
 514 commit 439f39128ec278ce78139d4b96ed098c68efa3f5
 515 Author: Yan, Zheng <zyan@redhat.com>
 516 Date:   Wed May 17 19:08:37 2017 +0800
 517
 518     mds: limit client writable range increment
 519
 520     For very large file, setting the writable range to '2 * file_size'
 521     causes file recovery to run a long time. To recover a 1T file, Filer
 522     needs to probe 2T~1T range.
 523
 524     Fixes: http://tracker.ceph.com/issues/19955
 525     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 526     (cherry picked from commit 538f35bef944b18e9bca2b15ed7f4e8807ef0554)
 527
 528     Conflicts:
 529             src/mds/Locker.h - in jewel, file_update_finish() has different
 530                 arguments than it does in master
 531
 532 commit 06cf9f365033f7913051bdf4060f0bc6fc0444d7
 533 Author: Yan, Zheng <zyan@redhat.com>
 534 Date:   Tue May 23 21:46:54 2017 +0800
 535
 536     osdc/Journaler: avoid executing on_safe contexts prematurely
 537
 538     Journaler::_do_flush() can skip flushing some data when prezered
 539     journal space isn't enough. Before updating Journaler::next_safe_pos,
 540     we need to check if Journaler::_do_flush() has flushed enough data.
 541
 542     Fixes: http://tracker.ceph.com/issues/20055
 543     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 544     (cherry picked from commit 6511e7a9e35a14216c03cd6921ca4d232274f953)
 545
 546 commit 2e299b50de4a297fee2aec21290632336d239857
 547 Author: Yan, Zheng <zyan@redhat.com>
 548 Date:   Wed Apr 12 16:00:18 2017 +0800
 549
 550     osdc/Journaler: make header write_pos align to boundary of flushed entry
 551
 552     This can speed up the process that detects and drops partial written
 553     entry in the log tail.
 554
 555     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 556     (cherry picked from commit 8ae2962b79903e217fda83cea4140af64b5d6883)
 557
 558     Conflicts:
 559             src/osdc/Journaler.cc - 8d4f6b92cba is not being backported to jewel
 560             src/osdc/Journaler.h - Journaler::Journaler initializer list is different in jewel, compared to master
 561
 562 commit 2469085d57a05933589165f6f99a67b2e28c7022
 563 Merge: 42c3fbc129 d57437e338
 564 Author: John Spray <jspray@redhat.com>
 565 Date:   Wed Jun 14 10:01:27 2017 -0400
 566
 567     Merge pull request #14672 from smithfarm/wip-19334-jewel
 568
 569     jewel: MDS heartbeat timeout during rejoin, when working with large amount of caps/inodes
 570
 571     Reviewed-by: John Spray <john.spray@redhat.com>
 572
 573 commit 8ac0e5c363bd6439071d26874b6714cf2376736f
 574 Author: Yan, Zheng <zyan@redhat.com>
 575 Date:   Fri May 12 10:38:51 2017 +0800
 576
 577     pybind: fix cephfs.OSError initialization
 578
 579     Traceback (most recent call last):
 580       File "<stdin>", line 1, in <module>
 581       File "cephfs.pyx", line 672, in cephfs.LibCephFS.open (/home/zhyan/Ceph/ceph-2/build/src/pybind/cephfs/pyrex/cephfs.c:10160)
 582       File "cephfs.pyx", line 155, in cephfs.OSError.__init__ (/home/zhyan/Ceph/ceph-2/build/src/pybind/cephfs/pyrex/cephfs.c:1889)
 583     TypeError: __init__() takes exactly 3 positional arguments (2 given)
 584
 585     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 586     (cherry picked from commit e6493f64ba4592b8dca54ece4464efa6c7f331a7)
 587
 588 commit 09b9410c2e69a466b001d92fc14eb44d768009f1
 589 Author: Yan, Zheng <zyan@redhat.com>
 590 Date:   Wed May 10 08:13:52 2017 +0800
 591
 592     pybind: fix open flags calculation
 593
 594     (O_WRONLY | O_RDWR) is invaild open flags
 595
 596     Fixes: http://tracker.ceph.com/issues/19890
 597     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
 598     (cherry picked from commit 2c25c99cb4572ffae97555a56b24a4c4097dcdec)
 599
 600 commit 42c3fbc129cbb60d447c1a6b0402a9def1656446
 601 Merge: 71d45e1905 7347f11939
 602 Author: John Spray <jspray@redhat.com>
 603 Date:   Wed Jun 14 09:43:05 2017 -0400
 604
 605     Merge pull request #14677 from smithfarm/wip-19665-jewel
 606
 607     jewel: mds: C_MDSInternalNoop::complete doesn't free itself
 608
 609     Reviewed-by: John Spray <john.spray@redhat.com>
 610
 611 commit 71d45e190528124a8ff7e4674f7bfb7c340f80ee
 612 Merge: 27c915f5d2 e6daee8a9f
 613 Author: John Spray <jspray@redhat.com>
 614 Date:   Wed Jun 14 09:35:08 2017 -0400
 615
 616     Merge pull request #15466 from smithfarm/wip-19762-jewel
 617
 618     jewel: cephfs: non-local quota changes not visible until some IO is done
 619
 620     Reviewed-by: John Spray <john.spray@redhat.com>
 621
 622 commit 27c915f5d21bd84502e35eb269d955fafc47de0b
 623 Merge: a76357622a db053da618
 624 Author: John Spray <jspray@redhat.com>
 625 Date:   Wed Jun 14 09:32:47 2017 -0400
 626
 627     Merge pull request #14700 from smithfarm/wip-19709-jewel
 628
 629     jewel: mds: enable start when session ino info is corrupt
 630
 631     Reviewed-by: John Spray <john.spray@redhat.com>
 632
 633 commit a76357622a2773850153aa1f6ea02b1737942c4b
 634 Merge: 85aab833f7 db86a24e79
 635 Author: John Spray <jspray@redhat.com>
 636 Date:   Wed Jun 14 09:31:22 2017 -0400
 637
 638     Merge pull request #14685 from smithfarm/wip-19675-jewel
 639
 640     jewel: cephfs: Test failure: test_data_isolated (tasks.cephfs.test_volume_client.TestVolumeClient)
 641
 642     Reviewed-by: John Spray <john.spray@redhat.com>
 643
 644 commit 85aab833f7f66ac81d8b12f4203fa215787f8d2a
 645 Merge: f6b395115e 7b9283beec
 646 Author: John Spray <jspray@redhat.com>
 647 Date:   Wed Jun 14 09:30:53 2017 -0400
 648
 649     Merge pull request #14684 from smithfarm/wip-19673-jewel
 650
 651     jewel: cephfs: mds is crushed, after I set about 400 64KB xattr kv pairs to a file
 652
 653     Reviewed-by: John Spray <john.spray@redhat.com>
 654
 655 commit f6b395115e4de15d73269ff6b96f2ee0fd0ea9c3
 656 Merge: ff60fceb5f b52c508861
 657 Author: John Spray <jspray@redhat.com>
 658 Date:   Wed Jun 14 09:30:13 2017 -0400
 659
 660     Merge pull request #14683 from smithfarm/wip-19671-jewel
 661
 662     jewel: cephfs: MDS assert failed when shutting down
 663
 664     Reviewed-by: John Spray <john.spray@redhat.com>
 665
 666 commit ff60fceb5f5f7f9df4a48da4ad8bd863b9a04e2d
 667 Merge: 6a6d57d2de 96e801fb53
 668 Author: John Spray <jspray@redhat.com>
 669 Date:   Wed Jun 14 09:29:50 2017 -0400
 670
 671     Merge pull request #14682 from smithfarm/wip-19668-jewel
 672
 673     jewel: cephfs: MDS goes readonly writing backtrace for a file whose data pool has been removed
 674
 675     Reviewed-by: John Spray <john.spray@redhat.com>
 676
 677 commit 6a6d57d2de674c5a971e999cb2731b2d8ae1b523
 678 Merge: 8260669efa f34489dd52
 679 Author: John Spray <jspray@redhat.com>
 680 Date:   Wed Jun 14 09:27:13 2017 -0400
 681
 682     Merge pull request #14679 from smithfarm/wip-19666-jewel
 683
 684     jewel: cephfs: The mount point break off when mds switch hanppened.
 685
 686     Reviewed-by: John Spray <john.spray@redhat.com>
 687
 688 commit 8260669efa74f41d3b1b1039fb7b34e070951c34
 689 Merge: 013529b61f 824b19a9a6
 690 Author: John Spray <jspray@redhat.com>
 691 Date:   Wed Jun 14 09:26:40 2017 -0400
 692
 693     Merge pull request #14676 from smithfarm/wip-19619-jewel
 694
 695     jewel: cephfs: MDS server crashes due to inconsistent metadata.
 696
 697     Reviewed-by: John Spray <john.spray@redhat.com>
 698
 699 commit 013529b61fc4fbf02656f7c6cb0baa1bc6004758
 700 Merge: b518522f64 eab56dae67
 701 Author: John Spray <jspray@redhat.com>
 702 Date:   Wed Jun 14 09:24:38 2017 -0400
 703
 704     Merge pull request #14674 from smithfarm/wip-19482-jewel
 705
 706     jewel: cephfs: No output for ceph mds rmfailed 0 --yes-i-really-mean-it command
 707
 708     Reviewed-by: John Spray <john.spray@redhat.com>
 709
 710 commit b518522f64b66f4a14618a9345b6314ca0f2c54c
 711 Merge: 388e0d1bc3 63f41d543f
 712 Author: John Spray <jspray@redhat.com>
 713 Date:   Wed Jun 14 09:23:27 2017 -0400
 714
 715     Merge pull request #14671 from smithfarm/wip-19044-jewel
 716
 717     jewel: tests: buffer overflow in test LibCephFS.DirLs
 718
 719     Reviewed-by: John Spray <john.spray@redhat.com>
 720
 721 commit 388e0d1bc35985a6916d3eb1ca5184a0907b6e8b
 722 Merge: d7c7ce7ebb 7146816065
 723 Author: John Spray <jspray@redhat.com>
 724 Date:   Wed Jun 14 09:23:13 2017 -0400
 725
 726     Merge pull request #14670 from smithfarm/wip-18949-jewel
 727
 728     jewel: mds: avoid reusing deleted inode in StrayManager::_purge_stray_logged
 729
 730     Reviewed-by: John Spray <john.spray@redhat.com>
 731
 732 commit d7c7ce7ebbb663dab0dfa8058c845d494d7615cc
 733 Merge: d717ef73bc d8b139b584
 734 Author: John Spray <jspray@redhat.com>
 735 Date:   Wed Jun 14 09:22:28 2017 -0400
 736
 737     Merge pull request #14669 from smithfarm/wip-18900-jewel
 738
 739     jewel: cephfs: Test failure: test_open_inode
 740
 741     Reviewed-by: John Spray <john.spray@redhat.com>
 742
 743 commit d717ef73bc0d5ef24551ec9157385c8b0521380b
 744 Merge: c2a3b7567f 36c86f71ef
 745 Author: John Spray <jspray@redhat.com>
 746 Date:   Wed Jun 14 09:21:43 2017 -0400
 747
 748     Merge pull request #14668 from smithfarm/wip-18705-jewel
 749
 750     jewel: mds: fragment space check can cause replayed request fail
 751
 752     Reviewed-by: John Spray <john.spray@redhat.com>
 753
 754 commit c2a3b7567fbe0b2f62bcd38cd9bb9a5a2a238743
 755 Merge: fd9256b770 5b56214519
 756 Author: John Spray <jspray@redhat.com>
 757 Date:   Tue Jun 13 19:05:55 2017 +0100
 758
 759     Merge pull request #14698 from smithfarm/wip-19677-jewel
 760
 761     jewel: cephfs: ceph-fuse does not recover after lost connection to MDS
 762
 763     Reviewed-by: John Spray <john.spray@redhat.com>
 764
 765 commit fd9256b77010066c934fd0016eb6d3f9c1fb54e1
 766 Merge: 26ada59c81 c49b114e8d
 767 Author: Abhishek L <abhishek.lekshmanan@gmail.com>
 768 Date:   Fri Jun 9 19:39:17 2017 +0200
 769
 770     Merge pull request #14766 from smithfarm/wip-19757-jewel
 771
 772     jewel: rgw: fix failed to create bucket if a non-master zonegroup has a single zone
 773
 774     Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 775     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 776
 777 commit 26ada59c810bfda5c16f2d935a2b98711e7b2c76
 778 Merge: fdd25c2bdb e552d91f73
 779 Author: Abhishek L <abhishek.lekshmanan@gmail.com>
 780 Date:   Fri Jun 9 19:38:46 2017 +0200
 781
 782     Merge pull request #14787 from linuxbox2/jewel-rgw-shard-limit-ck
 783
 784     jewel: rgw: add bucket size limit check to radosgw-admin
 785
 786
 787     Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 788     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 789
 790 commit fdd25c2bdb9ec6bb8e3061088b8782bfb2331bc4
 791 Merge: cfd6750416 fb3ee2efcc
 792 Author: Abhishek L <abhishek.lekshmanan@gmail.com>
 793 Date:   Fri Jun 9 19:38:23 2017 +0200
 794
 795     Merge pull request #14789 from mdw-at-linuxbox/wip-jewel-rgw-rvk
 796
 797     jewel: rgw: swift: disable revocation thread if sleep == 0 || cache_size == 0
 798
 799
 800     Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 801     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 802
 803 commit cfd6750416baeb2bbd3836bc7dc55d83fc214c20
 804 Merge: ce1fc3492e 86980a045b
 805 Author: Abhishek L <abhishek.lekshmanan@gmail.com>
 806 Date:   Fri Jun 9 19:36:02 2017 +0200
 807
 808     Merge pull request #14815 from smithfarm/wip-19786-jewel
 809
 810     jewel: rgw: failure to create s3 type subuser from admin rest api
 811
 812     Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
 813     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 814
 815 commit ce1fc3492e87c669f7059c2047a3bed077418a89
 816 Merge: 7ca0252560 aa99558934
 817 Author: Nathan Cutler <presnypreklad@gmail.com>
 818 Date:   Wed Jun 7 20:13:15 2017 +0200
 819
 820     Merge pull request #15312 from theanalyst/wip-20078
 821
 822     jewel: rgw: only append zonegroups to rest params if not empty
 823
 824     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 825
 826 commit 7ca0252560d1ff23384afb50a9c2ae2aad2ce85c
 827 Merge: 62c500f522 59bd6711a4
 828 Author: Nathan Cutler <presnypreklad@gmail.com>
 829 Date:   Wed Jun 7 13:48:06 2017 +0200
 830
 831     Merge pull request #15382 from theanalyst/wip-mem-leak2
 832
 833     jewel: rgw:fix memory leaks in data/md sync
 834
 835     Reviewed-by: Casey Bodley <cbodley@redhat.com>
 836
 837 commit aa0cd461df5fee6d143bc07440ec6de829761cef
 838 Author: Brad Hubbard <bhubbard@redhat.com>
 839 Date:   Mon May 22 13:21:25 2017 +1000
 840
 841     osd: Move scrub sleep timer to osdservice
 842
 843     PR 14886 erroneously creates a scrub sleep timer for every pg resulting
 844     in a proliferation of threads. Move the timer to the osd service so
 845     there can be only one.
 846
 847     Fixes: http://tracker.ceph.com/issues/19986
 848
 849     Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
 850     (cherry picked from commit f110a82437df79dc20207d296e8229fc0e9ce18b)
 851
 852     Conflicts:
 853             src/osd/PG.cc - ceph_clock_now requires a CephContext argmunent
 854             in Jewel
 855
 856 commit c47bd0562b1187ffb0b1b2c1ef5f105aa7951d10
 857 Author: Brad Hubbard <bhubbard@redhat.com>
 858 Date:   Mon Apr 24 14:10:47 2017 +1000
 859
 860     osd: Implement asynchronous scrub sleep
 861
 862     Rather than blocking the main op queue just do an async sleep.
 863
 864     Fixes: http://tracker.ceph.com/issues/19497
 865
 866     Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
 867     (cherry picked from commit 7af3e86c2e4992db35637864b83832535c94d0e6)
 868
 869 commit e6daee8a9fbc576da2a03550a81056d093a516c9
 870 Author: Nathan Cutler <ncutler@suse.com>
 871 Date:   Sun Jun 4 20:39:58 2017 +0200
 872
 873     Client.cc: adjust Client::_getattr calls
 874
 875     Signed-off-by: Nathan Cutler <ncutler@suse.com>
 876
 877 commit a2c7a2262ac8ecbea78f09e6e8e6a37498568d57
 878 Author: John Spray <john.spray@redhat.com>
 879 Date:   Wed Mar 15 19:36:08 2017 +0000
 880
 881     qa/cephfs: use getfattr/setfattr helpers
 882
 883     Signed-off-by: John Spray <john.spray@redhat.com>
 884     (cherry picked from commit dd43d3bc646aeab88486b0963fc83de0b18800c4)
 885
 886     Conflicts:
 887             qa/tasks/cephfs/test_data_scan.py: difference in the
 888                 self._mount.run_shell() call in NonDefaultLayout::write (which is
 889                 being dropped by this commit) - in jewel it has "sudo", and in
 890                 master it doesn't
 891
 892 commit 12aa35a6d50f612df77199ac4f35c7baeed0583e
 893 Author: John Spray <john.spray@redhat.com>
 894 Date:   Wed Jun 22 13:00:44 2016 +0100
 895
 896     tasks/cephfs: fix race while mounting
 897
 898     This could fail if the mount hadn't finished
 899     coming up.
 900
 901     Signed-off-by: John Spray <john.spray@redhat.com>
 902     (cherry picked from commit adfb757c898a80f18c15dafd02e29840c5931c87)
 903
 904 commit a7b699269b65c76361fcb1d10593812be40c7612
 905 Author: John Spray <john.spray@redhat.com>
 906 Date:   Wed Mar 15 19:26:30 2017 +0000
 907
 908     qa: add test for reading quotas from different clients
 909
 910     Fixes: http://tracker.ceph.com/issues/17939
 911     Signed-off-by: John Spray <john.spray@redhat.com>
 912     (cherry picked from commit 61617f8f10a6322603a9add77980865cd972ef97)
 913
 914 commit 8b8ee392b8093b9b140dbbe895691f69ae40440f
 915 Author: John Spray <john.spray@redhat.com>
 916 Date:   Wed Mar 15 17:51:44 2017 +0000
 917
 918     client: _getattr on quota_root before using in statfs
 919
 920     ...so that after someone adjusts the quota settings
 921     on an inode that another client is using as its mount root,
 922     the change is visible immediately on the other client.
 923
 924     Signed-off-by: John Spray <john.spray@redhat.com>
 925     (cherry picked from commit 3d25941aadd223669448d0f5d3c0bd1fefa72308)
 926
 927 commit dd7d59a08141d6a24b172c22f5e27c8962e25fb9
 928 Author: John Spray <john.spray@redhat.com>
 929 Date:   Wed Mar 15 15:32:47 2017 +0000
 930
 931     client: getattr before read on ceph.* xattrs
 932
 933     Previously we were returning values for quota, layout
 934     xattrs without any kind of update -- the user just got
 935     whatever happened to be in our cache.
 936
 937     Clearly this extra round trip has a cost, but reads of
 938     these xattrs are fairly rare, happening on admin
 939     intervention rather than in normal operation.
 940
 941     Fixes: http://tracker.ceph.com/issues/17939
 942     Signed-off-by: John Spray <john.spray@redhat.com>
 943     (cherry picked from commit 532dc4b68e538c189ef828f67cecd0d647a62250)
 944
 945 commit 62c500f52240eaa5faadd3795bd9ec84bdcbc6c7
 946 Merge: 2badc2416c 8dd93cabd5
 947 Author: David Zafman <dzafman@redhat.com>
 948 Date:   Fri Jun 2 09:54:22 2017 -0700
 949
 950     Merge pull request #15416 from dzafman/wip-20126
 951
 952     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 953
 954 commit 8dd93cabd52cbafc29a47862f343431eb6f1cfe3
 955 Author: David Zafman <dzafman@redhat.com>
 956 Date:   Wed May 31 15:39:19 2017 -0700
 957
 958     osd: Object level shard errors are tracked and used if no auth available
 959
 960     Shards with object mismatch are tracked to mark them inconsistent
 961     Fix test because storing omap_digest in object_info not behaving as before
 962
 963     Fixes: http://tracker.ceph.com/issues/20089
 964
 965     Signed-off-by: David Zafman <dzafman@redhat.com>
 966
 967     (cherry picked from commit 1cacbea763c7aabfeaaf4bd5e878301044184117)
 968
 969     Conflicts:
 970             src/test/osd/osd-scrub-repair.sh (no alloc_hint in object_info)
 971
 972 commit 59bd6711a47c354117a612dd4ef033d70d449383
 973 Author: weiqiaomiao <wei.qiaomiao@zte.com.cn>
 974 Date:   Wed Jun 1 17:20:49 2016 +0800
 975
 976     rgw:fix memory leaks
 977
 978     Signed-off-by: weiqiaomiao <wei.qiaomiao@zte.com.cn>
 979     (cherry picked from commit 73e5be2b6133cf4caa0e5e5c8c9eae748b785dbf)
 980
 981 commit b429fa1807062716c9705ddcf316ed9b2741cc43
 982 Author: Casey Bodley <cbodley@redhat.com>
 983 Date:   Fri May 5 14:56:40 2017 -0400
 984
 985     cls/rgw: list_plain_entries() stops before bi_log entries
 986
 987     list_plain_entries() was using encode_obj_versioned_data_key() to set
 988     its end_key, which gives a prefix of BI_BUCKET_OBJ_INSTANCE_INDEX[=2]
 989
 990     that range between start_key and end_key would not only span the
 991     BI_BUCKET_OBJS_INDEX[=0] prefixes, but BI_BUCKET_LOG_INDEX[=1] prefixes
 992     as well. this can result in list_plain_entries() trying and failing to
 993     decode a rgw_bi_log_entry as a rgw_bucket_dir_entry
 994
 995     Fixes: http://tracker.ceph.com/issues/19876
 996
 997     Signed-off-by: Casey Bodley <cbodley@redhat.com>
 998     (cherry picked from commit b29a1633a57abf443d5790c13d680d2917f86037)
 999
1000 commit 2badc2416c9e16babbc91364502dcb03877e88bf
1001 Merge: 5d2a68eb90 1f895c2403
1002 Author: Sage Weil <sage@newdream.net>
1003 Date:   Tue May 30 09:39:47 2017 -0500
1004
1005     Merge pull request #15360 from liewegas/wip-jewel-master-mixed
1006
1007     qa/workunits/rados/test-upgrade-*: whitelist tests for master
1008
1009     Reviewed-by: Kefu Chai <kchai@redhat.com>
1010
1011 commit 1f895c24030b79ae9f2eae16b09582fc72928b6a
1012 Author: Sage Weil <sage@redhat.com>
1013 Date:   Tue May 30 09:58:09 2017 -0400
1014
1015     qa/workunits/rados/test-upgrade-*: whitelist tests for master
1016
1017     The jewel-x upgrade test now runs this script against a mixed cluster on
1018     a machine with code from master installed.  That means we have to skip
1019     any new tests that will fail on a mixed cluster.
1020
1021     Signed-off-by: Sage Weil <sage@redhat.com>
1022
1023 commit 5d2a68eb903f533689d1a1cbc410ca940baeaff3
1024 Merge: 998d1ee4f5 81e35b9416
1025 Author: Nathan Cutler <presnypreklad@gmail.com>
1026 Date:   Sat May 27 09:21:44 2017 +0200
1027
1028     Merge pull request #15208 from liewegas/wip-sortbitwise-jewel
1029
1030     mon: fix 'sortbitwise' warning on jewel
1031
1032     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1033
1034 commit 998d1ee4f503cb4875283fe8d2140f3b10a26643
1035 Merge: 54bc1e13ea 99c65bbc18
1036 Author: Nathan Cutler <presnypreklad@gmail.com>
1037 Date:   Sat May 27 02:46:55 2017 +0200
1038
1039     Merge pull request #14851 from yehudasa/wip-rgw-support-ragweed-jewel
1040
1041     jewel: rgw: add apis to support ragweed suite
1042
1043     Reviewed-by: Nathan Cutler <ncutler@suse.com>
1044
1045 commit aa99558934c6143cb35e722148b87f35ccecc397
1046 Author: Yehuda Sadeh <yehuda@redhat.com>
1047 Date:   Tue Aug 23 10:22:42 2016 -0700
1048
1049     rgw: rest conn functions cleanup, only append zonegroup if not empty
1050
1051     Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1052     (cherry picked from commit 4e41af19846db75081cb0ddb7b33dc2bb9321ace)
1053
1054 commit 5d90798432a6446636699fc03b2f548010b1212f
1055 Author: Karol Mroz <kmroz@suse.com>
1056 Date:   Thu Mar 17 10:32:14 2016 +0100
1057
1058     rgw: rest and http client code to use param vectors
1059
1060     Replaces param/header lists with vectors. In these cases, we're only ever
1061     adding to the back of the list, so a vector should be more efficient.
1062     Also moves param_pair_t/param_vec_t higher up the include chain for
1063     cleaner function signatures.
1064
1065     Signed-off-by: Karol Mroz <kmroz@suse.com>
1066     (cherry picked from commit d4a2527872e0f5c3ae2874bb7d0ff459ae83cfd4)
1067
1068     Conflicts:
1069             src/rgw/rgw_http_client.cc
1070     trivial ws conflict on rebase
1071
1072 commit 54bc1e13ea19642c8d5893f29e96ee5c053f9c59
1073 Merge: 966f222917 43327f83ef
1074 Author: Alfredo Deza <alfredo@deza.pe>
1075 Date:   Thu May 25 12:45:30 2017 -0400
1076
1077     Merge pull request #14765 from smithfarm/wip-18972-jewel
1078
1079     jewel: ceph-disk does not support cluster names different than 'ceph'
1080
1081     Reviewed-by: Alfredo Deza <adeza@redhat.com>
1082
1083 commit 81e35b941659a0f6dff8a935c27c7dd6d5cc4213
1084 Author: Sage Weil <sage@redhat.com>
1085 Date:   Wed May 24 09:48:11 2017 -0400
1086
1087     qa/suites/rados/singleton-nomsgr/*: set sortbitwise after upgrade
1088
1089     Signed-off-by: Sage Weil <sage@redhat.com>
1090
1091 commit f2814e4dbb2821a415310559fce405c2eab23947
1092 Author: huanwen ren <ren.huanwen@zte.com.cn>
1093 Date:   Tue Dec 27 10:54:45 2016 +0000
1094
1095     mon/OSDMonitor: fixup sortbitwise flag warning
1096
1097     "ceph -s" does not report warning when using
1098     command "ceph osd unset sortbitwise" to drop
1099     sortbitwise flag.
1100     we should use "osdmap.get_up_osd_features() &
1101     CEPH_FEATURE_OSD_BITWISE_HOBJ_SORT"
1102     instead of "(osdmap.get_features(CEPH_ENTITY_TYPE_OSD, NULL) &
1103     CEPH_FEATURE_OSD_BITWISE_HOBJ_SORT)",
1104     because osdmap.get_features only get local "features"
1105
1106     Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
1107     (cherry picked from commit c25ee187e28724846d0011cd8145e16956d3636e)
1108
1109 commit 02617188688eebde759c375a2257e076e4538491
1110 Author: Sage Weil <sage@redhat.com>
1111 Date:   Tue Nov 1 12:09:57 2016 -0400
1112
1113     mon: remove config option to disable no sortbitwise warning
1114
1115     We'll require this soon.
1116
1117     Signed-off-by: Sage Weil <sage@redhat.com>
1118     (cherry picked from commit 04e3319e8aecde9ca58ccb7c89016f9079c7d657)
1119
1120 commit 966f2229178c5d72722982d0f6a40f4d01210d9a
1121 Merge: 6537fc741c 4ceaa7cce9
1122 Author: Nathan Cutler <presnypreklad@gmail.com>
1123 Date:   Wed May 17 09:29:21 2017 +0200
1124
1125     Merge pull request #13450 from dreamhost/wip-18887-jewel
1126
1127     jewel: msg: IPv6 Heartbeat packets are not marked with DSCP QoS - simple messenger
1128
1129     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1130
1131 commit bb79663490468b4ac2832aa9e3cbac019c1f712d
1132 Author: Jan Fajerski <jfajerski@suse.com>
1133 Date:   Thu Apr 20 18:38:43 2017 +0200
1134
1135     fs: normalize file open flags internally used by cephfs
1136
1137     The file open flags (O_foo) are platform specific. Normalize these flags
1138     before they are send to the MDS. For processing of client messages the
1139     MDS should only compare to these normalized flags.
1140     Otherwise this can lead to bogus flags getting transmitted on ppc64.
1141
1142     Signed-off-by: Jan Fajerski <jfajerski@suse.com>
1143     (cherry picked from commit 88d2da5e93198e69435e288ce00d216d5fe27f80)
1144
1145     Conflicts:
1146             src/client/Client.cc
1147               Conflicts can be resolved by choosing changes from HEAD and
1148               adding a call to ceph_flags_sys2wire where flags are logged.
1149             src/mds/Server.cc
1150               Conflicts can be resolved by choosing changes from HEAD and
1151               while making sure that the MDS compares request O_ flags the
1152               the CEPH_O_ flags, since all wire O_ flags are normalized.
1153
1154 commit 4ceaa7cce9c9132d47564d79204b48b1d02e531c
1155 Author: Robin H. Johnson <robin.johnson@dreamhost.com>
1156 Date:   Wed May 3 22:31:40 2017 -0700
1157
1158     msg/simple/Pipe: manual backport of fix in PR#14795
1159
1160     Manual backport of errno fixup from PR#14795
1161     (6f1037e22c2a304795895498cdc955e0ef80f8e3), as noted by
1162     https://github.com/ceph/ceph/pull/13450#discussion_r114696885.
1163
1164     Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
1165
1166 commit 6537fc741c3267472e607e581c002854d8eafd46
1167 Merge: af31b453f4 82ea0971b3
1168 Author: Nathan Cutler <presnypreklad@gmail.com>
1169 Date:   Wed May 3 13:04:27 2017 +0200
1170
1171     Merge pull request #14667 from smithfarm/wip-18699-jewel
1172
1173     jewel: client: fix the cross-quota rename boundary check conditions
1174
1175     Reviewed-by: Gregory Farnum <gfarnum@redhat.com>
1176
1177 commit af31b453f404cb73ee7bdb8b6b02d71ad8aab7e7
1178 Merge: 472034cbe0 335258f975
1179 Author: Nathan Cutler <presnypreklad@gmail.com>
1180 Date:   Wed May 3 10:33:24 2017 +0200
1181
1182     Merge pull request #14332 from shinobu-x/wip-19396-jewel
1183
1184      jewel: Objecter::epoch_barrier isn't respected in _op_submit()
1185
1186     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1187
1188 commit 472034cbe0735131f3419205e9f7e6530377bfba
1189 Merge: afe98ba074 acf608a903
1190 Author: Nathan Cutler <presnypreklad@gmail.com>
1191 Date:   Wed May 3 10:32:37 2017 +0200
1192
1193     Merge pull request #14204 from dzafman/wip-18533-jewel
1194
1195     jewel: core: two instances of omap_digest mismatch
1196
1197     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1198
1199 commit afe98ba07436ea105f4c9c42a52795ce3757419e
1200 Merge: 0353a91bfd 043d70461c
1201 Author: Nathan Cutler <presnypreklad@gmail.com>
1202 Date:   Wed May 3 10:31:38 2017 +0200
1203
1204     Merge pull request #13884 from shinobu-x/wip-19119-jewel
1205
1206     jewel: pre-jewel "osd rm" incrementals are misinterpreted
1207
1208     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1209
1210 commit 0353a91bfd799077ddd87de901e8a07f5da6a13d
1211 Merge: 460b12c259 608785a007
1212 Author: Nathan Cutler <presnypreklad@gmail.com>
1213 Date:   Wed May 3 10:30:32 2017 +0200
1214
1215     Merge pull request #13647 from mslovy/wip-19083-jewel
1216
1217     jewel: osd: preserve allocation hint attribute during recovery
1218
1219     Reviewed-by: Gregory Farnum <gfarnum@redhat.com>
1220
1221 commit 460b12c259f5563d9d1b2477149fe79486ba5bcd
1222 Merge: 630cfca36c 905c4acb99
1223 Author: Nathan Cutler <presnypreklad@gmail.com>
1224 Date:   Fri Apr 28 10:24:35 2017 +0200
1225
1226     Merge pull request #14791 from smithfarm/wip-19774-jewel
1227
1228     jewel: osd: promote throttle parameters are reversed
1229
1230     Reviewed-by: Sage Weil <sage@redhat.com>
1231
1232 commit 630cfca36c5a42a58882966c5598752aac9ff54e
1233 Merge: 013e781b45 3ec1a9bf16
1234 Author: Nathan Cutler <presnypreklad@gmail.com>
1235 Date:   Fri Apr 28 10:23:45 2017 +0200
1236
1237     Merge pull request #14763 from smithfarm/wip-19562-jewel
1238
1239     jewel: api_misc: [  FAILED  ] LibRadosMiscConnectFailure.ConnectFailure
1240
1241     Reviewed-by: Sage Weil <sage@redhat.com>
1242
1243 commit 99c65bbc1899663d0b23e2cec3d6e516d0e95ad7
1244 Author: Yehuda Sadeh <yehuda@redhat.com>
1245 Date:   Mon Jan 9 13:04:43 2017 -0800
1246
1247     rgw: new rest api to retrieve object layout
1248
1249     Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1250     (cherry picked from commit 2768583dc486109e49d209243675b99fdd39e92c)
1251
1252 commit 33745a342433716ad8a07ef846dbea9b57de5017
1253 Author: Yehuda Sadeh <yehuda@redhat.com>
1254 Date:   Thu Jan 5 13:47:24 2017 -0800
1255
1256     rgw: rest api to read zone config params
1257
1258     Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1259     (cherry picked from commit a220a2efbfd675e6abf14ae33c21005bcbf6dadf)
1260
1261 commit 013e781b45e28af15da4ddc32025aad713f128e8
1262 Merge: d144d99f00 b698d1fa4c
1263 Author: Nathan Cutler <presnypreklad@gmail.com>
1264 Date:   Thu Apr 27 21:46:41 2017 +0200
1265
1266     Merge pull request #14481 from dillaman/wip-19468-jewel
1267
1268     jewel: librbd: is_exclusive_lock_owner API should ping OSD
1269
1270     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
1271
1272 commit d144d99f008776dc02c5838feb5a9eb9b84a400b
1273 Merge: 0f0cd82eda dbe90c79b8
1274 Author: Nathan Cutler <presnypreklad@gmail.com>
1275 Date:   Thu Apr 27 21:05:11 2017 +0200
1276
1277     Merge pull request #14666 from smithfarm/wip-19612-jewel
1278
1279     jewel: librbd: Issues with C API image metadata retrieval functions
1280
1281     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
1282
1283 commit 0f0cd82edabcb2feb0b29793e5b45eb424f1c42d
1284 Merge: eba821ce9c 216156b5d4
1285 Author: Nathan Cutler <presnypreklad@gmail.com>
1286 Date:   Thu Apr 27 21:04:27 2017 +0200
1287
1288     Merge pull request #14664 from smithfarm/wip-19325-jewel
1289
1290     jewel: rbd: [api] temporarily restrict (rbd_)mirror_peer_add from adding multiple peers
1291
1292     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
1293
1294 commit eba821ce9c61bcb197895e6949ce30a5c07097fa
1295 Merge: 25e29c9215 b8fd297eb6
1296 Author: Nathan Cutler <presnypreklad@gmail.com>
1297 Date:   Thu Apr 27 20:46:08 2017 +0200
1298
1299     Merge pull request #14694 from ceph/wip-bp-systemd
1300
1301     [backport] qa/tasks: systemd test backport to jewel
1302
1303     Reviewed-by: Nathan Cutler <ncutler@suse.com>
1304
1305 commit 25e29c921557883da4bf48ff2496bcec6448e6ec
1306 Merge: d7b63e21ca 7c6c3c753c
1307 Author: Nathan Cutler <presnypreklad@gmail.com>
1308 Date:   Thu Apr 27 08:44:26 2017 +0200
1309
1310     Merge pull request #13544 from shinobu-x/wip-18932-jewel
1311
1312     jewel: tests: 'ceph auth import -i' overwrites caps, should alert user before overwrite
1313
1314     Reviewed-by: Kefu Chai <kchai@redhat.com>
1315
1316 commit acf608a9034e915e38ccea6002ee808c46620433
1317 Author: David Zafman <dzafman@redhat.com>
1318 Date:   Mon Mar 20 17:28:45 2017 -0700
1319
1320     filestore, tools: Fix logging of DBObjectMap check() repairs
1321
1322     Signed-off-by: David Zafman <dzafman@redhat.com>
1323     (cherry picked from commit 1704f62c0831e6b07138f7dd14a89fef3c9ed2c1)
1324
1325 commit fecc52338b2a58bf1730f7b7a3e4a293e45160d2
1326 Author: David Zafman <dzafman@redhat.com>
1327 Date:   Fri Mar 3 15:04:02 2017 -0800
1328
1329     osd: Simplify DBObjectMap by no longer creating complete tables
1330
1331     Bump the version for new maps to 3
1332     Make clone less efficient but simpler
1333     Add rename operation (use instead of clone/unlink)
1334     For now keep code that understands version 2 maps
1335
1336     Signed-off-by: David Zafman <dzafman@redhat.com>
1337     (cherry picked from commit 738156a99ed1caf61e5a8230eb8048360056c08e)
1338
1339     No ghobject_t::operator>() so use Kraken cmp_bitwise() instead
1340     Need to use MIN_GHOBJ/MAX_GHOBJ instead of std::min/std::max
1341
1342 commit 6902c3141eeaefaacd92f33877cf319872f626c7
1343 Author: David Zafman <dzafman@redhat.com>
1344 Date:   Wed Feb 15 16:17:32 2017 -0800
1345
1346     ceph-osdomap-tool: Fix seg fault with large amount of check error output
1347
1348     Signed-off-by: David Zafman <dzafman@redhat.com>
1349     (cherry picked from commit 1dda0411f4fbb14ce1e0062da9f14ec3af505d39)
1350
1351 commit 4a3e4bcf40cd004fc53e7be467a29084dedc3e1c
1352 Author: David Zafman <dzafman@redhat.com>
1353 Date:   Wed Feb 15 15:02:33 2017 -0800
1354
1355     osd: Add automatic repair for DBObjectMap bug
1356
1357     Add repair command to ceph-osdomap-tool too
1358
1359     Under some situations the previous rm_keys() code would
1360     generated a corrupt complete table.  There is no way to
1361     figure out what the table should look like now.  By removing
1362     the entries we fix the corruption and aren't much worse off
1363     because the corruption caused some deleted keys to re-appear.
1364
1365     This doesn't breaking the parent/child relationship during
1366     repair because some of the keys may still be contained
1367     in the parent.
1368
1369     Signed-off-by: David Zafman <dzafman@redhat.com>
1370     (cherry picked from commit 4cd3c74c928a32e065ed9543d6c91d8718a6ae3d)
1371
1372     Conflicts:
1373             src/os/filestore/DBObjectMap.h (trivial)
1374
1375 commit d4f0ac0a405266f638b25ec475c0110741e3c431
1376 Author: David Zafman <dzafman@redhat.com>
1377 Date:   Wed Feb 15 14:59:40 2017 -0800
1378
1379     ceph-osdomap-tool: Fix tool exit status
1380
1381     Signed-off-by: David Zafman <dzafman@redhat.com>
1382     (cherry picked from commit 666f14ed90655a2d1bedde8561949625db7a9e6c)
1383
1384 commit 5f36c319cd7fa4be15efd27b8aabbebc99d8999c
1385 Author: Samuel Just <sjust@redhat.com>
1386 Date:   Fri Feb 10 15:51:42 2017 -0800
1387
1388     DBObjectMap: rewrite rm_keys and merge_new_complete
1389
1390     Leverage the updated in_complete_region and needs_parent to simplify
1391     these methods.
1392
1393     Signed-off-by: Samuel Just <sjust@redhat.com>
1394     (cherry picked from commit c4dffb68eaafe724f7fdae93a4285a7f8003ea29)
1395
1396 commit 1fe4b856a37b29c85a9317e514c7f15f8e4905d5
1397 Author: Samuel Just <sjust@redhat.com>
1398 Date:   Fri Feb 10 15:50:57 2017 -0800
1399
1400     DBObjectMap: strengthen in_complete_region post condition
1401
1402     Previously, in_complete_region didn't guarantee anything about
1403     where it left complete_iter pointing.  It will be handy for
1404     complete_iter to be pointing at the lowest interval which ends
1405     after to_test.  Make it so.
1406
1407     Signed-off-by: Samuel Just <sjust@redhat.com>
1408     (cherry picked from commit 97b35f4d7d4862da4b6f50ecaef0d292a671fd04)
1409
1410 commit 85f2151fec991e5db13d8e6f44b27e092605fb35
1411 Author: Samuel Just <sjust@redhat.com>
1412 Date:   Fri Feb 10 15:48:57 2017 -0800
1413
1414     DBObjectMap: fix next_parent()
1415
1416     The previous implementation assumed that
1417     lower_bound(parent_iter->key()) always leaves the iterator
1418     on_parent().  There isn't any guarantee, however, that that
1419     key isn't present on the child as well.
1420
1421     Signed-off-by: Samuel Just <sjust@redhat.com>
1422     (cherry picked from commit 74a7631d0938d7b44894f022224eab10a90d5cec)
1423
1424 commit 484ccda5c53a8ee151b9f97687edd160ad7ebbd7
1425 Author: Samuel Just <sjust@redhat.com>
1426 Date:   Thu Feb 9 10:47:59 2017 -0800
1427
1428     test_object_map: add tests to trigger some bugs related to 18533
1429
1430     Signed-off-by: Samuel Just <sjust@redhat.com>
1431     (cherry picked from commit f131dbcf5bb17107c029f942a57e9bf4432a26ee)
1432
1433 commit cdeb690869bf6dd52226476c21514cf03ff37d6e
1434 Author: David Zafman <dzafman@redhat.com>
1435 Date:   Tue Feb 14 12:40:33 2017 -0800
1436
1437     test: Add ceph_test_object_map to make check tests
1438
1439     Signed-off-by: David Zafman <dzafman@redhat.com>
1440     (cherry picked from commit 0e97a01bd7291458881ee53cece2d887f6333669)
1441
1442 commit cf5d588d39498c8d65bc64bd2935aecac8546e40
1443 Author: David Zafman <dzafman@redhat.com>
1444 Date:   Wed Feb 8 18:56:27 2017 -0800
1445
1446     ceph-osdomap-tool: Add --debug and only show internal logging if enabled
1447
1448     Signed-off-by: David Zafman <dzafman@redhat.com>
1449     (cherry picked from commit 5fb2b2d13953979e5da9f571ab8c4b0b510b8368)
1450
1451 commit 4c4a06ff525d9fa2271099db73701c7994054d36
1452 Author: David Zafman <dzafman@redhat.com>
1453 Date:   Wed Feb 8 18:55:48 2017 -0800
1454
1455     osd: DBOjectMap::check: Dump complete mapping when inconsistency found
1456
1457     Signed-off-by: David Zafman <dzafman@redhat.com>
1458     (cherry picked from commit fcf1e17c645e8fad5216c3e59627c817e5c858c7)
1459
1460 commit 6c128ff8c6944e57059008959f49bd03635c5417
1461 Author: David Zafman <dzafman@redhat.com>
1462 Date:   Wed Feb 8 15:38:51 2017 -0800
1463
1464     test_object_map: Use ASSERT_EQ() for check() so failure doesn't stop testing
1465
1466     Signed-off-by: David Zafman <dzafman@redhat.com>
1467     (cherry picked from commit 053a273cbc02d6902a4bb1f11db1ea946498df3a)
1468
1469 commit aa769a9d08ac7490f5063bb44c3a44c77a5d7232
1470 Author: David Zafman <dzafman@redhat.com>
1471 Date:   Wed Feb 8 10:02:40 2017 -0800
1472
1473     tools: Check for overlaps in internal "complete" table for DBObjectMap
1474
1475     Changed check to return an error count and fix tool error message
1476
1477     Signed-off-by: David Zafman <dzafman@redhat.com>
1478     (cherry picked from commit e5e8eb962db6187ea19b96ba29ac83469c90b4ea)
1479
1480     Conflicts:
1481             src/os/filestore/DBObjectMap.h (trivial)
1482
1483 commit 761ee7c6af8802ab6b668a0b7ccaa819b2764456
1484 Author: David Zafman <dzafman@redhat.com>
1485 Date:   Wed Feb 8 09:40:49 2017 -0800
1486
1487     tools: Add dump-headers command to ceph-osdomap-tool
1488
1489     Signed-off-by: David Zafman <dzafman@redhat.com>
1490     (cherry picked from commit f4101591ad701a62fe027c4744ca8ea505f44bdc)
1491
1492     Conflicts:
1493             src/os/filestore/DBObjectMap.h (trivial)
1494
1495 commit 117db1c6d6952d181614f4fe22f0b1866eed6f10
1496 Author: David Zafman <dzafman@redhat.com>
1497 Date:   Mon Feb 6 21:09:42 2017 -0800
1498
1499     tools: Add --oid option to ceph-osdomap-tool
1500
1501     Signed-off-by: David Zafman <dzafman@redhat.com>
1502     (cherry picked from commit 2d94889e9ee3359017b1efd560f3557ce03ccee6)
1503
1504 commit 4d8120d3b3975ce101f6272240e5e43bcfe0e742
1505 Author: David Zafman <dzafman@redhat.com>
1506 Date:   Mon Feb 6 21:31:18 2017 -0800
1507
1508     osd: Remove unnecessary assert and assignment in DBObjectMap
1509
1510     Fix and add comment(s)
1511
1512     Signed-off-by: David Zafman <dzafman@redhat.com>
1513     (cherry picked from commit 937e6a03ea4692cc44d53faa0615f8e808c9eb03)
1514
1515 commit 86980a045b7176428062a19758d145e38e515b40
1516 Author: snakeAngel2015 <li.jieA@h3c.com>
1517 Date:   Mon Jul 18 14:51:37 2016 +0800
1518
1519     rgw: add suport for creating S3 type subuser of admin rest api
1520
1521     Fixes: http://tracker.ceph.com/issues/16682
1522
1523     The original code cannot support create s3 type subuser of admin rest api as when i execute the following command:
1524
1525       ./s3curl.pl --id=personal --put -- http://radosgw.h3c.com:8000/admin/user?subuser\&uid=yrf2\&subuser=yrf2:yrf1\&key-type=s3 -v
1526
1527     it would return msg as follows :
1528
1529              < HTTP/1.1 403 Forbidden
1530             < Date: Thu, 14 Jul 2016 07:04:40 GMT
1531             * Server Apache/2.4.7 (Ubuntu) is not blacklisted
1532             < Server: Apache/2.4.7 (Ubuntu)
1533             < x-amz-request-id: tx00000000000000006608f-0057873988-8551-slave
1534             < Accept-Ranges: bytes
1535             < Content-Length: 114
1536             < Content-Type: application/json
1537             <
1538             * Connection #0 to host slave.com left intact
1539             {"Code":"InvalidAccessKeyId","RequestId":"tx00000000000000006608f-0057873988-8551-slave","HostId":"8551-slave-us"}
1540
1541      But i have modified the codes for support it ,and it will return actual msg as follows :
1542
1543       "subusers": [
1544             {
1545                 "id": "yrf2:yrf1",
1546                 "permissions": "<none>"
1547             }
1548         ],
1549         "keys": [
1550             {
1551                 "user": "yrf2",
1552                 "access_key": "B46PXYFEWUX0IMHGKP8C",
1553                 "secret_key": "2JYxywXizqwiiMd74UXrJdSJMPNlBtYwF5z8rNvh"
1554             },
1555             {
1556                 "user": "yrf2:yrf1",
1557                 "access_key": "INO55WXJ7JQ1ZZGSAB6B",
1558                 "secret_key": "GgCKEfF9hArV2hglunbO7KtvKZnbhmsDpqjSj5DL"
1559             }
1560         ],
1561
1562     Please check it ,thanks .
1563
1564     Signed-off-by: snakeAngel2015 <li.jieA@h3c.com>
1565     (cherry picked from commit 6535f6ad2137ee55bf5531e865c05aa10bd39bd0)
1566
1567 commit d7b63e21ca9818e21afeb2945b882e24b7a9b10b
1568 Merge: a3fae531cd f32b5c613a
1569 Author: Nathan Cutler <presnypreklad@gmail.com>
1570 Date:   Wed Apr 26 21:32:31 2017 +0200
1571
1572     Merge pull request #14809 from tchaikov/wip-18193-jewel
1573
1574     jewel: tests: test/librados/tmap_migrate: g_ceph_context->put() upon return
1575
1576     Reviewed-by: Nathan Cutler <ncutler@suse.com>
1577
1578 commit a3fae531cdb8c2c64c66d38c947ca139f4ede83d
1579 Merge: 89d6ddb1f5 6c4826606d
1580 Author: Nathan Cutler <presnypreklad@gmail.com>
1581 Date:   Wed Apr 26 21:32:09 2017 +0200
1582
1583     Merge pull request #14701 from smithfarm/wip-18193-jewel
1584
1585     jewel: core: transient jerasure unit test failures
1586
1587     Reviewed-by: Kefu Chai <kchai@redhat.com>
1588
1589 commit 89d6ddb1f5ee08b0f019dfaadc5857a788c92a09
1590 Merge: c5f1fce21d be9e83281b
1591 Author: Matt Benjamin <mbenjamin@redhat.com>
1592 Date:   Wed Apr 26 14:28:29 2017 -0400
1593
1594     Merge pull request #14776 from linuxbox2/jewel-pullup-civet-chunked
1595
1596     [DNM] jewel: pullup civet chunked
1597
1598 commit c5f1fce21d0996cce751b6bcca5e57da2cafc135
1599 Merge: 4d97e0ba8e 97cd21afc5
1600 Author: Nathan Cutler <presnypreklad@gmail.com>
1601 Date:   Wed Apr 26 19:32:07 2017 +0200
1602
1603     Merge pull request #14416 from smithfarm/wip-19557-jewel
1604
1605     jewel: tests: upgrade/hammer-x failing with OSD has the store locked when Thrasher runs ceph-objectstore-tool on down PG
1606
1607     Reviewed-by: Kefu Chai <kchai@redhat.com>
1608
1609 commit be9e83281b8c765cd111d5687a516fcd3ca521a5
1610 Author: Matt Benjamin <mbenjamin@redhat.com>
1611 Date:   Tue Apr 25 09:44:39 2017 -0400
1612
1613     civetweb:  pullup chunked encoding by Marcus
1614
1615     Fixes: http://tracker.ceph.com/issues/19736
1616
1617     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
1618
1619 commit 608785a0079f807ff860c56d96b1b67bf6a2ed74
1620 Author: yaoning <yaoning@unitedstack.com>
1621 Date:   Fri Jun 24 09:51:07 2016 +0800
1622
1623     os: make zero values noops for set_alloc_hint() in FileStore
1624
1625     Signed-off-by: yaoning <yaoning@unitedstack.com>
1626     (cherry picked from commit e2ec24f61b55457caccefecd56f9f08b98264802)
1627
1628 commit f32b5c613a8acd50e32747c3581131d28d209efa
1629 Author: Kefu Chai <kchai@redhat.com>
1630 Date:   Wed Apr 26 22:58:30 2017 +0800
1631
1632     test/librados/tmap_migrate: g_ceph_context->put() upon return
1633
1634     Signed-off-by: Kefu Chai <kchai@redhat.com>
1635     Conflict: test/librados/tmap_migrate.cc
1636       this change is not cherry-picked from master, because tmap_migrate was
1637     removed in master. so we are applying the same change in
1638     cb1cda96713b2ec0f6418c4cbe3d964c2020729c to this test.
1639
1640 commit 905c4acb99f9ea78ff615034dae969ab089bda06
1641 Author: Mark Nelson <mnelson@redhat.com>
1642 Date:   Tue May 3 09:56:47 2016 -0500
1643
1644     Fix reveresed promote throttle default parameters.
1645
1646     Signed-off-by: Mark Nelson <mnelson@redhat.com>
1647     (cherry picked from commit 793ceac2f3d5a2c404ac50569c44a21de6001b62)
1648
1649 commit fb3ee2efcc13c37db90faa1bc8bad584bab22efa
1650 Author: Marcus Watts <mwatts@redhat.com>
1651 Date:   Thu Apr 13 05:33:55 2017 -0400
1652
1653     rgw: swift: disable revocation thread if sleep == 0 || cache_size == 0
1654
1655     Keystone tokens can be revoked.  This causes them to fail
1656     validation.  However, in ceph, we cache them.  As long as
1657     they're in the cache we trust them.  To find revoked tokens
1658     there's a call OSI-PKI/revoked but that's only useful for
1659     pki tokens.  Installations using fernet/uuid may not even
1660     have the proper credentials to support the call, in which
1661     case the call blows up in various ways filling up logs
1662     with complaints.
1663
1664     This code makes the revocation thread optional; by disabling it,
1665     the complaints go away.  A further fix is in the works
1666     to use other more modern calls available in modern keystone
1667     installations to properly deal with non-PKI/PKIZ tokens.
1668
1669     (NB: jewel has this logic in src/rgw/rgw_swift.cc not in src/rgw/rgw_keystone.h)
1670
1671     To disable the revocation thread, use at least one of these:
1672             rgw_keystone_token_cache_size = 0
1673                     using this will cause tokens to be validated on every call.
1674     You may instead want to set
1675             rgw_keystone_revocation_interval = 0
1676                     using just this will disable the revocation thread,
1677                     but leaves the cache in use.  That avoids the extra
1678                     validation overhead, but means token revocation won't
1679                     work very well.
1680
1681     Fixes: http://tracker.ceph.com/issues/9493
1682     Fixes: http://tracker.ceph.com/issues/19499
1683
1684     Signed-off-by: Marcus Watts <mwatts@redhat.com>
1685     (cherry picked from commit 003291a8cbca455c0e8731f66759395a0bb1f555)
1686
1687 commit 4d97e0ba8ebc89b7797b0936a9e046ef59cc3899
1688 Merge: cc820a0d4c c2efeb4b62
1689 Author: Nathan Cutler <presnypreklad@gmail.com>
1690 Date:   Tue Apr 25 21:06:26 2017 +0200
1691
1692     Merge pull request #14686 from smithfarm/wip-19686-jewel
1693
1694     jewel: osd: Give requested scrubs a higher priority
1695
1696     Reviewed-by: David Zafman <dzafman@redhat.com>
1697
1698 commit e552d91f73d996c44821f5fbfb28cfc3e5cddc9a
1699 Author: Matt Benjamin <mbenjamin@redhat.com>
1700 Date:   Sat Nov 5 13:13:47 2016 -0400
1701
1702     rgw: add bucket size limit check to radosgw-admin
1703
1704     The change adds a new list of all buckets x all users, with
1705     fields for bucket name, tenant name, current num_objects,
1706     current num_shards, current objects per shard, and the
1707     corresponding fill_status--the latter consisting of 'OK',
1708     'WARN <n>%', or 'OVER <n>%.'
1709
1710     The warning check is relative to two new tunables.  The threshold
1711     max objects per shard is set as rgw_bucket_safe_max_objects_per_shard,
1712     which defaults to 100K.  The value rgw_bucket_warning_threshold is
1713     a percent of the current safe max at which to warn (defaults to
1714     90% of full).
1715
1716     From review:
1717
1718     * fix indentation (rgw_admin)
1719     * if user a user_id is provided, check only buckets for that user
1720     * update shard warn pct to be pct-of-fill (not 100 - pct-of-fill)
1721     * print only buckets near or over per-shard limit, if --warnings-only
1722     * s/bucket limitcheck/bucket limit check */
1723     * sanity shard limit should be 90, not 10 (because that changed)
1724     * fixes for memleaks and other points found by cbodley
1725
1726     Fixes: http://tracker.ceph.com/issues/17925
1727
1728     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
1729     (cherry picked from commit 7bc144ce36fedc16a3dedc54598b0d75fb8c68bc)
1730
1731 commit cc820a0d4c2676799383aeaf49a9269e104853ce
1732 Merge: 5ee54cfe8f d079b91479
1733 Author: Nathan Cutler <presnypreklad@gmail.com>
1734 Date:   Tue Apr 25 17:38:21 2017 +0200
1735
1736     Merge pull request #14605 from asheplyakov/19476-jewel
1737
1738     jewel: rgw: don't return skew time in pre-signed url
1739
1740     Reviewed-by: Casey Bodley <cbodley@redhat.com>
1741
1742 commit 5ee54cfe8f0a955b11edb4819a652ee420f43e5d
1743 Merge: 37254aee6c c05bd1cb1f
1744 Author: Nathan Cutler <presnypreklad@gmail.com>
1745 Date:   Tue Apr 25 17:37:16 2017 +0200
1746
1747     Merge pull request #14660 from smithfarm/wip-19478-jewel
1748
1749     jewel: rgw: zonegroupmap set does not work
1750
1751     Reviewed-by: Casey Bodley <cbodley@redhat.com>
1752
1753 commit 37254aee6c3f725cf60deb5c7a3fc28405c51216
1754 Merge: 09919f938f faeb8088ac
1755 Author: Nathan Cutler <presnypreklad@gmail.com>
1756 Date:   Tue Apr 25 17:36:23 2017 +0200
1757
1758     Merge pull request #14607 from asheplyakov/19607-jewel
1759
1760     jewel: rgw: multisite: fetch_remote_obj() gets wrong version when copying from remote
1761
1762     Reviewed-by: Casey Bodley <cbodley@redhat.com>
1763
1764 commit 09919f938f3c1e9e81963f676a9dd13114d54e73
1765 Merge: d5e1345a2a 527911fab7
1766 Author: Nathan Cutler <presnypreklad@gmail.com>
1767 Date:   Tue Apr 25 15:50:26 2017 +0200
1768
1769     Merge pull request #14587 from asheplyakov/19617-jewel
1770
1771     jewel: mon/MonClient: make get_mon_log_message() atomic
1772
1773     Reviewed-by: Sage Weil <sage@redhat.com>
1774
1775 commit 6c4826606dc81fdd847959a49454c69a958bb1d8
1776 Author: Kefu Chai <kchai@redhat.com>
1777 Date:   Mon Dec 5 20:23:21 2016 +0800
1778
1779     test/ceph_crypto: do not read ceph.conf in global_init()
1780
1781     ForkDeathTest.MD5 expect an empty output while global_init() complains
1782     if ceph.conf is missing if 0 is passed in as the `flags`. this test
1783     passes if ceph.conf is in current working directory, but jenkins does
1784     not prepare this file for this test.
1785
1786     Fixes: http://tracker.ceph.com/issues/18128
1787     Signed-off-by: Kefu Chai <kchai@redhat.com>
1788     (cherry picked from commit c72a2271a8012a66d7bbccf5766a73da5bb878d6)
1789
1790     Conflicts:
1791             src/test/ceph_crypto.cc - jewel does not have
1792                  5af29540675b674c1985ff98b28a783ed124acf6
1793
1794 commit aca2659952528aa1b40b02828293342128657194
1795 Author: Loic Dachary <ldachary@redhat.com>
1796 Date:   Thu Dec 8 12:40:42 2016 +0100
1797
1798     tests: fix erasure-code premature deallocation of cct
1799
1800     The setup function returns before the run function, the cct variable
1801     must be a data member, not a local variable that gets de-allocated
1802     before run() starts.
1803
1804     Signed-off-by: Loic Dachary <loic@dachary.org>
1805     (cherry picked from commit efa1e54362423d4cfd1541fb8c68237b7b9ebbe3)
1806
1807 commit 8bddd427d9a4bca75a352ca333847dbd6d2c369c
1808 Author: Pan Liu <pan.liu@istuary.com>
1809 Date:   Thu Feb 16 22:17:52 2017 +0800
1810
1811     rbd-nbd: no need create asok file for unmap and list-mapped commands.
1812
1813     Fixes: http://tracker.ceph.com/issues/17951
1814     Signed-off-by: Pan Liu <liupan1111@gmail.com>
1815     (cherry picked from commit 72352653d585ef89043a4ece371b5c0cb3f6f32a)
1816
1817 commit 328bfbd25c63b1b0e253865abedada7a9e5858e4
1818 Author: Jason Dillaman <dillaman@redhat.com>
1819 Date:   Tue Nov 29 12:36:00 2016 -0500
1820
1821     rbd-nbd: restart parent process logger after forking
1822
1823     Fixes: http://tracker.ceph.com/issues/18070
1824     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1825     (cherry picked from commit 29baf254d72cc593572b5a6215360ba51e3be198)
1826
1827 commit 192e7bcdcd1bc02b40b29eff86a335cc8919f663
1828 Author: Kefu Chai <kchai@redhat.com>
1829 Date:   Wed Nov 23 19:45:57 2016 +0800
1830
1831     crushtool: do not release g_ceph_context at exit
1832
1833     it is but a work around of occasionally timeout.
1834
1835     Signed-off-by: Kefu Chai <kchai@redhat.com>
1836     (cherry picked from commit d305cc51b18cbf4b2757bbacb5d43324461306a9)
1837
1838 commit 8a2f27cc632c26d7c2b8e8528b4d459b1d78705b
1839 Author: Kefu Chai <kchai@redhat.com>
1840 Date:   Tue Nov 15 14:21:03 2016 +0800
1841
1842     common,test: g_ceph_context->put() upon return
1843
1844     prior to this change, global_init() could create a new CephContext
1845     and assign it to g_ceph_context. it's our responsibilty to release
1846     the CephContext explicitly using cct->put() before the application
1847     quits. but sometimes, we fail to do so.
1848
1849     in this change, global_init() will return an intrusive_ptr<CephContext>,
1850     which calls `g_ceph_context->put()` in its dtor. this ensures that
1851     the CephContext is always destroyed before main() returns. so the
1852     log is flushed before _log_exp_length is destroyed.
1853
1854     there are two cases where global_pre_init() is called directly.
1855     - ceph_conf.cc: g_ceph_context->put() will be called by an intrusive_ptr<>
1856       deleter.
1857     - rgw_main.cc: global_init() is called later on on the success code
1858       path, so it will be taken care of.
1859
1860     Fixes: http://tracker.ceph.com/issues/17762
1861     Signed-off-by: Kefu Chai <kchai@redhat.com>
1862     (cherry picked from commit cb1cda96713b2ec0f6418c4cbe3d964c2020729c)
1863
1864     Conflicts:
1865             src/ceph_fuse.cc
1866             src/ceph_mgr.cc
1867             src/global/global_init.cc
1868             src/rgw/rgw_main.cc
1869             src/test/compressor/test_compression.cc
1870             src/test/compressor/test_compression_plugin.cc
1871             src/test/compressor/test_compression_plugin_snappy.cc
1872             src/test/compressor/test_compression_plugin_zlib.cc
1873             src/test/compressor/test_compression_snappy.cc
1874             src/test/compressor/test_compression_zlib.cc
1875             src/test/erasure-code/TestErasureCode.cc
1876             src/test/erasure-code/TestErasureCodeExample.cc
1877             src/test/erasure-code/TestErasureCodeIsa.cc
1878             src/test/erasure-code/TestErasureCodeJerasure.cc
1879             src/test/erasure-code/TestErasureCodeLrc.cc
1880             src/test/erasure-code/TestErasureCodePlugin.cc
1881             src/test/erasure-code/TestErasureCodePluginIsa.cc
1882             src/test/erasure-code/TestErasureCodePluginJerasure.cc
1883             src/test/erasure-code/TestErasureCodePluginLrc.cc
1884             src/test/erasure-code/TestErasureCodePluginShec.cc
1885             src/test/erasure-code/TestErasureCodeShec.cc
1886             src/test/erasure-code/TestErasureCodeShec_thread.cc
1887             src/test/fio/fio_ceph_objectstore.cc
1888             src/test/librados/misc.cc
1889             src/test/mon/PGMap.cc
1890             src/test/msgr/test_async_networkstack.cc
1891             src/test/msgr/test_userspace_event.cc
1892             src/test/objectstore/Allocator_test.cc
1893             src/test/objectstore/BitAllocator_test.cc
1894             src/test/objectstore/test_bluefs.cc
1895             src/test/objectstore/test_bluestore_types.cc
1896             src/test/objectstore/test_memstore_clone.cc
1897             src/test/osd/TestPGLog.cc
1898             src/test/rgw/test_http_manager.cc
1899             src/test/rgw/test_rgw_compression.cc
1900             src/test/test_mempool.cc
1901             src/tools/rados/rados.cc
1902
1903 commit d5e1345a2a6bd8456417db6bb60d61871165e6bd
1904 Merge: 256f48f463 5096fc9c7c
1905 Author: Nathan Cutler <presnypreklad@gmail.com>
1906 Date:   Tue Apr 25 15:42:43 2017 +0200
1907
1908     Merge pull request #14665 from smithfarm/wip-19610-jewel
1909
1910     jewel: [librados_test_stub] cls_cxx_map_get_XYZ methods don't return correct value
1911
1912     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
1913
1914 commit 256f48f46354f11a2f238f24fa9d890b55f4f4fc
1915 Merge: 4f67da1b9d 21a83e1276
1916 Author: Nathan Cutler <presnypreklad@gmail.com>
1917 Date:   Tue Apr 25 09:20:19 2017 +0200
1918
1919     Merge pull request #14653 from smithfarm/wip-19662-jewel
1920
1921     jewel: rgw_file: fix event expire check, don't expire directories being read
1922
1923     Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>
1924
1925 commit 4f67da1b9d554a1a5371665573b1f56d3f903de3
1926 Merge: c2452c5364 33af18e592
1927 Author: Loic Dachary <loic@dachary.org>
1928 Date:   Tue Apr 25 09:17:33 2017 +0200
1929
1930     Merge pull request #14635 from smithfarm/wip-19690-jewel
1931
1932     jewel: doc: Improvements to crushtool manpage
1933
1934     Reviewed-by: Loic Dachary <ldachary@redhat.com>
1935
1936 commit 040ff013d5c8aded8beedc59bbc4f5afccc64e46
1937 Author: Kefu Chai <kchai@redhat.com>
1938 Date:   Wed Nov 16 11:56:09 2016 +0800
1939
1940     crushtool: s/exit(EXIT_FAILURE)/return EXIT_FAILURE/
1941
1942     so the destructor(s) can be called.
1943
1944     Signed-off-by: Kefu Chai <kchai@redhat.com>
1945     (cherry picked from commit e01b89ed6be6b99fec5c725f4bc5769b42468cac)
1946
1947     Conflicts:
1948             src/tools/crushtool.cc - jewel does not have 17feefbcb3105553b763cb7ce123b20b77b95857
1949
1950 commit 8e993e6e43ecd6e2b444961d1e6c945081b64207
1951 Author: Kefu Chai <kchai@redhat.com>
1952 Date:   Wed Nov 16 11:19:04 2016 +0800
1953
1954     global/signal_handler: reset injected segv after test
1955
1956     ~CephContext() => ~TypedSingletonWrapper() => ~MempoolObs =>
1957     unregister_command() => ldout() << "unregister_command" =>
1958     Log::submit_entry() => *(volatile int *)(0) = 0xdead;
1959
1960     Signed-off-by: Kefu Chai <kchai@redhat.com>
1961     (cherry picked from commit d932c8f2f23263924103a900714db82ee87f6eef)
1962
1963     Conflicts:
1964             src/log/Log.cc - jewel has "namespace log" instead of "namespace
1965                              logging" (trivial resolution)
1966
1967 commit f1c0042b831826e5bd72ada79a4918a9c26bda24
1968 Author: Kefu Chai <kchai@redhat.com>
1969 Date:   Tue Nov 15 14:42:35 2016 +0800
1970
1971     test_cors.cc: fix the mem leak
1972
1973     Signed-off-by: Kefu Chai <kchai@redhat.com>
1974     (cherry picked from commit c8a3777203482cabf6739a8ba69b127df8697628)
1975
1976 commit c49b114e8dde6c3f3c5a5b663bd8299b4f2feafb
1977 Author: weiqiaomiao <wei.qiaomiao@zte.com.cn>
1978 Date:   Tue Sep 6 16:34:52 2016 +0800
1979
1980     rgw: fix failed to create bucket if a non-master zonegroup has a single zone
1981
1982     If a non-master zonegroup has a single zone, the metadata sync thread not running and
1983     the non-master zonegroup can't sync user from master zonegroup,
1984     so we can't create bucket(or other metadata update) in it
1985     because the authenticated user not found in the zone of non-master zonegroup.
1986
1987     Signed-off-by: weiqiaomiao <wei.qiaomiao@zte.com.cn>
1988     (cherry picked from commit 949af79b21098e6410bc29274cf36eae2d89faea)
1989
1990     Conflicts:
1991             src/rgw/rgw_rados.cc - retain d32654b7cd60ccc4e23d3f05b9e4385a697bacd6
1992                 which was merged after this commit
1993
1994 commit 43327f83efcbc5ce54be866a30da8dd1e58d6707
1995 Author: Loic Dachary <ldachary@redhat.com>
1996 Date:   Wed Feb 22 01:49:12 2017 +0100
1997
1998     ceph-disk: dmcrypt activate must use the same cluster as prepare
1999
2000     When dmcrypt is used, the fsid cannot be retrieved from the data
2001     partition because it is encrypted. Store the fsid in the lockbox to
2002     enable dmcrypt activation using the same logic as regular activation.
2003
2004     The fsid is used to retrive the cluster name that was used during
2005     prepare, reason why activation does not and must not have a --cluster
2006     argument.
2007
2008     Fixes: http://tracker.ceph.com/issues/17821
2009
2010     Signed-off-by: Loic Dachary <ldachary@redhat.com>
2011     (cherry picked from commit 7f66672b675abbc0262769d32a38112c781fefac)
2012
2013     Conflicts:
2014             src/ceph-disk/ceph_disk/main.py - in master, self.create_key() takes an
2015                 argument (self.args.cluster) but in jewel it takes no argument
2016
2017 commit c2452c53641fca416268c2e31c774b8b7e609c88
2018 Merge: 82b8c89e47 01d04e28db
2019 Author: Nathan Cutler <presnypreklad@gmail.com>
2020 Date:   Tue Apr 25 07:20:41 2017 +0200
2021
2022     Merge pull request #13608 from smithfarm/wip-19063-jewel
2023
2024     jewel: tests: eliminate race condition in Thrasher constructor
2025
2026     Reviewed-by: Kefu Chai <kchai@redhat.com>
2027     Reviewed-by: David Zafman <dzafman@redhat.com>
2028
2029 commit 3ec1a9bf16e2c305096e11223aaa8db94dc4084d
2030 Author: Sage Weil <sage@redhat.com>
2031 Date:   Fri Mar 31 10:06:42 2017 -0400
2032
2033     ceph_test_librados_api_misc: fix stupid LibRadosMiscConnectFailure.ConnectFailure test
2034
2035     Sometimes the cond doesn't time out and it wakes up instead.  Just repeat
2036     the test many times to ensure that at least once it times out (usually
2037     it doesn't; it's pretty infrequent that it doesn't).
2038
2039     Fixes: http://tracker.ceph.com/issues/15368
2040     Signed-off-by: Sage Weil <sage@redhat.com>
2041     (cherry picked from commit 8bc197400d94ee2716d3f2fa454247379a676cf9)
2042
2043 commit 82b8c89e477610641a6e21123b7c929e9c560729
2044 Merge: 28c7ce595a 30c9527353
2045 Author: Nathan Cutler <presnypreklad@gmail.com>
2046 Date:   Mon Apr 24 22:27:33 2017 +0200
2047
2048     Merge pull request #14195 from cbodley/wip-19353
2049
2050     jewel: rgw: use separate http_manager for read_sync_status
2051
2052     Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2053
2054 commit 28c7ce595ab8fef7c0076f091f249f6b3548ad54
2055 Merge: c05ecff3f0 2e50fe1684
2056 Author: Nathan Cutler <presnypreklad@gmail.com>
2057 Date:   Mon Apr 24 22:26:50 2017 +0200
2058
2059     Merge pull request #14066 from asheplyakov/19321-bp-jewel
2060
2061     jewel: rgw: fix break inside of yield in RGWFetchAllMetaCR
2062
2063     Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2064
2065 commit c05ecff3f0cb5af2f3aa52b68188742796335ddb
2066 Merge: 3240cbf4ef dc4e7a1a86
2067 Author: Nathan Cutler <presnypreklad@gmail.com>
2068 Date:   Mon Apr 24 22:26:14 2017 +0200
2069
2070     Merge pull request #14064 from asheplyakov/19211-bp-jewel
2071
2072     jewel: rgw: "cluster [WRN] bad locator @X on object @X...." in cluster log
2073
2074     Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2075
2076 commit 3240cbf4ef60733f55800039051e85f2a7ace61d
2077 Merge: d4672acbb8 85fbb00f6e
2078 Author: Nathan Cutler <presnypreklad@gmail.com>
2079 Date:   Mon Apr 24 22:25:21 2017 +0200
2080
2081     Merge pull request #13842 from smithfarm/wip-19145-jewel
2082
2083     jewel: rgw: a few cases where rgw_obj is incorrectly initialized
2084
2085     Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2086
2087 commit d4672acbb82daf81eaf259e659dd627ec9f3bc79
2088 Merge: 61ed719d75 ec0668c201
2089 Author: Nathan Cutler <presnypreklad@gmail.com>
2090 Date:   Mon Apr 24 22:24:29 2017 +0200
2091
2092     Merge pull request #13837 from smithfarm/wip-19048-jewel
2093
2094     jewel: rgw: multisite: some yields in RGWMetaSyncShardCR::full_sync() resume in incremental_sync()
2095
2096     Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2097
2098 commit 61ed719d75bf3606fa9e3ae8c8ed776aa2ee4313
2099 Merge: f5e51db564 ced799f9c6
2100 Author: Nathan Cutler <presnypreklad@gmail.com>
2101 Date:   Mon Apr 24 22:23:24 2017 +0200
2102
2103     Merge pull request #13724 from asheplyakov/18626-bp-jewel
2104
2105     jewel: rgw: Use decoded URI when verifying TempURL
2106
2107     Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2108
2109 commit 66c3db7aee5b53f83e87ee8c8f081ab9b0336177
2110 Author: Casey Bodley <cbodley@redhat.com>
2111 Date:   Wed Mar 8 16:31:34 2017 -0500
2112
2113     rgw: data sync skips slo data when syncing the manifest object
2114
2115     Fixes: http://tracker.ceph.com/issues/19027
2116
2117     Signed-off-by: Casey Bodley <cbodley@redhat.com>
2118     (cherry picked from commit 8b69847d7b3e92c70090d1dddf7cea5c44fb6b20)
2119
2120     Conflicts: bucket cleanup, overrides
2121
2122 commit 303a62f7f15c69413165ed604869909587714a94
2123 Author: Casey Bodley <cbodley@redhat.com>
2124 Date:   Mon Mar 13 11:33:02 2017 -0400
2125
2126     rgw: RGWGetObj applies skip_manifest flag to SLO
2127
2128     Signed-off-by: Casey Bodley <cbodley@redhat.com>
2129     (cherry picked from commit 987377ae34382e107e1d54f0bfc1121fcedb4513)
2130
2131 commit f3d99ae8aafaa65e91cd233b5ce8054678d8fa11
2132 Author: Casey Bodley <cbodley@redhat.com>
2133 Date:   Tue Feb 21 10:27:13 2017 -0500
2134
2135     rgw: allow system users to read SLO parts
2136
2137     multisite data sync relies on fetching the object as the system user
2138
2139     Fixes: http://tracker.ceph.com/issues/19027
2140
2141     Signed-off-by: Casey Bodley <cbodley@redhat.com>
2142     (cherry picked from commit d50d18c500fd5dd89e7cada1162cf453b36df370)
2143
2144     Conflicts: auth rework
2145
2146 commit f5e51db5644d4eafdbe3b2d541582a17febed139
2147 Merge: c90cfb7a32 eac0e27193
2148 Author: Nathan Cutler <presnypreklad@gmail.com>
2149 Date:   Fri Apr 21 22:51:59 2017 +0200
2150
2151     Merge pull request #14643 from smithfarm/wip-revert-14427
2152
2153     Wip revert 14427
2154
2155     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2156
2157 commit db053da61807e26876d97786550a769295c7955a
2158 Author: John Spray <john.spray@redhat.com>
2159 Date:   Mon Mar 27 12:56:31 2017 +0100
2160
2161     mds: validate prealloc_inos on sessions after load
2162
2163     Mitigates http://tracker.ceph.com/issues/16842
2164
2165     Signed-off-by: John Spray <john.spray@redhat.com>
2166     (cherry picked from commit c39aaf90ed1b23343eba2b341bb8ee6a50a4ea74)
2167
2168     Conflicts:
2169             src/mds/InoTable.cc - no 5259683e7819c22c14b21b1dd678a33e14574f21 in jewel
2170             src/mds/InoTable.h - no 5259683e7819c22c14b21b1dd678a33e14574f21 in jewel
2171
2172 commit 2b5eb8fa141fa8bd9173dee206ec3530d702fc3a
2173 Author: John Spray <john.spray@redhat.com>
2174 Date:   Mon Mar 27 12:33:59 2017 +0100
2175
2176     mds: operator<< for Session
2177
2178     Use this to get a nice human readable name
2179     when available (also including the session id in
2180     parentheses)
2181
2182     Signed-off-by: John Spray <john.spray@redhat.com>
2183     (cherry picked from commit 0f89787d8312f132ebb621f16c44e950b17a395a)
2184
2185 commit 5b562145197833c6a6f48f006dfb13b6e38f9084
2186 Author: Henrik Korkuc <henrik@uber.com>
2187 Date:   Sun Feb 19 11:44:20 2017 +0200
2188
2189     client/Client.cc: add feature to reconnect client after MDS reset
2190
2191     Client.cc marks session as stale instead of reconecting after received
2192     reset from MDS. On MDS side session is closed so MDS is ignoring cap
2193     renew. This adds option to reconnect stale client sessions instead of
2194     just marking sessions stale.
2195
2196     Fixes: http://tracker.ceph.com/issues/18757
2197
2198     Signed-off-by: Henrik Korkuc <henrik@kirneh.eu>
2199     (cherry picked from commit e0bbc704676ef4aed510daff075ef63c9e73b7b3)
2200
2201 commit 8f21038d30097622e319ab986631b03f87f5d907
2202 Author: Kefu Chai <kchai@redhat.com>
2203 Date:   Tue Mar 21 12:49:45 2017 +0800
2204
2205     doc: cephfs: fix the unexpected indent warning
2206
2207     Signed-off-by: Kefu Chai <kchai@redhat.com>
2208     (cherry picked from commit e423f0b59711422b40c4b3de0bdc73b0947c04d3)
2209
2210 commit f9a1954e113e807b5f9ee6fd56351adc2841730a
2211 Author: Barbora Ančincová <bancinco@redhat.com>
2212 Date:   Thu Feb 16 10:45:36 2017 +0100
2213
2214     doc: additional edits in FUSE client config
2215
2216     Signed-off-by: Bara Ancincova (bara@redhat.com)
2217     (cherry picked from commit b6cad3364c020abd3acf906643fa0b6cbb862a0a)
2218
2219 commit 018649f8a4628881c706a6df9b3eef48403fe91e
2220 Author: Barbora Ančincová <bancinco@redhat.com>
2221 Date:   Thu Jan 26 12:23:34 2017 +0100
2222
2223     doc: Dirty data are not the same as corrupted data
2224
2225     Signed-off-by: Bara Ancincova (bara@redhat.com)
2226     (cherry picked from commit 80db40f8559128baadad42b925ae813e51a31409)
2227
2228 commit 1d8a5b6d64a26a36fb3a227c52908631d95bef79
2229 Author: Barbora Ančincová <bancinco@redhat.com>
2230 Date:   Mon Jan 23 16:34:55 2017 +0100
2231
2232     doc: minor changes in fuse client config reference
2233
2234     Signed-off-by: Bara Ancincova (bara@redhat.com)
2235     (cherry picked from commit e57605681f10436f4b2c85e95179a2904b8c80da)
2236
2237 commit 1ae46b2b94280b8b4a1db89b2cbed8f12bcbcc53
2238 Author: Patrick Donnelly <pdonnell@redhat.com>
2239 Date:   Sun Jul 24 23:21:29 2016 -0400
2240
2241     doc: add client config ref
2242
2243     Fixes: http://tracker.ceph.com/issues/16743
2244
2245     Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2246     (cherry picked from commit 9ad2ccf29830d5309336fc7de877b6926e5dbacd)
2247
2248 commit b8fd297eb64cd29c58596d0a5b0be8fbe9d94a72
2249 Author: Vasu Kulkarni <vasu@redhat.com>
2250 Date:   Tue Apr 11 13:51:47 2017 -0700
2251
2252     use sudo to check check health
2253
2254     Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
2255     (cherry picked from commit 7af157ad4ce7f7e2b8de97ee10eeaf64b9099bc0)
2256
2257 commit 1b91ffc0fbe76c5475f17d4e15ea295ee3680688
2258 Author: Vasu Kulkarni <vasu@redhat.com>
2259 Date:   Wed Mar 29 09:27:20 2017 -0700
2260
2261     Add reboot case for systemd test
2262
2263     test systemd units restart after reboot
2264
2265     Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
2266     (cherry picked from commit 7b587304a54d9b21041ffdfbc85fad8d87859c49)
2267
2268 commit 3d8d1dad8a698a9fb45be7b31c92f8b71ee67720
2269 Author: Vasu Kulkarni <vasu@redhat.com>
2270 Date:   Wed Mar 29 09:56:11 2017 -0700
2271
2272     Fix distro's, point to latest version
2273
2274     Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
2275     (cherry picked from commit 1947648669971c1bd1ca189870ed9b25bbd48d3a)
2276
2277 commit c90cfb7a327fcbb9508e617c2353becb7e5cb45f
2278 Merge: 327276cf3d 9b77b16b88
2279 Author: Nathan Cutler <presnypreklad@gmail.com>
2280 Date:   Thu Apr 20 22:05:57 2017 +0200
2281
2282     Merge pull request #14602 from asheplyakov/19646-jewel
2283
2284     jewel: ceph-disk: enable directory backed OSD at boot time
2285
2286     Reviewed-by: Loic Dachary <ldachary@redhat.com>
2287
2288 commit 327276cf3d4fb7a2558620a8a7d9cc90e5d6e5c5
2289 Merge: 7008c64c51 25e43ac256
2290 Author: Nathan Cutler <presnypreklad@gmail.com>
2291 Date:   Thu Apr 20 22:04:41 2017 +0200
2292
2293     Merge pull request #14449 from smithfarm/wip-test-doc-oversight
2294
2295     tests: fix oversight in yaml comment
2296
2297     Reviewed-by: Loic Dachary <ldachary@redhat.com>
2298
2299 commit f34489dd52f07aeb88e1dbf361ead63f09bb9c65
2300 Author: YunfeiGuan <yunfei.guan@xtaotech.com>
2301 Date:   Mon Apr 10 05:48:47 2017 +0000
2302
2303     cephfs: fix mount point break off problem after mds switch occured
2304
2305     The hot-standby become active as we expected but the mount piont broken strangely
2306     when the active mds is down. The root reason is the new mds use last_cap_renews
2307     decoded from ESesson::replay in find_idle_sessions and wrongly killed the session.
2308     Maybe we should reset session->last_cap_renew to the current time when server send
2309     OPEN to client in reconnect stage.
2310
2311     Fixes: http://tracker.ceph.com/issues/19437
2312     Signed-off-by: Guan yunfei <yunfei.guan@xtaotech.com>
2313     (cherry picked from commit 4ef830c5d6f22bf0d4f82a8624c772ecbbda44a6)
2314
2315     Conflicts:
2316             src/mds/Server.cc (leave '\n' in because jewel does not have
2317             693132eb00b1803d5e97a79908521d5a6903e9f8; jewel ceph_clock_now takes a
2318             CephContext object)
2319
2320 commit 7008c64c5143a3ea19d6679a3e521897f74ff69a
2321 Merge: d62644abcd f8aa6be06c
2322 Author: Nathan Cutler <presnypreklad@gmail.com>
2323 Date:   Thu Apr 20 19:26:52 2017 +0200
2324
2325     Merge pull request #13606 from smithfarm/wip-19062-jewel
2326
2327     jewel: build/ops: enable build of ceph-resource-agents package on rpm-based os
2328
2329     Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
2330
2331 commit d62644abcde0fd99da24922ce7638d173952ceb4
2332 Merge: 8df8960553 482bd1adab
2333 Author: Nathan Cutler <presnypreklad@gmail.com>
2334 Date:   Thu Apr 20 18:18:05 2017 +0200
2335
2336     Merge pull request #14654 from smithfarm/wip-19461-jewel
2337
2338     jewel: admin ops: fix the quota section
2339
2340     Reviewed-by: Casey Bodley <cbodley@redhat.com>
2341
2342 commit c2efeb4b62e6dc4e6975561eb8b20cfca66f7237
2343 Author: David Zafman <dzafman@redhat.com>
2344 Date:   Mon Apr 17 14:58:02 2017 -0700
2345
2346     osd: Give requested scrub work a higher priority
2347
2348     Once started we now queue scrub work at higher priority than
2349     scheduled scrubs.
2350
2351     Fixes: http://tracker.ceph.com/issues/15789
2352
2353     Signed-off-by: David Zafman <dzafman@redhat.com>
2354     (cherry picked from commit ebab8b1f4f67fbdec1e147c580329c1e2b5cf7cd)
2355
2356     Conflicts:
2357             src/osd/OSD.h - in jewel, the PGScrub() call is enclosed within
2358                 op_wq.queue(make_pair(...)) instead of enqueue_back()
2359
2360 commit db86a24e7906ca7c70c60b1752f1230d56361bcd
2361 Author: Yan, Zheng <zyan@redhat.com>
2362 Date:   Wed Feb 22 17:33:05 2017 +0800
2363
2364     client: wait for lastest osdmap when handling set file/dir layout
2365
2366     Fixes: http://tracker.ceph.com/issues/18914
2367     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
2368     (cherry picked from commit 76f5eb86cdd61dde4e6c7cfeb5cf34f0c0334f21)
2369
2370     Conflicts:
2371             src/client/Client.cc (jewel does not have 201c56039)
2372             src/client/Client.h (jewel does not have 201c56039)
2373
2374 commit 7b9283beec0c4bf8e2067558fc4e03e336849b77
2375 Author: Yang Honggang <joseph.yang@xtaotech.com>
2376 Date:   Thu Apr 13 20:09:07 2017 +0800
2377
2378     cephfs: fix write_buf's _len overflow problem
2379
2380     After I have set about 400 64KB xattr kv pair to a file,
2381     mds is crashed. Every time I try to start mds, it will crash again.
2382     The root reason is write_buf._len overflowed when doing
2383     Journaler::append_entry().
2384
2385     This patch try to fix this problem through the following changes:
2386
2387      1. limit file/dir's xattr size
2388      2. throttle journal entry append operations
2389
2390     Fixes: http://tracker.ceph.com/issues/19033
2391     Signed-off-by: Yang Honggang joseph.yang@xtaotech.com
2392     (cherry picked from commit eb915d0eeccbe523f8f70f6571880003ff459459)
2393
2394 commit b52c5088618011a569f157616c5c667c2fc1e9fe
2395 Author: John Spray <john.spray@redhat.com>
2396 Date:   Wed Mar 8 12:13:46 2017 +0000
2397
2398     mds: shut down finisher before objecter
2399
2400     Some of the finisher contexts would try to call into Objecter.
2401     We mostly are protected from this by mds_lock+the stopping
2402     flag, but at the Filer level there's no mds_lock, so in the
2403     case of file size probing we have a problem.
2404
2405     Fixes: http://tracker.ceph.com/issues/19204
2406     Signed-off-by: John Spray <john.spray@redhat.com>
2407     (cherry picked from commit 177a97d5c55ee6a2d5dcd3cf0893546190b10f7a)
2408
2409     Conflicts:
2410             src/mds/MDSRank.cc (no purge_queue.shutdown() in jewel because jewel
2411             does not have 8ebf7d95a9071de24bb1e56a6423c505169cb4de)
2412
2413 commit 96e801fb53941214c669ac01c1804a4d1f27086c
2414 Author: John Spray <john.spray@redhat.com>
2415 Date:   Tue Mar 28 14:13:33 2017 -0400
2416
2417     mds: ignore ENOENT on writing backtrace
2418
2419     We get ENOENT when a pool doesn't exist.  This can
2420     happen because we don't prevent people deleting
2421     former cephfs data pools whose files may not have
2422     had their metadata flushed yet.
2423
2424     http://tracker.ceph.com/issues/19401
2425     Signed-off-by: John Spray <john.spray@redhat.com>
2426     (cherry picked from commit 3fccc2372f2715d075b05e459140360cf6e7ca96)
2427
2428 commit 8df896055366d777ca309dd4a7840f2a2a5a72cc
2429 Merge: da888fa871 7468689314
2430 Author: Nathan Cutler <presnypreklad@gmail.com>
2431 Date:   Thu Apr 20 14:53:30 2017 +0200
2432
2433     Merge pull request #14680 from smithfarm/wip-19711-jewel
2434
2435     jewel: [test] test_notify.py: rbd.InvalidArgument: error updating features for image test_notify_clone2
2436
2437     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2438
2439 commit 74686893147f2d7e1fd1287de37587554fea96c6
2440 Author: Jason Dillaman <dillaman@redhat.com>
2441 Date:   Wed Apr 19 09:26:31 2017 -0400
2442
2443     test: rbd master/slave notify test should test active features
2444
2445     Fixes: http://tracker.ceph.com/issues/19692
2446     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2447     (cherry picked from commit 0dcba41cba96566d0b8da54cf0316d523b88ded2)
2448
2449 commit 7347f1193963666ce07ff9163064aeb0cf343d38
2450 Author: Yan, Zheng <zyan@redhat.com>
2451 Date:   Wed Apr 5 21:29:10 2017 +0800
2452
2453     mds: make C_MDSInternalNoop::complete() delete 'this'
2454
2455     Fixes: http://tracker.ceph.com/issues/19501
2456     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
2457     (cherry picked from commit 424e0c6744e7f63459ca0ff7deab751726aa30cd)
2458
2459     Conflicts:
2460             src/mds/MDSContext.h (omit "override" because jewel does not have
2461             1a91aeab987870b3ccbcf2f1e476fac8b534d449)
2462
2463 commit 824b19a9a66261a90c4e4de64e877a42438febd2
2464 Author: John Spray <john.spray@redhat.com>
2465 Date:   Wed Mar 29 19:38:37 2017 +0100
2466
2467     tools/cephfs: set dir_layout when injecting inodes
2468
2469     When we left this as zero, the MDS would interpret it was HASH_LINUX
2470     rather than the default HASH_RJENKINS.  Potentially that
2471     could cause problems if there perhaps were already dirfrags in
2472     the metadata pool that were set up using rjenkins.  Mainly
2473     it just seems more appropriate to explicitly set this field
2474     rather than hit the fallback behaviour.
2475
2476     Related: http://tracker.ceph.com/issues/19406
2477     Signed-off-by: John Spray <john.spray@redhat.com>
2478     (cherry picked from commit 7d6d542885bd29b71214f9ca52bd26e9183c5d01)
2479
2480 commit eab56dae6714706cf9650dff172b20a764f5481c
2481 Author: John Spray <john.spray@redhat.com>
2482 Date:   Thu Mar 9 13:15:46 2017 +0000
2483
2484     mon: fix hiding mdsmonitor informative strings
2485
2486     Local `stringstream ss` declarations were hiding
2487     the real variable used to feed back to the user.
2488
2489     Fixes: http://tracker.ceph.com/issues/16709
2490     Signed-off-by: John Spray <john.spray@redhat.com>
2491     (cherry picked from commit 00404ae9bd4cce0518a44d36d2d6a5612f4f9d04)
2492
2493 commit da888fa8713178cc23c1069ce2651e02c98aad05
2494 Merge: 8068b546af 2271cd8128
2495 Author: Kefu Chai <tchaikov@gmail.com>
2496 Date:   Thu Apr 20 19:10:33 2017 +0800
2497
2498     Merge pull request #14402 from shinobu-x/wip-17331-jewel
2499
2500      jewel: ceph-disk: ceph-disk list reports mount error for OSD having mount options with SELinux context
2501
2502     Reviewed-by: Loic Dachary <ldachary@redhat.com>
2503     Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
2504     Reviewed-by: Kefu Chai <kchai@redhat.com>
2505
2506 commit 8068b546afbe8e774c9d699205a32215e38c5d4f
2507 Merge: 118ccad0ba 754b4a482c
2508 Author: Nathan Cutler <presnypreklad@gmail.com>
2509 Date:   Thu Apr 20 13:06:33 2017 +0200
2510
2511     Merge pull request #13865 from smithfarm/wip-19158-jewel
2512
2513     jewel: rgw: health check errors out incorrectly
2514
2515     Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2516
2517 commit d57437e338984c5db84f9d16387e082ada5a0a33
2518 Author: John Spray <john.spray@redhat.com>
2519 Date:   Mon Mar 6 11:51:31 2017 +0000
2520
2521     mds: reset heartbeat in export_remaining_imported_caps
2522
2523     This loop can be very long.
2524
2525     Fixes: http://tracker.ceph.com/issues/19118
2526     Signed-off-by: John Spray <john.spray@redhat.com>
2527     (cherry picked from commit 85071f1509beba4a390730e6a3a4332484646d63)
2528
2529 commit 6adf1904ed7209b70328b20f248a701fbdd3127c
2530 Author: John Spray <john.spray@redhat.com>
2531 Date:   Mon Mar 6 11:24:50 2017 +0000
2532
2533     mds: heartbeat_reset in dispatch
2534
2535     Previously we only heartbeated in tick.  However, our locking is
2536     not guaranteed to be fair, so on a super-busy dispatch queue it may be
2537     possible for the heartbeat to time out while the tick() function
2538     is waiting for mds_lock.
2539
2540     Fixes: http://tracker.ceph.com/issues/19118
2541     Signed-off-by: John Spray <john.spray@redhat.com>
2542     (cherry picked from commit 819394549af10532419d88742fae3a69d2ea487d)
2543
2544 commit 63f41d543f8a5f1f55a12612d39c6a2a1cf9c114
2545 Author: Yan, Zheng <zyan@redhat.com>
2546 Date:   Wed Feb 15 11:45:26 2017 +0800
2547
2548     test/libcephfs: avoid buffer overflow when testing ceph_getdents()
2549
2550     The buffer size should be at least "2 * sizeof(struct dirent)".
2551     Otherwise, the code that checks dentry '..' overflow.
2552
2553     Fixes: http://tracker.ceph.com/issues/18941
2554     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
2555     (cherry picked from commit fa6671345b8f3a82dcd232f99e55a982b0a641f1)
2556
2557 commit 7146816065bea55b4e3fec59048a459bcff50f55
2558 Author: Zhi Zhang <willzzhang@tencent.com>
2559 Date:   Fri Feb 10 10:56:46 2017 +0800
2560
2561     mds/StrayManager: aviod reusing deleted inode in StrayManager::_purge_stray_logged
2562
2563     Signed-off-by: Zhi Zhang <zhangz.david@outlook.com>
2564     (cherry picked from commit 4978e57419482384279d7e784a625f5e5c10961a)
2565
2566 commit d8b139b5847cb46aa13486ee85e26f26421f36d6
2567 Author: John Spray <john.spray@redhat.com>
2568 Date:   Wed Feb 1 00:38:08 2017 +0000
2569
2570     tasks/cephfs: switch open vs. write in test_open_inode
2571
2572     Do the write after opening the file, so that we get good
2573     behaviour wrt the change in Mount.open_background that uses
2574     file existence to confirm that the open happened.
2575
2576     Signed-off-by: John Spray <john.spray@redhat.com>
2577     (cherry picked from commit a027dba78fc8bc84ae39d7998b386ce21c01e1bf)
2578
2579 commit e8ae80fc8721e500782c7e87a7cb89128c2a9687
2580 Author: John Spray <john.spray@redhat.com>
2581 Date:   Thu Jan 26 16:48:58 2017 +0000
2582
2583     qa: fix race in Mount.open_background
2584
2585     Previously a later remote call could end up executing
2586     before the remote python program in open_background
2587     had actually got as far as opening the file.
2588
2589     Fixes: http://tracker.ceph.com/issues/18661
2590     Signed-off-by: John Spray <john.spray@redhat.com>
2591     (cherry picked from commit c6d91dd91252e703d08b8ac62ac6a47ee82c0bed)
2592
2593 commit 36c86f71efc10f9a651aedf01dc6fb9d1bfed703
2594 Author: Yan, Zheng <zyan@redhat.com>
2595 Date:   Wed Jan 25 15:28:23 2017 +0800
2596
2597     mds: don't purge strays when mds is in clientreplay state
2598
2599     MDS does not trim log when it's in clientreplay state. If mds hang
2600     at clientreplay state (due to bug), purging strays can submit lots
2601     of log events and create very large mds log.
2602
2603     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
2604     (cherry picked from commit 86bbc7fff02668077f27d0924ba3efe6544b77f6)
2605
2606 commit 8b01cf33575783661a1e1151c8214d327a08b6f9
2607 Author: Yan, Zheng <zyan@redhat.com>
2608 Date:   Wed Jan 25 11:03:45 2017 +0800
2609
2610     mds: skip fragment space check for replayed request
2611
2612     when handling replayed request, stray directory can be different
2613     from the stray directory used by the original request. The fragment
2614     space check for stray directory can fail.
2615
2616     Fixes: http://tracker.ceph.com/issues/18660
2617     Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
2618     (cherry picked from commit afe889cbc5baab196567c2aad01f49fe90901fda)
2619
2620 commit 82ea0971b3cb07c32ec837cb85de63d4068a70d7
2621 Author: Greg Farnum <gfarnum@redhat.com>
2622 Date:   Wed Dec 14 12:09:44 2016 -0800
2623
2624     client: fix the cross-quota rename boundary check conditions
2625
2626     We were previously rejecting a rename if either of the involved directories
2627     was a quota root, even if the other directory was part of the same quota
2628     "tree". What we really want to do is identify the correct quota root
2629     (whether local or ancestral) for each directory and compare them. So
2630     now we do.
2631
2632     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2633     (cherry picked from commit 8e8892aa46accb519faa4bb9fecf66618f1b11b2)
2634
2635     Conflicts:
2636             src/client/Client.cc (do not pass perm to get_quota_root() because
2637             jewel does not have 3caa4d233633fb7a67747f2c79c4a0ab89112294)
2638
2639 commit dbe90c79b86743c7d143d33eb5389fcee0ac76ef
2640 Author: Mykola Golub <mgolub@mirantis.com>
2641 Date:   Tue Apr 11 22:31:43 2017 +0200
2642
2643     librbd: fix rbd_metadata_list and rbd_metadata_get
2644
2645     - properly check for val_len in rbd_metadata_list
2646     - don't expect output buffers are zero pre-filled
2647
2648     Fixes: http://tracker.ceph.com/issues/19588
2649     Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2650     (cherry picked from commit 75afc74ea681402e22b6dec8b83276d145fc786b)
2651
2652 commit 5096fc9c7c62e3043a9a0638eb2516792526fdd7
2653 Author: Jason Dillaman <dillaman@redhat.com>
2654 Date:   Wed Apr 12 10:47:28 2017 -0400
2655
2656     test/librados_test_stub: fixed cls_cxx_map_get_keys/vals return value
2657
2658     Fixes: http://tracker.ceph.com/issues/19597
2659     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2660     (cherry picked from commit 9ffd464dac102f684d6dfa78e58d2cb45e165ed6)
2661
2662 commit 216156b5d4e2666ca592eaab2211940028422bc0
2663 Author: Jason Dillaman <dillaman@redhat.com>
2664 Date:   Fri Mar 10 10:56:38 2017 -0500
2665
2666     rbd: prevent adding multiple mirror peers to a single pool
2667
2668     The rbd-mirror daemon does not currently support replication
2669     from multiple peers. Until that is supported, add a temporary
2670     restriction to prevent confusion.
2671
2672     Fixes: http://tracker.ceph.com/issues/19256
2673     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2674     (cherry picked from commit c0c9d1014d57b3d5b95e7513fcc38d04b9ea5165)
2675
2676 commit c05bd1cb1f570c42b1999fbc70c86b0a72a7fcb3
2677 Author: Orit Wasserman <owasserm@redhat.com>
2678 Date:   Wed Apr 5 13:31:08 2017 +0300
2679
2680     radosgw-admin: use zone id when creating a zone
2681
2682     Fixes: http://tracker.ceph.com/issues/19498
2683     Signed-off-by: Orit Wasserman <owasserm@redhat.com>
2684     (cherry picked from commit 3fea36d635fcba8ca584a1c0ec9f07840009402c)
2685
2686 commit ba81cbbdee1ccf95ceff56eef0a1b2b06be4024a
2687 Author: Casey Bodley <cbodley@redhat.com>
2688 Date:   Tue Mar 14 15:43:13 2017 -0400
2689
2690     qa: rgw task uses period instead of region-map
2691
2692     Signed-off-by: Casey Bodley <cbodley@redhat.com>
2693     (cherry picked from commit e3e3a71d1f1fb43bb4172ce2dfac9a28ca89df0f)
2694
2695 commit a755c95262e10fac774716e460e2ba2eaee9df70
2696 Author: Casey Bodley <cbodley@redhat.com>
2697 Date:   Tue Mar 14 14:18:15 2017 -0400
2698
2699     rgw-admin: remove deprecated regionmap commands
2700
2701     Fixes: http://tracker.ceph.com/issues/18725
2702
2703     Signed-off-by: Casey Bodley <cbodley@redhat.com>
2704     (cherry picked from commit 5830c1849a0c0110d17c37784808e456e6dcb7b3)
2705
2706     Conflicts:
2707             src/rgw/rgw_admin.cc (trivial resolution)
2708
2709 commit 482bd1adabf6f54c8f2bd73090d2e4c02d07f8a3
2710 Author: hrchu <petertc@gmail.com>
2711 Date:   Wed Mar 29 02:17:04 2017 +0000
2712
2713     doc: rgw: correct the quota section
2714
2715     Add the missing option and fix typo.
2716
2717     Fixes: http://tracker.ceph.com/issues/19397
2718
2719     Signed-off-by: Chu, Hua-Rong <hrchu@cht.com.tw>
2720     (cherry picked from commit 51a88267f0d7f51aeb62092949b66b9f6c062e15)
2721
2722 commit 21a83e1276e415e98a3780d0374bb9d4feb191d9
2723 Author: Matt Benjamin <mbenjamin@redhat.com>
2724 Date:   Wed Mar 15 16:35:16 2017 -0400
2725
2726     rgw_file:  remove unused rgw_key variable
2727
2728     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2729     (cherry picked from commit 1100a1c26e76485569cfebcf863b18cf908f6161)
2730
2731 commit ebad040b3ab62e702afc52edd9d99d984b24cdc1
2732 Author: Matt Benjamin <mbenjamin@redhat.com>
2733 Date:   Fri Apr 14 15:56:37 2017 -0400
2734
2735     rgw_file: fix readdir after dirent-change
2736
2737     Also, fixes link count computation off-by-one, update of state.nlink
2738     after computation, link computation reset at start, and a time print
2739     in debug log.
2740
2741     Fixes: http://tracker.ceph.com/issues/19634
2742
2743     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2744
2745     link count
2746
2747     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2748     (cherry picked from commit e0f80266ecd424bf9466579b3edc03911a7c5719)
2749
2750 commit dd9833cacaec20e5bd1a70ec46a427a7352eb5d6
2751 Author: Matt Benjamin <mbenjamin@redhat.com>
2752 Date:   Tue Apr 11 06:42:07 2017 -0400
2753
2754     rgw_file: don't expire directories being read
2755
2756     If a readdir expire event turns out to be older than last_readdir,
2757     just reschedule it (but actually, we should just discard it, as
2758     another expire event must be in queue.
2759
2760     Fixes: http://tracker.ceph.com/issues/19625
2761
2762     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2763     (cherry picked from commit 007b7451c26716c51207c161dc347e9a00da53f1)
2764
2765 commit 30a5e857aade2474c5bd621c57938dfe333c3ae5
2766 Author: Matt Benjamin <mbenjamin@redhat.com>
2767 Date:   Wed Mar 15 16:40:35 2017 -0400
2768
2769     rgw_file:  rgw_readdir:  return dot-dirs only when *offset is 0
2770
2771     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2772     (cherry picked from commit 61482c2b85a07519f2256b1a3f2b6d8aa99d5f06)
2773
2774 commit fe836bfb7a286c87c57576d29d6862be2514ada5
2775 Author: Matt Benjamin <mbenjamin@redhat.com>
2776 Date:   Tue Apr 11 05:56:13 2017 -0400
2777
2778     rgw_file:  chunked readdir
2779
2780     Adjust readdir callback path for new nfs-ganesha chunked readdir,
2781     including changes to respect the result of callback to not
2782     continue.
2783
2784     Pending introduction of offset name hint, our caller will just be
2785     completely enumerating, so it is possible to remove the offset map
2786     and just keep a last offset.
2787
2788     Fixes: http://tracker.ceph.com/issues/19624
2789
2790     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2791     (cherry picked from commit e0191d74e3aef06bf300df045a53a3952a71f651)
2792
2793 commit 16eeb8c2acc8a1f843fa87967c2ee4c0ed2cd8bd
2794 Author: Gui Hecheng <guihecheng@cmss.chinamobile.com>
2795 Date:   Fri Mar 31 10:42:40 2017 +0800
2796
2797     rgw_file: fix missing unlock in unlink
2798
2799     Fixes: http://tracker.ceph.com/issues/19435
2800
2801     Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
2802     (cherry picked from commit cb6808a6366a70f54d0cc16437d16aa1b7819c84)
2803
2804 commit 8c7cb8227c39b09060cdca4e11dec75cf7f2336c
2805 Author: Matt Benjamin <mbenjamin@redhat.com>
2806 Date:   Mon Mar 13 21:52:08 2017 -0400
2807
2808     rgw_file: implement reliable has-children check (unlink dir)
2809
2810     Bug report and discussion provided by
2811     Gui Hecheng <guihecheng@cmss.chinamobile.com> in nfs-ganesha upstream
2812     github.  Briefly, while a reliable check is potentially costly,
2813     it is necessary.
2814
2815     Fixes: http://tracker.ceph.com/issues/19270
2816
2817     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2818     (cherry picked from commit b05f1c6d61aa4501a971e87de6dcaf3e58c3d9b4)
2819
2820 commit 18f14dd86400b50f46930a9ef56666d82035507e
2821 Author: Matt Benjamin <mbenjamin@redhat.com>
2822 Date:   Tue Apr 4 20:16:13 2017 -0400
2823
2824     rgw_file: introduce rgw_lookup type hints
2825
2826     The new type hints optimize object type deduction, when the
2827     rgw_lookup is called from an rgw_readdir callback.
2828
2829     Fixes: http://tracker.ceph.com/issues/19623
2830
2831     Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
2832     (cherry picked from commit 2e66c7a7cc763c5c0d6f5db04855f60f2b2ceed3)
2833
2834 commit 118ccad0ba763b98c9ee3de225a947be5372dc0d
2835 Merge: bf30ecd7cb 65465356b5
2836 Author: Nathan Cutler <presnypreklad@gmail.com>
2837 Date:   Thu Apr 20 11:08:45 2017 +0200
2838
2839     Merge pull request #14383 from smithfarm/wip-19547-jewel
2840
2841     jewel: build/ops: rbdmap.service not included in debian packaging (jewel-only)
2842
2843     Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
2844
2845 commit bf30ecd7cb8dea43a9d55cb1b02c72bfda6bad09
2846 Merge: 376c5e4753 9e123e6d6c
2847 Author: Nathan Cutler <presnypreklad@gmail.com>
2848 Date:   Thu Apr 20 11:02:39 2017 +0200
2849
2850     Merge pull request #14143 from smithfarm/wip-19355-jewel
2851
2852     jewel: rgw: when converting region_map we need to use rgw_zone_root_pool
2853
2854     Reviewed-by: Casey Bodley <cbodley@redhat.com>
2855
2856 commit 376c5e47531a2803ac9b7ea236b989c528268a9f
2857 Merge: da306df4de e2ee70a8ad
2858 Author: Nathan Cutler <presnypreklad@gmail.com>
2859 Date:   Thu Apr 20 11:01:49 2017 +0200
2860
2861     Merge pull request #14136 from smithfarm/wip-19330-jewel
2862
2863     jewel: rgw: upgrade to multisite v2 fails if there is a zone without zone info
2864
2865     Reviewed-by: Casey Bodley <cbodley@redhat.com>
2866
2867 commit da306df4deaaa782c0bad72539affd3ee1ec7efc
2868 Merge: 57b210da41 5ee8feaba4
2869 Author: Nathan Cutler <presnypreklad@gmail.com>
2870 Date:   Thu Apr 20 11:00:35 2017 +0200
2871
2872     Merge pull request #13872 from smithfarm/wip-19163-jewel
2873
2874     jewel: doc: radosgw-admin: add the 'object stat' command to usage
2875
2876     Reviewed-by: Casey Bodley <cbodley@redhat.com>
2877
2878 commit 57b210da4185d624524d05bcd9ad01c2df16ca76
2879 Merge: db92019e9a 9cd7dd8490
2880 Author: Nathan Cutler <presnypreklad@gmail.com>
2881 Date:   Thu Apr 20 10:59:33 2017 +0200
2882
2883     Merge pull request #13863 from smithfarm/wip-19155-jewel
2884
2885     jewel: rgw: typo in rgw_admin.cc
2886
2887     Reviewed-by: Casey Bodley <cbodley@redhat.com>
2888
2889 commit db92019e9af43c2367fdf7865c0fab8cb38453e6
2890 Merge: 6dcd5fa847 6add2a457e
2891 Author: Nathan Cutler <presnypreklad@gmail.com>
2892 Date:   Thu Apr 20 10:56:35 2017 +0200
2893
2894     Merge pull request #13779 from smithfarm/wip-18866-jewel
2895
2896     jewel: rgw: 'radosgw-admin sync status' on master zone of non-master zonegroup
2897
2898     Reviewed-by: Casey Bodley <cbodley@redhat.com>
2899
2900 commit eac0e27193c67ffca60b5e7c61e7769ad8ace6aa
2901 Author: Nathan Cutler <ncutler@suse.com>
2902 Date:   Wed Apr 19 16:53:57 2017 +0200
2903
2904     Revert "osdc/Objecter: If osd full, it should pause read op which w/ rwordered flag."
2905
2906     This reverts commit 2d68822c784eb4d62d3b0198ed4ec04404dbffb3.
2907
2908     Signed-off-by: Nathan Cutler <ncutler@suse.com>
2909
2910 commit 0efe16d2566f0d6040f61fafd38c6661f08da1cd
2911 Author: Nathan Cutler <ncutler@suse.com>
2912 Date:   Wed Apr 19 16:53:49 2017 +0200
2913
2914     Revert "osdc/Objecter: resend RWORDERED ops on full"
2915
2916     This reverts commit f2474042ecd6560323673170c13f2cb964406e70.
2917
2918     Signed-off-by: Nathan Cutler <ncutler@suse.com>
2919
2920 commit 33af18e59298926d9636023faf871a191d582c1a
2921 Author: Nathan Cutler <ncutler@suse.com>
2922 Date:   Tue Apr 18 08:06:01 2017 +0200
2923
2924     doc: mention --show-mappings in crushtool manpage
2925
2926     Fixes: http://tracker.ceph.com/issues/19649
2927     Signed-off-by: Loic Dachary <ldachary@redhat.com>
2928     Signed-off-by: Nathan Cutler <ncutler@suse.com>
2929     (cherry picked from commit b48b6f4ed8b5f5b5852cbbfd5b3d5b650efb7f1b)
2930
2931 commit 6dcd5fa8474c3a43ffca2394557b11f4906ee1ff
2932 Merge: 7c006fc164 68fcb01211
2933 Author: Gregory Farnum <greg@gregs42.com>
2934 Date:   Wed Apr 19 02:47:27 2017 -0400
2935
2936     Merge pull request #14596 from gregsfortytwo/wip-17916-divergent
2937
2938     Wip 17916 divergent
2939
2940     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2941
2942 commit 97cd21afc54efe3afb482b041f9c34ab6cdc682e
2943 Author: Nathan Cutler <ncutler@suse.com>
2944 Date:   Sun Apr 9 20:11:27 2017 +0200
2945
2946     tests: Thrasher: handle "OSD has the store locked" gracefully
2947
2948     On slower machines (VPS, OVH) it takes time for the OSD to go down.
2949
2950     Fixes: http://tracker.ceph.com/issues/19556
2951     Signed-off-by: Nathan Cutler <ncutler@suse.com>
2952     (cherry picked from commit a5b19d2d73540b730392f8001c8601f2cecc1b51)
2953
2954 commit faeb8088ac3bd16bc04a1e5a55fec50285a7253f
2955 Author: Casey Bodley <cbodley@redhat.com>
2956 Date:   Wed Apr 5 16:19:57 2017 -0400
2957
2958     rgw: fix for null version_id in fetch_remote_obj()
2959
2960     commit 8b43c9781206c22d9aedb4beb8d669bf1e23169f fixed the wrong use of
2961     the dest_obj's version, but removed the check for "null" version
2962
2963     Signed-off-by: Casey Bodley <cbodley@redhat.com>
2964     (cherry picked from commit 915370776df5b964c2ee8d9f9329562919eef8d5)
2965
2966 commit 6180fcb4ec33bdade37f5693193712f8016b2560
2967 Author: Zhang Shaowen <zhangshaowen@cmss.chinamobile.com>
2968 Date:   Fri Mar 17 16:26:56 2017 +0800
2969
2970     rgw: version id doesn't work in fetch_remote_obj
2971
2972     Signed-off-by: Zhang Shaowen <zhangshaowen@cmss.chinamobile.com>
2973     (cherry picked from commit 8b43c9781206c22d9aedb4beb8d669bf1e23169f)
2974
2975     Conflicts:
2976             src/rgw/rgw_rados.cc: trivial: dest_obj.key.instance in master
2977               versus dest_obj.get_instance() in Jewel
2978
2979 commit d079b91479abfb474fad4fafe54c119f009ea572
2980 Author: liuchang0812 <liuchang0812@gmail.com>
2981 Date:   Fri Feb 10 18:02:03 2017 +0800
2982
2983     rgw: don't return skew time in pre-signed url
2984
2985     Fixes: http://tracker.ceph.com/issues/18828
2986
2987     Signed-off-by: liuchang0812 <liuchang0812@gmail.com>
2988     (cherry picked from commit dd8b348f4aad0124e8a4457117bf3f5f76af7bdb)
2989
2990 commit 9b77b16b888b8efbf7d50d333e1880a6ec70d87a
2991 Author: Loic Dachary <ldachary@redhat.com>
2992 Date:   Thu Apr 13 23:49:50 2017 +0200
2993
2994     ceph-disk: enable directory backed OSD at boot time
2995
2996     https://github.com/ceph/ceph/commit/539385b143feee3905dceaf7a8faaced42f2d3c6
2997     introduced a regression preventing directory backed OSD from starting at
2998     boot time.
2999
3000     For device backed OSD the boot sequence starts with ceph-disk@.service
3001     and proceeds to
3002
3003         systemctl enable --runtime ceph-osd@.service
3004
3005     where the --runtime ensure ceph-osd@12 is removed when the machine
3006     reboots so that it does not compete with the ceph-disk@/dev/sdb1 unit at
3007     boot time.
3008
3009     However directory backed OSD solely rely on the ceph-osd@.service unit
3010     to start at boot time and will therefore fail to boot.
3011
3012     The --runtime flag is selectively set for device backed OSD only.
3013
3014     Fixes: http://tracker.ceph.com/issues/19628
3015
3016     Signed-off-by: Loic Dachary <loic@dachary.org>
3017     (cherry picked from commit f425a127b7487d2093c8c943f0bcdec3d673d601)
3018
3019     Conflicts:
3020             src/ceph-disk/ceph_disk/main.py: trivial: Jewel does not support
3021                OpenRC and other inits, hence no corresponding 'elif'
3022
3023 commit 68fcb01211e064f0d200cc9c9576254e9a6b949c
3024 Author: Greg Farnum <gfarnum@redhat.com>
3025 Date:   Mon Apr 17 18:09:55 2017 -0700
3026
3027     pglog: require users set a config option before ignoring divergent_priors
3028
3029     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3030
3031 commit b9477303b010b3653934f77fa533df01aeff1c3c
3032 Author: Greg Farnum <gfarnum@redhat.com>
3033 Date:   Fri Apr 7 14:33:20 2017 -0700
3034
3035     osd: pglog: clean up divergent_priors off disk when running; don't assert on startup
3036
3037     Fixes: http://tracker.ceph.com/issues/17916
3038
3039     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3040
3041 commit 7c006fc1640d2f6b9dc9002bfd994ecdf25510ee
3042 Merge: e31a540dce 721b2083cd
3043 Author: Gregory Farnum <greg@gregs42.com>
3044 Date:   Mon Apr 17 17:41:19 2017 -0400
3045
3046     Merge pull request #14492 from gregsfortytwo/wip-jewel-snaptrim
3047
3048     Backport snap trimming improvements to Jewel
3049
3050     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3051
3052 commit 721b2083cd2ed8567d41aba6b17caf30b678e6b3
3053 Merge: a84dc8fe74 e31a540dce
3054 Author: Gregory Farnum <greg@gregs42.com>
3055 Date:   Mon Apr 17 17:39:59 2017 -0400
3056
3057     Merge branch 'jewel' into wip-jewel-snaptrim
3058
3059 commit a84dc8fe747d1e17c7910c9857f7f5d3d56e2dff
3060 Author: Greg Farnum <gfarnum@redhat.com>
3061 Date:   Mon Apr 17 14:32:38 2017 -0700
3062
3063     PendingReleaseNotes: discuss snap trim improvements
3064
3065     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3066
3067 commit 360a9d9af003c650cdf00534909d6488c702c413
3068 Author: Greg Farnum <gfarnum@redhat.com>
3069 Date:   Wed Apr 12 16:30:55 2017 -0700
3070
3071     PrimaryLogPG: reimplement osd_snap_trim_sleep within the state machine
3072
3073     Rather than blocking the main op queue, just pause for that amount of
3074     time between state machine cycles.
3075
3076     Also, add osd_snap_trim_sleep to a few of the thrasher yamls.
3077
3078     Signed-off-by: Samuel Just <sjust@redhat.com>
3079     (cherry picked from commit 2ed7759cfeb03e71f0fbd98fe7c2db2bb741861c)
3080
3081     Conflicts:
3082             src/osd/PrimaryLogPG.cc
3083
3084     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3085     (cherry picked from commit 67336454a4cee66522bc0ca01b2c58b8960f75ec)
3086
3087     Conflicts:
3088             qa/suites/rados/thrash/thrashers/default.yaml
3089             qa/suites/rados/thrash/thrashers/pggrow.yaml
3090             src/osd/OSD.h
3091             src/osd/ReplicatedPG.cc
3092             src/osd/ReplicatedPG.h
3093
3094     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3095
3096 commit 18dbf6a0245e35dbbdb5ddb760182795b37983c0
3097 Author: Samuel Just <sjust@redhat.com>
3098 Date:   Thu Jan 26 15:41:21 2017 -0800
3099
3100     rados: check that pool is done trimming before removing it
3101
3102     Signed-off-by: Samuel Just <sjust@redhat.com>
3103     (cherry picked from commit 4aebf59d906fa3e03d21bdac182f89fe3cd4c802)
3104     (cherry picked from commit 34398c29b3c57f00d932cf96570f882dce64a82b)
3105
3106 commit 7f78450bd184335ba0098f3ded0fefa2c79a5dd5
3107 Author: Greg Farnum <gfarnum@redhat.com>
3108 Date:   Tue Apr 11 14:04:19 2017 -0700
3109
3110     osd/ReplicatedPG: limit the number of concurrently trimming pgs
3111
3112     This patch introduces an AsyncReserver for snap trimming to limit the
3113     number of pgs on any single OSD which can be trimming, as with backfill.
3114     Unlike backfill, we don't take remote reservations on the assumption
3115     that the set of pgs with trimming work to do is already well
3116     distributed, so it doesn't seem worth the implementation overhead to get
3117     reservations from the peers as well.
3118
3119     Signed-off-by: Samuel Just <sjust@redhat.com>
3120     (cherry picked from commit 21cc515adfb225ba70f1d80b1b76f0345c214c22)
3121
3122     Conflicts:
3123             src/osd/PrimaryLogPG.cc
3124             src/osd/PrimaryLogPG.h
3125
3126     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3127
3128     (cherry picked from commit 68ea24396ca6450d4d8042a7c5f51306b7d199fa)
3129     (cherry picked from commit c7176b869898c870b56b1762958652d801af4c4c)
3130
3131     Conflicts: Many. As evidenced by involving two distinct patches
3132     in this one commit, it wasn't a clean backport.
3133
3134     Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3135
3136 commit 527911fab78b4752313a4a2a5d3ab0ae736bc50f
3137 Author: Kefu Chai <kchai@redhat.com>
3138 Date:   Mon Apr 10 14:53:46 2017 +0800
3139
3140     mon/MonClient: make get_mon_log_message() atomic
3141
3142     * LogClient: move reset_session() into get_mon_log_message() and add a
3143       "flush" param to the latter. so it can get_mon_log_message()
3144       atomically. otherwise another call changing the log queue could sneak
3145       in between reset_session() and get_mon_log_message().
3146     * MonClient: add a "flush" param to do_send() so we can reset the
3147       LogClient session once we are connected to a monitor.
3148
3149     Fixes: http://tracker.ceph.com/issues/19427
3150     Signed-off-by: Kefu Chai <kchai@redhat.com>
3151     (cherry picked from commit 5215e291da2b527d85e129eda86043490843178e)
3152
3153     Conflicts:
3154             src/mon/MonClient.cc: handle_auth: replaced 'log_client->reset_session();
3155             send_log();' sequence with newly introduced 'send_log(true);' like
3156             the original patch does
3157
3158 commit e31a540dcea96b3d5b4f7ecd20e2d54e81a68e2b
3159 Merge: 7c36d1650f 06916a8798
3160 Author: Nathan Cutler <presnypreklad@gmail.com>
3161 Date:   Fri Apr 14 22:28:51 2017 +0200
3162
3163     Merge pull request #13834 from smithfarm/wip-18969-jewel
3164
3165     jewel: rgw: Change loglevel to 20 for 'System already converted' message
3166
3167     Reviewed-by: Casey Bodley <cbodley@redhat.com>
3168     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3169
3170 commit 7c36d1650fdb7a357ff724b37cf59b4da413f76b
3171 Merge: 0e3aa2cb01 4c1f302f7d
3172 Author: Nathan Cutler <presnypreklad@gmail.com>
3173 Date:   Fri Apr 14 22:26:51 2017 +0200
3174
3175     Merge pull request #13833 from smithfarm/wip-18908-jewel
3176
3177     jewel: rgw: the swift container acl does not support field .ref
3178
3179     Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
3180
3181 commit b698d1fa4ce4aca5e392eeec600e3357c3cf71a9
3182 Author: Jason Dillaman <dillaman@redhat.com>
3183 Date:   Thu Mar 16 12:28:41 2017 -0400
3184
3185     librbd: is_exclusive_lock_owner API should ping OSD
3186
3187     This is required to detect if a peer has been silently blacklisted
3188     and is therefore no longer the lock owner.
3189
3190     Fixes: http://tracker.ceph.com/issues/19287
3191     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3192     (cherry picked from commit e15db05960a284bdf3701256722299d553cfd5aa)
3193
3194     Conflicts:
3195             src/librbd/ManagedLock.[h|cc]: logic moved to ExclusiveLock
3196
3197     (cherry picked from commit 7e30b630e2806c73ea503871599f958b58df7934)
3198
3199 commit 0e3aa2cb011be8a2af69040b94ee240ea6c1e663
3200 Merge: 8d5a5ddfec 419c9926d9
3201 Author: Nathan Cutler <presnypreklad@gmail.com>
3202 Date:   Thu Apr 13 11:09:27 2017 +0200
3203
3204     Merge pull request #13214 from ovh/bp-osd-updateable-throttles-jewel
3205
3206     jewel: osd: allow client throttler to be adjusted on-fly, without restart
3207
3208     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3209
3210 commit 8d5a5ddfecc7c020eeb8aeb927e007bc04885534
3211 Merge: 091aaa2ab7 bcd3c906e5
3212 Author: Nathan Cutler <presnypreklad@gmail.com>
3213 Date:   Thu Apr 13 10:44:23 2017 +0200
3214
3215     Merge pull request #14326 from shinobu-x/wip-15025-jewel
3216
3217     jewel: osd: new added OSD always down when full flag is set
3218
3219     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3220
3221 commit d311eea6bc5b2c88984cfc16340970ea1459b74a
3222 Author: Robin H. Johnson <robin.johnson@dreamhost.com>
3223 Date:   Sat Feb 11 10:32:53 2017 -0800
3224
3225     msg/simple/Pipe: support IPv6 QoS.
3226
3227     Extend DSCP marking for heartbeat packets to IPv6, as commit
3228     9b9a682fe035c985e416ee1c112fa58f9045a27c only implemented
3229     support for IPv4.
3230
3231     Conflicts: Cherry-picked 91a29bc490fdfbbef0875fa620c7ba1a1a6492ae from master to avoid conflict.
3232     Backport: jewel, luminious
3233     Fixes: http://tracker.ceph.com/issues/18887
3234     Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
3235     (cherry picked from commit 2d6021fbf7a728f73c2998be17e9224f14b83a30)
3236
3237 commit 332b5174c769f395074255e075de8d2cc1ee4021
3238 Author: Jason Dillaman <dillaman@redhat.com>
3239 Date:   Thu Mar 16 12:27:08 2017 -0400
3240
3241     pybind: fix incorrect exception format strings
3242
3243     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3244     (cherry picked from commit 68617455f534a612ade1331f43b032ab524704ae)
3245
3246 commit 091aaa2ab768858e840e2d05e0896c229ce69984
3247 Merge: 3f2e4cd2d5 d30c4d55ad
3248 Author: Nathan Cutler <presnypreklad@gmail.com>
3249 Date:   Wed Apr 12 11:38:48 2017 +0200
3250
3251     Merge pull request #13874 from smithfarm/wip-19171-jewel
3252
3253     jewel: doc: rgw S3 create bucket should not do response in json
3254
3255     Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
3256
3257 commit 3f2e4cd2d53f1cbdbdb1fc8687c5ffe0d46b346b
3258 Merge: ea0bc6c553 0e11a938c5
3259 Author: Nathan Cutler <presnypreklad@gmail.com>
3260 Date:   Wed Apr 12 11:06:40 2017 +0200
3261
3262     Merge pull request #13492 from shinobu-x/wip-18516-jewel
3263
3264     jewel: build/ops: "osd marked itself down" will not recognised if host runs mon + osd on shutdown/reboot
3265
3266     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3267
3268 commit ea0bc6c553454f7641a2594013412aee142fbd11
3269 Merge: 845972f4de d012c381e8
3270 Author: Nathan Cutler <presnypreklad@gmail.com>
3271 Date:   Wed Apr 12 11:01:23 2017 +0200
3272
3273     Merge pull request #13254 from shinobu-x/wip-14609-jewel
3274
3275     jewel: common: radosstriper: protect aio_write API from calls with 0 bytes
3276
3277     Reviewed-by: Kefu Chai <kchai@redhat.com>
3278
3279 commit 845972f4de1d803aa4dab0e1afaed693bbba088f
3280 Merge: a3deef997f cfa37d6a16
3281 Author: Nathan Cutler <presnypreklad@gmail.com>
3282 Date:   Wed Apr 12 10:52:10 2017 +0200
3283
3284     Merge pull request #13489 from shinobu-x/wip-18955-jewel
3285
3286     jewel: ceph-disk: bluestore --setgroup incorrectly set with user
3287
3288     Reviewed-by: Loic Dachary <ldachary@redhat.com>
3289     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3290
3291 commit a3deef997ff0800b9e3e2d141cfc6fdc73cac837
3292 Merge: 702edb5519 39aab763a4
3293 Author: Nathan Cutler <presnypreklad@gmail.com>
3294 Date:   Wed Apr 12 10:51:42 2017 +0200
3295
3296     Merge pull request #14070 from smithfarm/wip-19339-jewel
3297
3298     jewel: tests: dummy suite fails in OpenStack
3299
3300     Reviewed-by: Loic Dachary <ldachary@redhat.com>
3301
3302 commit 702edb5519e67bc5f8c5b65c6f63c9635cd758cf
3303 Merge: f509ccc4b7 a20d2b89ee
3304 Author: Nathan Cutler <presnypreklad@gmail.com>
3305 Date:   Wed Apr 12 10:51:12 2017 +0200
3306
3307     Merge pull request #14329 from smithfarm/wip-19493-jewel
3308
3309     jewel: ceph-disk: Racing between partition creation & device node creation
3310
3311     Reviewed-by: Loic Dachary <ldachary@redhat.com>
3312     Reviewed-by: Sébastien Han <seb@redhat.com>
3313
3314 commit f509ccc4b734697267cec35ff57dd7f1c5aaaddb
3315 Merge: c8c4bff2af f2474042ec
3316 Author: Nathan Cutler <presnypreklad@gmail.com>
3317 Date:   Wed Apr 12 10:48:29 2017 +0200
3318
3319     Merge pull request #14427 from smithfarm/wip-19140-jewel
3320
3321     jewel: osdc/Objecter: If osd full, it should pause read op which w/ rwordered flag
3322
3323     Reviewed-by: Kefu Chai <kchai@redhat.com>
3324
3325 commit c8c4bff2afa7fe0339a97f32e69bb9d0546f1318
3326 Merge: 349baea1a4 b5b441abaa
3327 Author: Nathan Cutler <presnypreklad@gmail.com>
3328 Date:   Wed Apr 12 10:48:08 2017 +0200
3329
3330     Merge pull request #14324 from shinobu-x/wip-19371-jewel
3331
3332     jewel: common: monitor creation with IPv6 public network segfaults
3333
3334     Reviewed-by: Kefu Chai <kchai@redhat.com>
3335
3336 commit 349baea1a4486e475e0381a6b316d64a6ce0139c
3337 Merge: dd466b7d9a 72e2476a13
3338 Author: Nathan Cutler <presnypreklad@gmail.com>
3339 Date:   Wed Apr 12 10:45:35 2017 +0200
3340
3341     Merge pull request #14112 from shinobu-x/wip-19192-jewel
3342
3343     jewel: tools: brag fails to count "in" mds
3344
3345     Reviewed-by: Kefu Chai <kchai@redhat.com>
3346
3347 commit dd466b7d9acb03c8830bdd83b3b73602c68083c2
3348 Merge: b8f8bd0a94 ee06517547
3349 Author: Nathan Cutler <presnypreklad@gmail.com>
3350 Date:   Wed Apr 12 10:45:11 2017 +0200
3351
3352     Merge pull request #14150 from smithfarm/wip-18823-jewel
3353
3354     jewel: tests: run-rbd-unit-tests.sh assert in lockdep_will_lock, TestLibRBD.ObjectMapConsistentSnap
3355
3356     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3357
3358 commit b8f8bd0a949c917b119eca91aec95c6a971a1fb4
3359 Merge: 222916a375 1cc8d0d085
3360 Author: Nathan Cutler <presnypreklad@gmail.com>
3361 Date:   Wed Apr 12 10:44:47 2017 +0200
3362
3363     Merge pull request #14152 from smithfarm/wip-18893-jewel
3364
3365     jewel: librbd: Incomplete declaration for ContextWQ in librbd/Journal.h
3366
3367     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3368
3369 commit 222916a3758b0253bcb851b5e955f6970f171306
3370 Merge: 49f84b1a14 b85677397e
3371 Author: Nathan Cutler <presnypreklad@gmail.com>
3372 Date:   Wed Apr 12 10:44:25 2017 +0200
3373
3374     Merge pull request #14154 from smithfarm/wip-18948-jewel
3375
3376     jewel: tests: additional rbd-mirror test stability improvements
3377
3378     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3379
3380 commit 49f84b1a14295e80ef35204ed691b3131c5f744c
3381 Merge: 2a232d43a1 f6489d01ca
3382 Author: Nathan Cutler <presnypreklad@gmail.com>
3383 Date:   Wed Apr 12 10:44:04 2017 +0200
3384
3385     Merge pull request #14148 from smithfarm/wip-18778-jewel
3386
3387     jewel: rbd: rbd --pool=x rename y z does not work
3388
3389     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3390
3391 commit 2a232d43a18b47399f96abeac3ac800f204a9eaf
3392 Merge: 413ac584d6 8bed107b84
3393 Author: Nathan Cutler <presnypreklad@gmail.com>
3394 Date:   Wed Apr 12 10:43:41 2017 +0200
3395
3396     Merge pull request #14083 from smithfarm/wip-19357-jewel
3397
3398     jewel: rbd: systemctl stop rbdmap unmaps all rbds and not just the ones in /etc/ceph/rbdmap
3399
3400     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3401
3402 commit 413ac584d63fd469ea28defa68c39538444d01b1
3403 Merge: 23d595b1b0 547e867628
3404 Author: Nathan Cutler <presnypreklad@gmail.com>
3405 Date:   Wed Apr 12 10:43:21 2017 +0200
3406
3407     Merge pull request #13154 from smithfarm/wip-18496-jewel
3408
3409     jewel: librbd: Possible deadlock performing a synchronous API action while refresh in-progress
3410
3411     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3412
3413 commit 23d595b1b0fb874adfd1507d86db261cf6a193a9
3414 Merge: 4add6f5580 915dbace5d
3415 Author: Nathan Cutler <presnypreklad@gmail.com>
3416 Date:   Wed Apr 12 10:42:54 2017 +0200
3417
3418     Merge pull request #13244 from smithfarm/wip-18775-jewel
3419
3420     jewel: rbd: qemu crash triggered by network issues
3421
3422     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3423
3424 commit 4add6f5580b7d65571bc426e95fac35be7150ff4
3425 Merge: 37ab19cc29 1a4e1e09b1
3426 Author: Nathan Cutler <presnypreklad@gmail.com>
3427 Date:   Wed Apr 12 10:42:23 2017 +0200
3428
3429     Merge pull request #13809 from asheplyakov/18321-bp-jewel
3430
3431     jewel: librbd: remove image header lock assertions
3432
3433     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3434
3435 commit 37ab19cc29d61970e08af1b4627137cfcaa99474
3436 Merge: f7c04e3ca6 8d0140a9ed
3437 Author: Nathan Cutler <presnypreklad@gmail.com>
3438 Date:   Wed Apr 12 10:41:56 2017 +0200
3439
3440     Merge pull request #13107 from smithfarm/wip-18669-jewel
3441
3442     jewel: tests: [  FAILED  ] TestLibRBD.ImagePollIO in upgrade:client-upgrade-kraken-distro-basic-smithi
3443
3444     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3445
3446 commit f7c04e3ca69c7fc134b267e825277eccea228c9b
3447 Merge: d2909bdb4c a18a2dd108
3448 Author: Nathan Cutler <presnypreklad@gmail.com>
3449 Date:   Wed Apr 12 10:39:26 2017 +0200
3450
3451     Merge pull request #13585 from asheplyakov/jewel-bp-16585
3452
3453     jewel: msg: set close on exec flag
3454
3455     Reviewed-by: Kefu Chai <kchai@redhat.com>
3456
3457 commit d2909bdb4cf135de850cc865f735cc61eb3d4ea6
3458 Merge: cd748603da 6d47615c11
3459 Author: Nathan Cutler <presnypreklad@gmail.com>
3460 Date:   Wed Apr 12 03:56:22 2017 +0200
3461
3462     Merge pull request #14371 from tchaikov/wip-19429-jewel
3463
3464     jewel: tests: clone workunit using the branch specified by task
3465
3466     Nathan Cutler <ncutler@suse.com>
3467
3468 commit cd748603da43c0b7377b5ef07aff618711b4879c
3469 Merge: 1a20c12355 dd25a8f36b
3470 Author: Nathan Cutler <presnypreklad@gmail.com>
3471 Date:   Wed Apr 12 03:54:11 2017 +0200
3472
3473     Merge pull request #14325 from shinobu-x/wip-18619-jewel
3474
3475     jewel: osd: degraded and misplaced status output inaccurate
3476
3477     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3478
3479 commit 1a20c123556e2f566af7e87ab3da6fc2ca328bb9
3480 Merge: 4838c4db4c 7fdf4d41c5
3481 Author: Nathan Cutler <presnypreklad@gmail.com>
3482 Date:   Wed Apr 12 03:53:25 2017 +0200
3483
3484     Merge pull request #14236 from smithfarm/wip-19392-jewel
3485
3486     jewel: mon: remove bad rocksdb option
3487
3488     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3489
3490 commit 4838c4db4c610926edc599677cb7e74f3bdc3077
3491 Merge: e26b7033eb 3860ccf16d
3492 Author: Nathan Cutler <presnypreklad@gmail.com>
3493 Date:   Wed Apr 12 03:52:41 2017 +0200
3494
3495     Merge pull request #14181 from mslovy/wip-19394-jewel
3496
3497     jewel: osd: bypass readonly ops when osd full
3498
3499     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3500
3501 commit e26b7033ebd4a09210d482a54677806fe2c03505
3502 Merge: 389150bcc3 819af9e413
3503 Author: Nathan Cutler <presnypreklad@gmail.com>
3504 Date:   Wed Apr 12 03:49:49 2017 +0200
3505
3506     Merge pull request #14113 from shinobu-x/wip-19319-jewel
3507
3508      jewel: cli: RadosImport::import should return an error if Rados::connect fails
3509
3510     Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
3511     Reviewed-by: David Zafman <dzafman@redhat.com>
3512
3513 commit 389150bcc376702022cf15c0c79dbe856a97dffa
3514 Merge: a8b10082f5 90de64bd81
3515 Author: Nathan Cutler <presnypreklad@gmail.com>
3516 Date:   Wed Apr 12 03:45:33 2017 +0200
3517
3518     Merge pull request #14047 from asheplyakov/reindex-on-pg-split
3519
3520     jewel: osd: reindex properly on pg log split
3521
3522     Reviewed-by: Sage Weil <sage@redhat.com>
3523     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3524
3525 commit a8b10082f5761ea1132b85bd916a84f87e0276e2
3526 Merge: 32ed9b7897 ae498e84ff
3527 Author: Nathan Cutler <presnypreklad@gmail.com>
3528 Date:   Wed Apr 12 03:45:15 2017 +0200
3529
3530     Merge pull request #14044 from mslovy/wip-19311-jewel
3531
3532     jewel: core: os/filestore: fix clang static check warn use-after-free
3533
3534     Reviewed-by: Sage Weil <sage@redhat.com>
3535     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3536
3537 commit 32ed9b789733e21a2d597c7016eabf95763e50ae
3538 Merge: 6705e911a2 335b5fa4a9
3539 Author: Nathan Cutler <presnypreklad@gmail.com>
3540 Date:   Wed Apr 12 03:44:25 2017 +0200
3541
3542     Merge pull request #13932 from asheplyakov/18911-bp-jewel
3543
3544     jewel: rbd-nbd: check /sys/block/nbdX/size to ensure kernel mapped correctly
3545
3546     Reviewed-by: Mykola Golub <mgolub@mirantis.com>
3547     Reviewed-by: Jason Dillaman <dillaman@redhat.com>
3548
3549 commit 6705e911a229cee50d5ac992c7dd9af834f2400d
3550 Merge: 3d21a0080d 714eb863c3
3551 Author: Nathan Cutler <presnypreklad@gmail.com>
3552 Date:   Wed Apr 12 03:42:13 2017 +0200
3553
3554     Merge pull request #13831 from jan--f/wip-19206-jewel
3555
3556     jewel: fs: Invalid error code returned by MDS is causing a kernel client WARNING
3557
3558     Reviewed-by: John Spray <john.spray@redhat.com>
3559
3560 commit 3d21a0080dfd00c0b89ed23fd8049e46c3041af4
3561 Merge: 8a6d64331b 6b5322c5f6
3562 Author: Nathan Cutler <presnypreklad@gmail.com>
3563 Date:   Wed Apr 12 03:41:36 2017 +0200
3564
3565     Merge pull request #13827 from tchaikov/wip-19185-jewel
3566
3567     jewel: osd: ReplicatedPG: try with pool's use-gmt setting if hitset archive not found
3568
3569     Reviewed-by: Sage Weil <sage@redhat.com>
3570     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3571
3572 commit 8a6d64331b73a4a24a59f4dcdb6771f1dc638879
3573 Merge: f96392a589 cebba011e5
3574 Author: Nathan Cutler <presnypreklad@gmail.com>
3575 Date:   Wed Apr 12 03:40:09 2017 +0200
3576
3577     Merge pull request #13788 from shinobu-x/wip-18235-jewel
3578
3579     jewel: core: os/filestore/HashIndex: be loud about splits
3580
3581     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3582
3583 commit f96392a58970ab1e81653d5657d3d680120a02b6
3584 Merge: 8fe6ffcfef 1d054c3856
3585 Author: Nathan Cutler <presnypreklad@gmail.com>
3586 Date:   Wed Apr 12 03:39:27 2017 +0200
3587
3588     Merge pull request #13786 from shinobu-x/wip-19129-jewel
3589
3590     jewel: build/ops: ceph-base package missing dependency for psmisc
3591
3592     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3593
3594 commit 8fe6ffcfef8918fd8634e87255ad3fe7a544aa16
3595 Merge: 6f589a159e b249fd5bd8
3596 Author: Nathan Cutler <presnypreklad@gmail.com>
3597 Date:   Wed Apr 12 03:38:34 2017 +0200
3598
3599     Merge pull request #13732 from liewegas/wip-19119-jewel
3600
3601     jewel: doc: PendingReleaseNotes: note about 19119
3602
3603     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3604
3605 commit 6f589a159e4eb5f6c96634a742acaf6cd6e174c2
3606 Merge: b8f2d35884 8c7a1df251
3607 Author: Nathan Cutler <presnypreklad@gmail.com>
3608 Date:   Wed Apr 12 03:36:21 2017 +0200
3609
3610     Merge pull request #13541 from shinobu-x/wip-18929-jewel
3611
3612      jewel: osd: restrict want_acting to up+acting on recovery completion
3613
3614     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3615     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3616
3617 commit b8f2d35884a52586a28f1ff4eaf99c8c3ba1c43f
3618 Merge: 40d1443d68 2d17092fab
3619 Author: Nathan Cutler <presnypreklad@gmail.com>
3620 Date:   Wed Apr 12 03:34:54 2017 +0200
3621
3622     Merge pull request #13477 from asheplyakov/jewel-bp-18951
3623
3624     jewel: osd: --flush-journal: sporadic segfaults on exit
3625
3626     Reviewed-by: Sage Weil <sage@redhat.com>
3627     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3628
3629 commit 40d1443d68b788737f83c10db3811bc37dbe8dda
3630 Merge: 50e863e0f4 07501dec6f
3631 Author: Nathan Cutler <presnypreklad@gmail.com>
3632 Date:   Wed Apr 12 03:33:25 2017 +0200
3633
3634     Merge pull request #13261 from shinobu-x/wip-18587-jewel
3635
3636     jewel: mon: OSDMonitor: make 'osd crush move ...' work on osds
3637
3638     Reviewed-by: Josh Durgin <jdurgin@redhat.com>
3639     Reviewed-by: Nathan Cutler <ncutler@suse.com>
3640
3641 commit 25e43ac25619d883d5a04e2df1cf6f57fea73fd0
3642 Author: Nathan Cutler <ncutler@suse.com>
3643 Date:   Tue Apr 11 08:23:23 2017 +0200
3644
3645     tests: fix oversight in yaml comment
3646
3647     When the file was copied from the hammer version, the word "hammer"
3648     was not changed to "infernalis".
3649
3650     Signed-off-by: Nathan Cutler <ncutler@suse.com>
3651
3652     This cannot be cherry-picked from master because the test has been dropped.
3653
3654 commit f2474042ecd6560323673170c13f2cb964406e70
3655 Author: Sage Weil <sage@redhat.com>
3656 Date:   Thu Mar 2 21:20:08 2017 -0600
3657
3658     osdc/Objecter: resend RWORDERED ops on full
3659
3660     Our condition for respecting the FULL flag is complex, and involves
3661     the WRITE | RWORDERED flags vs the FULL_FORCE | FULL_TRY flags.  Previously,
3662     we could block a read bc of RWORDRED but not resend it later.
3663
3664     Fix by capturing the complex condition in a respects_full() bool and using
3665     it both for the blocking-on-send and resending-on-possibly-notfull-later
3666     checks.
3667
3668     Fixes: http://tracker.ceph.com/issues/19133
3669     Signed-off-by: Sage Weil <sage@redhat.com>
3670     (cherry picked from commit c4b73f19a7be13ff412eef804efcd8c18ed4dae6)
3671
3672 commit 2d68822c784eb4d62d3b0198ed4ec04404dbffb3
3673 Author: Jianpeng Ma <jianpeng.ma@intel.com>
3674 Date:   Thu May 5 23:44:57 2016 +0800
3675
3676     osdc/Objecter: If osd full, it should pause read op which w/ rwordered flag.
3677
3678     Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
3679     (cherry picked from commit 07b2a22210e26eac1b2825c30629788da05e5e12)
3680
3681 commit 2271cd81282f3f026316134cbab630f3e4f47782
3682 Author: Brad Hubbard <bhubbard@redhat.com>
3683 Date:   Mon Apr 3 13:37:17 2017 +1000
3684
3685     ceph-disk: Populate mount options when running "list"
3686
3687     Also tidy up by moving duplicated code into a function
3688
3689     Fixes: http://tracker.ceph.com/issues/17331
3690     Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
3691     (cherry picked from commit 7943ab2e01e24f2dfc5b6f1d3ffdc8a49e01af45)
3692
3693 commit 65465356b5e7cb6cba67f8cbb81259d21e888dfb
3694 Author: Ken Dreyer <kdreyer@redhat.com>
3695 Date:   Tue Feb 16 12:56:34 2016 -0700
3696
3697     debian: replace SysV rbdmap with systemd service
3698
3699     Stop shipping /etc/init.d/rbdmap in the Debian packages. Ship the
3700     rbdmap.service systemd unit file instead.
3701
3702     The corresponding change has already been made for RPMs, in
3703     9224ac2ad25f7d017916f58b642c0ea25305c3e5.
3704
3705     For Upstart-based systems (eg Ubuntu Trusty), the Debian packages
3706     already contain rbdmap.conf.
3707
3708     (This gets us a tiny bit closer to being able to remove the rbdmap SysV
3709     script from our tree entirely.)
3710
3711     Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
3712     (cherry picked from commit 839807118dda2fb4d57ed9d50ec46e3ee0e2820a)
3713
3714     Conflicts:
3715         debian/ceph-common.install (retain /etc/init.d/rbdmap so jewel users can choose sysvinit or systemd)
3716         debian/rules (retain /etc/init.d/rbdmap so jewel users can choose sysvinit or systemd)
3717
3718 commit 6d47615c11a216733ba368f0dbd9a0a1b9c8fd35
3719 Author: Kefu Chai <kchai@redhat.com>
3720 Date:   Sat Apr 1 23:04:22 2017 +0800
3721
3722     qa/tasks/workunit.py: use "overrides" as the default settings of workunit
3723
3724     otherwise the settings in "workunit" tasks are always overridden by the
3725     settings in template config. so we'd better follow the way of how
3726     "install" task updates itself with the "overrides" settings: it uses the
3727     "overrides" as the *defaults*.
3728
3729     Fixes: http://tracker.ceph.com/issues/19429
3730     Signed-off-by: Kefu Chai <kchai@redhat.com>
3731     (cherry picked from commit 47080150a17d238f38d9da824d227393ad767aad)
3732
3733 commit fdc71e75cd6361be49c4c20e77d3bdff017b38bf
3734 Author: Kefu Chai <kchai@redhat.com>
3735 Date:   Thu Mar 30 12:37:01 2017 +0800
3736
3737     tasks/workunit.py: specify the branch name when cloning a branch
3738
3739     c1309fb failed to specify a branch when cloning using --depth=1, which
3740     by default clones the HEAD. and we can not "git checkout" a specific
3741     sha1 if it is not HEAD, after cloning using '--depth=1', so in this
3742     change, we dispatch "tag", "branch", "HEAD" using three Refspec classes.
3743
3744     Signed-off-by: Kefu Chai <kchai@redhat.com>
3745     Signed-off-by: Dan Mick <dan.mick@redhat.com>
3746     (cherry picked from commit 9ca7ccf5f1739f731da8bf31260594aea3a2932d)
3747
3748     Conflicts:
3749             qa/tasks/workunit.py (trivial resolution)
3750
3751 commit 6c14a803894d71bae858705855551a44bdf9bf67
3752 Author: Dan Mick <dan.mick@redhat.com>
3753 Date:   Tue Mar 28 20:08:13 2017 -0700
3754
3755     tasks/workunit.py: when cloning, use --depth=1
3756
3757     Help avoid killing git.ceph.com.  A depth 1 clone takes about
3758     7 seconds, whereas a full one takes about 3:40 (much of it
3759     waiting for the server to create a huge compressed pack)
3760
3761     Signed-off-by: Dan Mick <dan.mick@redhat.com>
3762     (cherry picked from commit c1309fbef300a062138ac40eb5d3e5081b833072)
3763
3764 commit f8aa6be06cc628b1cf64c9196c30045d020d657e
3765 Author: Nathan Cutler <ncutler@suse.com>
3766 Date:   Fri Apr 7 06:59:13 2017 +0200
3767
3768     build/ops: rpm: move $CEPH_EXTRA_CONFIGURE_ARGS to right place
3769
3770     Signed-off-by: Nathan Cutler <ncutler@suse.com>
3771
3772 commit bb3eb4284fe74cbf78e4a406c7b5f67a8e3c84b3
3773 Author: Nathan Cutler <ncutler@suse.com>
3774 Date:   Fri Apr 7 06:48:51 2017 +0200
3775
3776     build/ops: rpm: explicitly provide --with-ocf to configure
3777
3778     Fixes: http://tracker.ceph.com/issues/19546
3779     Signed-off-by: Nathan Cutler <ncutler@suse.com>
3780
3781     (Note: This cannot be cherry-picked because master uses cmake, but
3782     the fix does bring the jewel spec file into better alignment its master
3783     counterpart, at least as far as this one little bit is concerned.)
3784
3785 commit 30c952735327d05b6049e9364c1a053ebf651751
3786 Author: lu.shasha <lu.shasha@eisoo.com>
3787 Date:   Mon Feb 27 15:52:43 2017 +0800
3788
3789     rgw: use separate http_manager for read_sync_status
3790
3791     concurrent users of read_sync_status() use different cr managers, when get_resource must
3792     use http_manager related to the cr manager.
3793
3794     Fixes: http://tracker.ceph.com/issues/19236
3795
3796     Signed-off-by: Shasha Lu <lu.shasha@eisoo.com>
3797     (cherry picked from commit c412024889f8995d98096ac863bafee71624bd70)
3798
3799 commit 87cb8474f523be1b281882c46a8a3597977a51c9
3800 Author: Casey Bodley <cbodley@redhat.com>
3801 Date:   Thu Jul 28 20:20:29 2016 -0400
3802
3803     rgw: pass cr registry to managers
3804
3805     Signed-off-by: Casey Bodley <cbodley@redhat.com>
3806     (cherry picked from commit ef4d7eab11fb5d2a41c9c28b9eb8b075aaff0d05)
3807
3808      Conflicts:
3809             src/rgw/rgw_rados.cc: removed ref to RGWSyncLogTrimThread (not backported)
3810
3811 commit 1a6d7c0506d4bcda775dda05bb357d4d5695dabb
3812 Author: Casey Bodley <cbodley@redhat.com>
3813 Date:   Fri Jul 22 11:00:16 2016 -0400
3814
3815     rgw: use separate cr manager for read_sync_status
3816
3817     RGWCoroutinesManager::run() is not reentrant, so concurrent users of
3818     read_sync_status() must use different managers
3819
3820     Signed-off-by: Casey Bodley <cbodley@redhat.com>
3821     (cherry picked from commit 6b1e40d7a21c549b55e6576ec56973c8d3c092d1)
3822
3823 commit c466adee9c8f7a97ff7e99bee56da8ce51bf0f00
3824 Author: Casey Bodley <cbodley@redhat.com>
3825 Date:   Thu Jul 21 23:46:20 2016 -0400
3826
3827     rgw: change read_sync_status interface
3828
3829     RGWDataSyncStatusManager::read_sync_status() now operates on the given
3830     parameter, rather than its internal member variable. this allows
3831     multiple concurrent readers, which is needed for the rest interface
3832
3833     Signed-off-by: Casey Bodley <cbodley@redhat.com>
3834     (cherry picked from commit b7cd4e0e8b879b5e528de75bea3307585b96cbf2)
3835
3836 commit 36921a3f3d01547508dc6270f6b19f2576e067d9
3837 Author: Casey Bodley <cbodley@redhat.com>
3838 Date:   Wed Jun 8 11:24:11 2016 -0400
3839
3840     rgw: don't ignore ENOENT in RGWRemoteDataLog::read_sync_status()
3841
3842     rest handlers for sync status need to return ENOENT errors. the only
3843     other callers are in radosgw-admin, so the ENOENT errors are ignored at
3844     those call sites instead
3845
3846     Signed-off-by: Casey Bodley <cbodley@redhat.com>
3847     (cherry picked from commit 2cc533b30c0f23c0750ea8d02c51b3b3d3b4821a)
3848
3849 commit b249fd5bd816a63b445db12c8f846cfda199c8b8
3850 Author: Sage Weil <sage@redhat.com>
3851 Date:   Wed Mar 1 13:18:44 2017 -0600
3852
3853     PendingReleaseNotes: warning about 'osd rm ...' and #19119
3854
3855     See http://tracker.ceph.com/issues/19119
3856
3857     Signed-off-by: Sage Weil <sage@redhat.com>
3858     (cherry picked from commit be96003c464481d8e84825178d600234a0d64d22)
3859
3860 commit 335258f975a8e8539774e8cb22690d746ec90d9f
3861 Author: Ilya Dryomov <idryomov@gmail.com>
3862 Date:   Tue Mar 28 11:49:08 2017 +0200
3863
3864     osdc/Objecter: respect epoch barrier in _op_submit()
3865
3866     Epoch barrier instructs us to avoid sending (i.e. pause) any OSD ops
3867     until we see a barrier epoch.  The only thing epoch_barrier check in
3868     target_should_be_paused() does is keep already paused ops paused.  We
3869     need to actually pause incoming OSD ops in _op_submit().
3870
3871     Fixes: http://tracker.ceph.com/issues/19396
3872     Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
3873     (cherry picked from commit f8e8efc0a53d7bd807cc0c2178aef7c4bed62ab7)
3874
3875 commit a20d2b89ee13e311cf1038c54ecadae79b68abd5
3876 Author: Erwan Velu <erwan@redhat.com>
3877 Date:   Fri Mar 31 14:54:33 2017 +0200
3878
3879     ceph-disk: Adding retry loop in get_partition_dev()
3880
3881     There is very rare cases where get_partition_dev() is called before the actual partition is available in /sys/block/<device>.
3882
3883     It appear that waiting a very short is usually enough to get the partition beein populated.
3884
3885     Analysis:
3886     update_partition() is supposed to be enough to avoid any racing between events sent by parted/sgdisk/partprobe and
3887     the actual creation on the /sys/block/<device>/* entrypoint.
3888     On our CI that race occurs pretty often but trying to reproduce it locally never been possible.
3889
3890     This patch is almost a workaround rather than a fix to the real problem.
3891     It offer retrying after a very short to be make a chance the device to appear.
3892     This approach have been succesful on the CI.
3893
3894     Note his patch is not changing the timing when the device is perfectly created on time and just differ by a 1/5th up to 2 seconds when the bug occurs.
3895
3896     A typical output from the build running on a CI with that code.
3897             command_check_call: Running command: /usr/bin/udevadm settle --timeout=600
3898             get_dm_uuid: get_dm_uuid /dev/sda uuid path is /sys/dev/block/8:0/dm/uuid
3899             get_partition_dev: Try 1/10 : partition 2 for /dev/sda does not in /sys/block/sda
3900             get_partition_dev: Found partition 2 for /dev/sda after 1 tries
3901             get_dm_uuid: get_dm_uuid /dev/sda uuid path is /sys/dev/block/8:0/dm/uuid
3902             get_dm_uuid: get_dm_uuid /dev/sda2 uuid path is /sys/dev/block/8:2/dm/uuid
3903
3904     fixes: #19428
3905
3906     Signed-off-by: Erwan Velu <erwan@redhat.com>
3907     (cherry picked from commit 93e7b95ed8b4c78daebf7866bb1f0826d7199075)
3908
3909 commit 2d5d0aec60ec9689d44a53233268e9b9dd25df95
3910 Author: Erwan Velu <erwan@redhat.com>
3911 Date:   Wed Mar 22 10:11:44 2017 +0100
3912
3913     ceph-disk: Reporting /sys directory in get_partition_dev()
3914
3915     When get_partition_dev() fails, it reports the following message :
3916         ceph_disk.main.Error: Error: partition 2 for /dev/sdb does not appear to exist
3917     The code search for a directory inside the /sys/block/get_dev_name(os.path.realpath(dev)).
3918
3919     The issue here is the error message doesn't report that path when failing while it might be involved in.
3920
3921     This patch is about reporting where the code was looking at when trying to estimate if the partition was available.
3922
3923     Signed-off-by: Erwan Velu <erwan@redhat.com>
3924     (cherry picked from commit 413c9fcfbe8e6ab33d73b8428090ccacc33c5d15)
3925
3926 commit bcd3c906e5b57e7f44df1963c6e11b78ff89482c
3927 Author: Mingxin Liu <mingxin@xsky.com>
3928 Date:   Mon Mar 13 23:41:58 2017 +0800
3929
3930     osd: don't share osdmap with objecter when preboot
3931
3932     Signed-off-by: Mingxin Liu <mingxin@xsky.com>
3933     (cherry picked from commit a5a3644eecc49b4eea890c6999fe87536495dcbe)
3934
3935 commit dd25a8f36bef1901f3ce6193cfcbdaf7ab2424a1
3936 Author: David Zafman <dzafman@redhat.com>
3937 Date:   Wed Jan 18 08:33:40 2017 -0800
3938
3939     osd: Calculate degraded and misplaced more accurately
3940
3941     Calculate num_object_copies based on the larger of pool size,
3942     up set size and acting set size.
3943
3944     Calculate num_objects_degraded as the difference between num_object_copies
3945     and all copies found on acting set and backfilling up set OSDs.
3946
3947     Calculate num_objects_misplaced as all copies on acting set OSDs not in up set
3948     less copies that have been backfilled to up set OSDs.
3949
3950     Fixes: http://tracker.ceph.com/issues/18619
3951
3952     Signed-off-by: David Zafman <dzafman@redhat.com>
3953     (cherry picked from commit 8423bc40759cca137f61e7b755411719a84369d4)
3954
3955 commit b5b441abaa852e85ddefd8b22835c9b85898cc06
3956 Author: Fabian Grünbichler <f.gruenbichler@proxmox.com>
3957 Date:   Wed Mar 22 16:13:50 2017 +0100
3958
3959     common: fix segfault in public IPv6 addr picking
3960
3961     sockaddr is only 16 bytes big, so declaring net as sockaddr
3962     and then casting to sockaddr_in6 in case of IPv6 cannot
3963     work.
3964
3965     using sockaddr_storage works for both IPv4 and IPv6, and is
3966     used in other code parts as well.
3967
3968     note that the tests did not find this issue as they declared
3969     the bigger structs and casted the references to (sockaddr *)
3970
3971     Fixes: http://tracker.ceph.com/issues/19371
3972     Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
3973     (cherry picked from commit ae2ee3d3835fe25b35eeb1a841ee5234cd69eb65)
3974
3975 commit 7fdf4d41c5bef14269cb302301a08d5a3a57a768
3976 Author: Sage Weil <sage@redhat.com>
3977 Date:   Mon May 2 15:29:12 2016 -0400
3978
3979     mon: remove bad rocksdb option
3980
3981     Signed-off-by: Sage Weil <sage@redhat.com>
3982     (cherry picked from commit 0ac671ece258e509f71a05253e62a9878e279840)
3983
3984 commit 3860ccf16d7dfb137face9886c3d7f29cd527835
3985 Author: Jianpeng Ma <jianpeng.ma@intel.com>
3986 Date:   Thu May 5 23:07:06 2016 +0800
3987
3988     osd: bypass readonly ops when osd full.
3989
3990     Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
3991     (cherry picked from commit e2a0ae8e88e6b7354b14adb503fd8ba8525bee39)
3992     See: http://tracker.ceph.com/issues/19394
3993
3994     Signed-off-by: yaoning <yaoning@unitedstack.com>
3995
3996 commit b85677397ef9a3fe16c087e67d3f752851bbe070
3997 Author: Jason Dillaman <dillaman@redhat.com>
3998 Date:   Mon Nov 21 15:31:43 2016 -0500
3999
4000     qa/workunits/rbd: resolve potential rbd-mirror race conditions
4001
4002     Fixes: http://tracker.ceph.com/issues/18935
4003     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4004     (cherry picked from commit 63eae97afc1a92412525468263fb8696a243ebac)
4005
4006 commit 1cc8d0d08560af4e8785d7c4c6a925b1fc9f988e
4007 Author: Boris Ranto <branto@redhat.com>
4008 Date:   Wed Feb 8 23:47:57 2017 +0100
4009
4010     librbd: Include WorkQueue.h since we use it
4011
4012     We use m_work_queue of type ContextWQ in handle_update function but we
4013     do not include common/WorkQueue.h that defines ContextWQ. This results
4014     in dereference of an incomplete type and causes build error in latest
4015     Fedora rawhide (future 26).
4016
4017     Fixes: http://tracker.ceph.com/issues/18862
4018
4019     Signed-off-by: Boris Ranto <branto@redhat.com>
4020     (cherry picked from commit 480f82847ad1fc7959f1fe5a90761a5a24550993)
4021
4022 commit ee06517547ae174472d739f966c0a27d3a97d742
4023 Author: Jason Dillaman <dillaman@redhat.com>
4024 Date:   Wed Jan 18 20:54:22 2017 -0500
4025
4026     librbd: avoid possible recursive lock when racing acquire lock
4027
4028     Fixes: http://tracker.ceph.com/issues/17447
4029     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4030     (cherry picked from commit 5e46e8eb664f573bd70ae7c96a6d9a98b0deb09e)
4031
4032 commit f6489d01ca41d6979b5de28e3cde6b43fcaa8edb
4033 Author: Gaurav Kumar Garg <garg.gaurav52@gmail.com>
4034 Date:   Mon Jan 30 13:03:20 2017 +0100
4035
4036     rbd: destination pool should be source pool if it is not specified
4037
4038     Currently if user perform image rename operation and user give pool
4039     name as a optional parameter (--pool=<pool_name>) then currently
4040     its taking this optional pool name for source pool and making
4041     destination pool name default pool name.
4042     With this fix if user provide pool name as a optional pool name
4043     parameter then it  will consider both soruce and destination pool
4044     name as optional parameter pool name.
4045
4046     Fixes: http://tracker.ceph.com/issues/18326
4047
4048     Reported-by: МАРК КОРЕНБЕРГ <socketpair@gmail.com>
4049     Signed-off-by: Gaurav Kumar Garg <garg.gaurav52@gmail.com>
4050     (cherry picked from commit 01f23aa99fb694da326ab408e75b33c640ce660b)
4051
4052 commit 9e123e6d6c8bbd54514b498df5a22d961f0cefbb
4053 Author: Orit Wasserman <owasserm@redhat.com>
4054 Date:   Sun Mar 12 12:11:28 2017 +0200
4055
4056     rgw: use rgw_zone_root_pool for region_map like is done in hammer
4057
4058     Fixes: http://tracker.ceph.com/issues/19195
4059     Signed-off-by: Orit Wasserman <owasserm@redhat.com>
4060     (cherry picked from commit c91dd6d9efd148e0fe0f027dde537e977de9aa26)
4061
4062 commit e2ee70a8ad51992bbd763d2465f6d8a01dad6a31
4063 Author: Orit Wasserman <owasserm@redhat.com>
4064 Date:   Thu Mar 9 13:03:24 2017 +0200
4065
4066     rgw: skip conversion of zones without any zoneparams
4067
4068     Fixes: http://tracker.ceph.com/issues/19231
4069     Signed-off-by: Orit Wasserman <owasserm@redhat.com>
4070     (cherry picked from commit 36cf5a5c8179c6313346b2e29286c537c6fefce8)
4071
4072 commit c7d292bf6714d7aaf10412e5109badb90f5dc208
4073 Author: Orit Wasserman <owasserm@redhat.com>
4074 Date:   Thu Mar 9 11:16:26 2017 +0200
4075
4076     rgw: better debug information for upgrade
4077
4078     Signed-off-by: Orit Wasserman <owasserm@redhat.com>
4079     (cherry picked from commit e9f3bf8eab1dd46a92f54b0f7afe1f4c0e4204db)
4080
4081 commit 11f5c841c7698e6239017478fa05f742b7c0ab1c
4082 Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
4083 Date:   Tue Jan 31 18:01:32 2017 +0100
4084
4085     rgw/rgw_rados.cc: prefer ++operator for non-primitive iterators
4086
4087     Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
4088     (cherry picked from commit 7086cf9a73f2ec1eb96c0e752beb1b74fca18570)
4089
4090 commit 819af9e4139997cd845dc24a137d43218d8a40a8
4091 Author: Brad Hubbard <bhubbard@redhat.com>
4092 Date:   Tue Mar 21 12:22:20 2017 +1000
4093
4094     tools/rados: Check return value of connect
4095
4096     Fail gracefully if Rados::connect returns an error.
4097
4098     Fixes: http://tracker.ceph.com/issues/19319
4099     Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
4100     (cherry picked from commit c119091ef0844e4a1ddd790a8bfef8f06bb57d58)
4101
4102 commit 72e2476a130e14abcd541ff61328454cb69ad9c3
4103 Author: Kefu Chai <kchai@redhat.com>
4104 Date:   Mon Mar 6 11:33:27 2017 +0800
4105
4106     brag: count the number of mds in fsmap not in mdsmap
4107
4108     this change was introduced in 4e9b953
4109
4110     Fixes: http://tracker.ceph.com/issues/19192
4111     Signed-off-by: Peng Zhang <zphj1987@gmail.com>
4112     (cherry picked from commit 2d25a9c0c760664d3de33ecca0e0272c1031cd46)
4113
4114 commit 3cb192779a1c3662d27bba7715eb31c5f7b6a5b7
4115 Author: Oleh Prypin <oleh@pryp.in>
4116 Date:   Thu Jun 30 00:51:50 2016 +0300
4117
4118     brag: Assume there are 0 MDS instead of crashing when data is missing
4119
4120     Signed-off-by: Oleh Prypin <oleh@pryp.in>
4121
4122 commit 8bed107b84efdc8c735245cdfb51bfd8d07da13b
4123 Author: David Disseldorp <ddiss@suse.de>
4124 Date:   Fri Feb 10 19:19:46 2017 +0100
4125
4126     doc: update description of rbdmap unmap[-all] behaviour
4127
4128     Fixes: http://tracker.ceph.com/issues/18884
4129
4130     Signed-off-by: David Disseldorp <ddiss@suse.de>
4131     (cherry picked from commit f987396e126d5e61240a6645ffed439f79b072b4)
4132
4133 commit da4e0b56c60f4bc2c67daa5dfe4d5255ab8bfc03
4134 Author: Nathan Cutler <ncutler@suse.com>
4135 Date:   Thu Dec 15 18:23:41 2016 +0100
4136
4137     doc: add verbiage to rbdmap manpage
4138
4139     Fixes: http://tracker.ceph.com/issues/18262
4140     Signed-off-by: Nathan Cutler <ncutler@suse.com>
4141     (cherry picked from commit fbac4a081547d83bb2436cd60b0b7ee7250f8a6c)
4142
4143 commit 167d4fd7ccf0cdac536f95250bbfa3e9879ab769
4144 Author: David Disseldorp <ddiss@suse.de>
4145 Date:   Fri Feb 10 17:50:12 2017 +0100
4146
4147     rbdmap: unmap RBDMAPFILE images unless called with unmap-all
4148
4149     When called with a "map" parameter, the rbdmap script iterates the list
4150     of images present in RBDMAPFILE (/etc/ceph/rbdmap), and maps each entry.
4151     When called with "unmap", rbdmap currently iterates *all* mapped RBD
4152     images and unmaps each one, regardless of whether it's listed in the
4153     RBDMAPFILE or not.
4154
4155     This commit adds functionality such that only RBD images listed in the
4156     configuration file are unmapped. This behaviour is the new default for
4157     "rbdmap unmap". A new "unmap-all" parameter is added to offer the old
4158     unmap-all-rbd-images behaviour, which is used by the systemd service.
4159
4160     Fixes: http://tracker.ceph.com/issues/18884
4161
4162     Signed-off-by: David Disseldorp <ddiss@suse.de>
4163     (cherry picked from commit e58413abf408cbe254232e563f3e30d2dc0d707c)
4164
4165 commit 39aab763a44e45e025c311cdfff95116df11a4c4
4166 Author: Nathan Cutler <ncutler@suse.com>
4167 Date:   Thu Dec 15 13:01:02 2016 +0100
4168
4169     Revert "dummy: reduce run time, run user.yaml playbook"
4170
4171     This reverts commit d4e3cec1851ae35889127b90912e133178085bc6.
4172
4173     Fixes: http://tracker.ceph.com/issues/18259
4174     Signed-off-by: Nathan Cutler <ncutler@suse.com>
4175     (cherry picked from commit a8a2a8c3e68b910dbaeb3186576898bf9f89f9fd)
4176
4177 commit 2e50fe1684f73ebe96969c341242b6f20c8470a0
4178 Author: Casey Bodley <cbodley@redhat.com>
4179 Date:   Thu Oct 20 15:01:01 2016 -0400
4180
4181     rgw: fix break inside of yield in RGWFetchAllMetaCR
4182
4183     the yield macro is implemented with for/switch, so the breaks in
4184     RGWFetchAllMetaCR weren't being applied to the for loop as expected -
4185     so any of these breaks send RGWFetchAllMetaCR into an infinite loop
4186
4187     removed the yield {} block, so that breaks will apply to the for loop as
4188     intended, then added a single yield; statement to allow the
4189     entries_index consumer to run one per iteration
4190
4191     Fixes: http://tracker.ceph.com/issues/17655
4192
4193     Signed-off-by: Casey Bodley <cbodley@redhat.com>
4194     (cherry picked from commit 190bd385a7be52867d65740c410884f5c8cbc21f)
4195
4196 commit dc4e7a1a865ea0ae7362c1b6a7a542aa5f72107d
4197 Author: Casey Bodley <cbodley@redhat.com>
4198 Date:   Fri Mar 3 12:10:40 2017 -0500
4199
4200     rgw: delete_system_obj() fails on empty object name
4201
4202     Signed-off-by: Casey Bodley <cbodley@redhat.com>
4203     (cherry picked from commit 67401193f871db95a6045915fa59dce8c5dd1012)
4204
4205 commit e9a577c8c535702d1eb285429978bdbb395e2d5c
4206 Author: Casey Bodley <cbodley@redhat.com>
4207 Date:   Fri Mar 3 11:42:45 2017 -0500
4208
4209     rgw: if user.email is empty, dont try to delete
4210
4211     Fixes: http://tracker.ceph.com/issues/18980
4212
4213     Signed-off-by: Casey Bodley <cbodley@redhat.com>
4214     (cherry picked from commit 022ecf0fcc8e44912c8758ee1d9a452dc23cbbce)
4215
4216 commit 90de64bd81fedcb9540e40d50420e169a4a81248
4217 Author: Alexey Sheplyakov <asheplyakov@mirantis.com>
4218 Date:   Mon Mar 20 14:05:17 2017 +0400
4219
4220     jewel: osd/PGLog: reindex properly on pg log split
4221
4222     When PGLog::IndexedLog::split_into runs it builds the list, which means
4223     the old indices are wrong (point to bad memory), but index() will not
4224     rebuild them because ever since b858e86 we won't rebuild them if they
4225     are already built. Fix that by calling unindex() before the split.
4226
4227     Based on 643ae42cf27f16dd6ed4e1402acc0483bb9fca74. Notice that both
4228     the child and the parent log are re-indexed in Jewel, so the only
4229     problem is missing unindex().
4230
4231     Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
4232
4233 commit ae498e84ffcff7424721f0d2704ec739d1cc092d
4234 Author: liuchang0812 <liuchang0812@gmail.com>
4235 Date:   Tue Dec 20 13:21:40 2016 +0800
4236
4237     os/filestore: fix clang static check warn use-after-free
4238
4239     Signed-off-by: liuchang0812 <liuchang0812@gmail.com>
4240     (cherry picked from commit 1d359455b3dd6abb383542ba596a03f14ac54dbd)
4241     See: http://tracker.ceph.com/issues/19311
4242
4243     Signed-off-by: yaoning <yaoning@unitedstack.com>
4244
4245 commit 335b5fa4a9694620546422f9a02bdcc16549d7cc
4246 Author: Mykola Golub <mgolub@mirantis.com>
4247 Date:   Thu Feb 2 11:11:35 2017 +0100
4248
4249     rbd-nbd: check /sys/block/nbdX/size to ensure kernel mapped correctly
4250
4251     Fixes: http://tracker.ceph.com/issues/18335
4252     Signed-off-by: Mykola Golub <mgolub@mirantis.com>
4253     (cherry picked from commit 596e5ea8a5df72002672eef0a6d20572ca6f60f0)
4254
4255     Conflicts:
4256             qa/workunits/rbd/rbd-nbd.sh: the original commit removes
4257               TOO_LARGE_IMAGE test, do the same thing
4258             src/tools/rbd_nbd/rbd-nbd.cc: help git to add
4259               "include/stringify.h"
4260
4261     Other changes:
4262             src/tools/rbd_nbd/rbd-nbd.cc: #include <fstream> so
4263               the code compiles
4264
4265     Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
4266
4267 commit ced799f9c6558482d538f8dec854c62162685ad0
4268 Author: Michal Koutný <mkoutny@suse.com>
4269 Date:   Thu Jan 26 16:08:09 2017 -0500
4270
4271     rgw: Use decoded URI when verifying TempURL
4272
4273     Instead of calliing url_decode directly, we reuse s->decoded_uri that is
4274     initialized in RGWREST::preprocess().
4275
4276     Fixes: http://tracker.ceph.com/issues/18590
4277
4278     Adapted from 4e1318f4dcbfd64c3ec94f4addf6e38ddd6c013a. Cherry-picking
4279     that patch requires a quite a number of unrelated changes, hence this
4280     patch does s/s->info.request_uri/s->decoded_uri/ to keep the fix as
4281     minimal as possible.
4282
4283     Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
4284
4285 commit 043d70461c1eb874d9185f9bd671930fad05ff65
4286 Author: Ilya Dryomov <idryomov@gmail.com>
4287 Date:   Wed Mar 1 17:19:04 2017 +0100
4288
4289     osd/OSDMap: don't set weight to IN when OSD is destroyed
4290
4291     Since commit 4e28f9e63644 ("osd/OSDMap: clear osd_info, osd_xinfo on
4292     osd deletion"), weight is set to IN when OSD is deleted.  This changes
4293     the result of applying an incremental for clients, not just OSDs.
4294     Because CRUSH computations are obviously affected, pre-4e28f9e63644
4295     servers disagree with post-4e28f9e63644 clients on object placement,
4296     resulting in misdirected requests.
4297
4298     Fixes: http://tracker.ceph.com/issues/19119
4299     Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
4300     (cherry picked from commit a6009d1039a55e2c77f431662b3d6cc5a8e8e63f)
4301
4302 commit 6b5322c5f62f4b90c4206c6ddcc70d090fa7eeb9
4303 Author: Kefu Chai <kchai@redhat.com>
4304 Date:   Tue Mar 7 18:49:46 2017 +0800
4305
4306     osd/ReplicatedPG: try with pool's use-gmt setting if hitset archive not found
4307
4308     due to http://tracker.ceph.com/issues/19185, a hammer OSD could store
4309     pg_hit_set_info_t with "use-gmt = true" even the pool setting is false.
4310     so we use the pool setting as a fallback if the hitset archive is not
4311     found locally and the pool.use_gmt does not match with hitset.use_gmt.
4312
4313     Fixes: http://tracker.ceph.com/issues/19185
4314     Signed-off-by: Kefu Chai <kchai@redhat.com>
4315
4316     Conflicts:
4317       osd/ReplicatedPG.cc:  This cannot be cherry-picked from master
4318         because: hammer should upgrade to jewel first before moving to a
4319         higher version. so there is no necessary to include this workaround
4320         in releases later than jewel.
4321
4322 commit d30c4d55ad52e2b63fdbd06ce256d92fc3fd36c9
4323 Author: Abhishek Lekshmanan <abhishek@suse.com>
4324 Date:   Thu Feb 16 17:40:50 2017 +0100
4325
4326     doc: rgw: make a note abt system users vs normal users
4327
4328     Mention that system users don't behave like normal users in context of
4329     normal rest operations
4330
4331     Fixes: http://tracker.ceph.com/issues/18889
4332     Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
4333     (cherry picked from commit a47bcf70c9f51a6601b809cba219f5615b204d34)
4334
4335     Conflicts:
4336             doc/radosgw/multisite.rst (trivial whitespace difference)
4337
4338 commit 5ee8feaba469886f9e3bd3909475ffef62ba261d
4339 Author: root <root@ceph-node1.homeoffice.wal-mart.com>
4340 Date:   Tue Feb 7 14:37:36 2017 +0530
4341
4342     rgw: Let the object stat command be shown in the usage
4343
4344     Fixes: http://tracker.ceph.com/issues/19013
4345     Signed-off-by: Pavan Rallabhandi <PRallabhandi@walmartlabs.com>
4346     (cherry picked from commit 0fe76f83d19be098ef54fb0492a376fef3aa9e23)
4347
4348 commit 754b4a482cb0369215beed58103a1e241231cf77
4349 Author: root <root@ceph-node1.homeoffice.wal-mart.com>
4350 Date:   Tue Feb 21 16:33:29 2017 +0530
4351
4352     rgw: Correct the return codes for the health check feature
4353     Fixes: http://tracker.ceph.com/issues/19025
4354     Signed-off-by: Pavan Rallabhandi <PRallabhandi@walmartlabs.com>
4355
4356     (cherry picked from commit 4da2bf310f6d43423554c32e43ebf90ad2c3f3a9)
4357
4358 commit 9cd7dd84909abdb9e603ff3aeb9958cdab8c70ad
4359 Author: Ronak Jain <ronakjain@outlook.in>
4360 Date:   Wed Feb 22 12:03:46 2017 +0530
4361
4362     rgw: Fixes typo in rgw_admin.cc
4363
4364     Issue: http://tracker.ceph.com/issues/19026
4365     Signed-off-by: Ronak Jain <ronakjain@outlook.in>
4366     (cherry picked from commit 58837ef6ce8cbcfc2cac29d5f833b2cf62d8737a)
4367
4368 commit 85fbb00f6ef5f11bc5d615ccd8e2202ce3896fd1
4369 Author: Yehuda Sadeh <yehuda@redhat.com>
4370 Date:   Mon Feb 27 10:35:01 2017 -0800
4371
4372     rgw: don't init rgw_obj from rgw_obj_key when it's incorrect to do so
4373
4374     Fixes: http://tracker.ceph.com/issues/19096
4375
4376     rgw_obj_key currently deals with the bucket index key, and not
4377     representing a (name, instance, ns) tupple. Need to initialize
4378     it in two steps.
4379
4380     Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
4381     (cherry picked from commit 392c5d9dae6ba699014ffe6e1e67818fa62d7e41)
4382
4383 commit ec0668c201a71b4a17ef0ab3c5908f57229aa6ef
4384 Author: Casey Bodley <cbodley@redhat.com>
4385 Date:   Tue Nov 29 11:29:41 2016 -0500
4386
4387     rgw: fix for broken yields in RGWMetaSyncShardCR
4388
4389     Fixes: http://tracker.ceph.com/issues/18076
4390
4391     Signed-off-by: Casey Bodley <cbodley@redhat.com>
4392     (cherry picked from commit e62d48a9bf2e309eab1a863f167af5267ebcc371)
4393
4394 commit 6afe3efa4b636ede1cd77086cb2e70ed09fa2e95
4395 Author: Abhishek Lekshmanan <abhishek@suse.com>
4396 Date:   Mon Jul 25 11:21:11 2016 +0200
4397
4398     rgw: kill a compile warning for rgw_sync
4399
4400     killing the compile warning for
4401
4402     ```
4403     /ceph/src/rgw/rgw_sync.cc:1462:12:
4404     warning: suggest explicit braces to avoid ambiguous ‘else’ [-Wparentheses]
4405              if (can_adjust_marker) yield {
4406                 ^
4407     ```
4408
4409     Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
4410     Signed-off-by: Casey Bodley <cbodley@redhat.com>
4411     (cherry picked from commit 831640bb46621a6f003ad562cef7928ffa9a7ad3)
4412
4413 commit 06916a8798439ec033294d791749ce7381d92f51
4414 Author: Vikhyat Umrao <vumrao@redhat.com>
4415 Date:   Mon Feb 13 23:07:25 2017 +0530
4416
4417     rgw: change log level to 20 for 'System already converted' message
4418
4419     Fixes: http://tracker.ceph.com/issues/18919
4420
4421     Signed-off-by: Vikhyat Umrao <vumrao@redhat.com>
4422     (cherry picked from commit 55b567c767830170d04de4cdc8f10aba30a3f379)
4423
4424 commit 4c1f302f7d71bedb0dead220f17eeb84e7e3f737
4425 Author: Jing Wenjun <jingwenjun@cmss.chinamobile.com>
4426 Date:   Wed Jan 11 05:28:43 2017 +0800
4427
4428     rgw: the swift container acl should support field .ref
4429
4430     On the openstack-swift. The container acl supports .ref, which is ignored on ceph swift.
4431
4432     Fixes: http://tracker.ceph.com/issues/18484
4433     Signed-off-by: Jing Wenjun <jingwenjun@cmss.chinamobile.com>
4434     (cherry picked from commit b06f9cd9f0900db7b0d0fbcaea69cdd0d4b10132)
4435
4436     Conflicts:
4437             src/rgw/rgw_acl_swift.cc - no extract_referer_urlspec() in jewel
4438                 see https://github.com/ceph/ceph/pull/8657
4439
4440 commit 714eb863c30df4e653068e6ea16630504e58b704
4441 Author: xie xingguo <xie.xingguo@zte.com.cn>
4442 Date:   Tue Jun 14 19:32:01 2016 +0800
4443
4444     server: negative error code when responding to client
4445
4446     As the comment suggests. Also a zero or positive return code
4447     shall indicates a success, which does not match our intention here.
4448
4449     Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
4450     (cherry picked from commit 26931f888ce4661765cca106b3a3dc66702266df)
4451     Signed-off-by: Jan Fajerski <jfajerski@suse.com>
4452
4453 commit 1a4e1e09b1e562bf97cfe96f5cb9f937b6987165
4454 Author: Jason Dillaman <dillaman@redhat.com>
4455 Date:   Tue Dec 13 14:10:58 2016 -0500
4456
4457     librbd: remove image header lock assertions
4458
4459     This assertions can sporadically fail if the watch is lost and
4460     recovered in the background. Upon a true loss of the lock, the
4461     client would either be blacklisted or it would have completed
4462     all in-flight ops before releasing.
4463
4464     Fixes: http://tracker.ceph.com/issues/18244
4465     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4466     (cherry picked from commit ce4f2a52ec0a794d89e7576b59c9b9aefe3db288)
4467
4468     Conflicts:
4469             src/librbd/operation/SnapshotCreateRequest.cc: rbd class
4470                does not support the snapshot namespaces in Jewel, skip
4471                the corresponding argument
4472
4473 commit cebba011e502f7009208bbddc295eb17f88f1bb9
4474 Author: Dan van der Ster <daniel.vanderster@cern.ch>
4475 Date:   Fri Dec 9 22:06:26 2016 +0100
4476
4477     os/filestore/HashIndex: be loud about splits
4478
4479     Filestore splits are a rare yet important enough event that an
4480     OSD should visibly report when they happen.
4481
4482     Without this reporting an operator could spend hours trying to
4483     understand the cause of any split-induced slow requests.
4484
4485     Fixes: http://tracker.ceph.com/issues/18235
4486     Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch>
4487     (cherry picked from commit 61c47acd3a1f3e01f0106d4a541bb7f28a1301d8)
4488
4489 commit 1d054c3856a63ceebe44f66ff83fda691c374f71
4490 Author: Nathan Cutler <ncutler@suse.com>
4491 Date:   Thu Mar 2 12:41:07 2017 +0100
4492
4493     build/ops: add psmisc dependency to ceph-base (deb and rpm)
4494
4495     Fixes: http://tracker.ceph.com/issues/19129
4496     Signed-off-by: Nathan Cutler <ncutler@suse.com>
4497     (cherry picked from commit 769b695465162bc8424abf8e2f259e6765b5bbff)
4498
4499     Conflicts:
4500             debian/control (jewel does not have f11acf2b 7e71cd2c)
4501
4502 commit 6add2a457e2826b71c0e9e82c6f6686cecbc4584
4503 Author: Jing Wenjun <jingwenjun@cmss.chinamobile.com>
4504 Date:   Fri Nov 25 21:31:22 2016 +0800
4505
4506     rgw: metadata sync info should be shown at master zone of slave zonegroup
4507
4508     When executing 'radosgw-admin sync status', the metadata sync info should be shown on the srceen at master zone of slave zonegroup.
4509
4510     Using the function store->is_meta_master() instead of 'zonegroup.is_master && zone.id == zonegroup.master_zone'
4511
4512     Fixes: http://tracker.ceph.com/issues/18091
4513     Signed-off-by: Jing Wenjun <jingwenjun@cmss.chinamobile.com>
4514     (cherry picked from commit c12d0af2f98b246a73cc3ee027449a22192795b3)
4515
4516 commit 0e11a938c5c9acd8a50efa9a154ea3bf21bcafc5
4517 Author: Boris Ranto <branto@redhat.com>
4518 Date:   Wed Jan 25 12:39:40 2017 +0100
4519
4520     systemd: Start OSDs after MONs
4521
4522     Currently, we start/stop OSDs and MONs simultaneously. This may cause
4523     problems especially when we are shutting down the system. Once the mon
4524     goes down it causes a re-election and the MONs can miss the message
4525     from the OSD that is going down.
4526
4527     Resolves: http://tracker.ceph.com/issues/18516
4528
4529     Signed-off-by: Boris Ranto <branto@redhat.com>
4530     (cherry picked from commit 7f4acf45dd0d86e7d9992a8c30e5876fb57b1914)
4531
4532     Conflicts:
4533             systemd/ceph-osd@.service (jewel does not have 4179aa8d)
4534
4535 commit 3bdd4398f1dcad0b7e22f1750ca524b97feca15a
4536 Author: yaoning <yaoning@unitedstack.com>
4537 Date:   Mon Jun 6 13:31:52 2016 +0800
4538
4539     osd: preserve allocation hint attribute during recovery
4540
4541     Signed-off-by: yaoning <yaoning@unitedstack.com>
4542     (cherry picked from commit e15be792960da6bac2bd469acf7d30007be61781)
4543
4544     Conflicts:
4545         src/osd/ReplicatedBackend.cc (in master, it contains alloc_hint_flags for set_alloc_hint)
4546         src/osd/ReplicatedPG.cc (in master, it contains alloc_hint_flags in object_info_t struct)
4547         src/osd/osd_types.cc (in master, it contains alloc_hint_flags in message serialization)
4548         alloc_hint_flags is used in master bluestore, filestore does not use alloc_hint_flags.
4549         therefore, remove alloc_hint_flags here in jewel
4550
4551     Signed-off-by: yaoning <yaoning@unitedstack.com>
4552
4553 commit 8d0140a9eda814beadf1f59c9b4205f30a1d2e35
4554 Author: Jason Dillaman <dillaman@redhat.com>
4555 Date:   Tue Jan 24 09:24:52 2017 -0500
4556
4557     librbd: improve debug logging for lock / watch state machines
4558
4559     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4560     (cherry picked from commit cc046597983bd491cc66081cc33d9046264fe24b)
4561
4562     Conflicts:
4563         NOTE: cherry-picked from kraken commit to avoid conflicts
4564
4565 commit 62ce3461c3b205eaa9062113526cf572184d0a27
4566 Author: Jason Dillaman <dillaman@redhat.com>
4567 Date:   Mon Jan 23 21:24:41 2017 -0500
4568
4569     test: use librados API to retrieve config params
4570
4571     The CephContext object is not ABI-stable, so it is necessary to
4572     use the "conf_get" librados methods to safely retrieve a setting.
4573
4574     Fixes: http://tracker.ceph.com/issues/18617
4575     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4576     (cherry picked from commit 2ed02f3cd56bf89984c3538ac3f21ec2321cd3b7)
4577
4578     Conflicts:
4579             src/test/librbd/test_librbd.cc (jewel does not have
4580                 006138e2d80b779d8c15b141002bb4b3852f6c4a or
4581                 cb3712e08cdc2c37a983b479f4692bbdfe83b220)
4582
4583 commit 01d04e28db7c2969b86df5b38a20b9eb156cf393
4584 Author: Nathan Cutler <ncutler@suse.com>
4585 Date:   Thu Feb 2 23:23:54 2017 +0100
4586
4587     tests: Thrasher: eliminate a race between kill_osd and __init__
4588
4589     If Thrasher.__init__() spawns the do_thrash thread before initializing the
4590     ceph_objectstore_tool property, do_thrash races with the rest
4591     of Thrasher.__init__() and in some cases do_thrash can call kill_osd() before
4592     Trasher.__init__() progresses much further. This can lead to an exception
4593     ("AttributeError: Thrasher instance has no attribute 'ceph_objectstore_tool'")
4594     being thrown in kill_osd().
4595
4596     This commit eliminates the race by making sure the ceph_objectstore_tool
4597     attribute is initialized before the do_thrash thread is spawned.
4598
4599     Fixes: http://tracker.ceph.com/issues/18799
4600     Signed-off-by: Nathan Cutler <ncutler@suse.com>
4601     (cherry picked from commit b519d38fb1967628ad8a1c46fcfb3f984de58790)
4602
4603     Conflicts:
4604         qa/tasks/ceph_manager.py (jewel has only one if statement after
4605             "self.thread = gevent.spawn(self.do_thrash)" while master has four;
4606             jewel lacks 66836c957ffd974dec136997e23261ec7de2f0aa which disables
4607             ceph-objectstore-tool testing in master)
4608
4609 commit 08a667883b68ccc72e3a4bc3013856deef1df93d
4610 Author: Nathan Cutler <ncutler@suse.com>
4611 Date:   Sat Feb 18 14:33:25 2017 +0100
4612
4613     rpm: build ceph-resource-agents by default
4614
4615     To align with debian build
4616
4617     Fixes: http://tracker.ceph.com/issues/17613
4618     Signed-off-by: Nathan Cutler <ncutler@suse.com>
4619     (cherry picked from commit 3e157bf16c3020ac11cb26df5df3ed331faf3c25)
4620
4621 commit d22becab0f2c541584ce891d392760a5c4f1d153
4622 Author: Yan Jun <yan.jun8@zte.com.cn>
4623 Date:   Thu Jul 14 19:10:29 2016 +0800
4624
4625     msg/simple: cleanups
4626
4627     should save the `errno` which may be changed by `ldout` and/or `<<` operator
4628
4629     Signed-off-by: Yan Jun <yan.jun8@zte.com.cn>
4630     (cherry picked from commit 91a29bc490fdfbbef0875fa620c7ba1a1a6492ae)
4631     Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
4632
4633 commit a18a2dd108678d2e4b57e08b559c1f9a262d6923
4634 Author: Kefu Chai <kchai@redhat.com>
4635 Date:   Fri Jun 17 13:58:55 2016 +0800
4636
4637     msg/simple: set close on exec on server sockets
4638
4639     mds execv() when handling the "respawn" command, to avoid fd leakage,
4640     and enormous CLOSE_WAIT connections after respawning, we need to set
4641     FD_CLOEXEC flag for the socket fds.
4642
4643     Fixes: http://tracker.ceph.com/issues/16390
4644     Signed-off-by: Kefu Chai <kchai@redhat.com>
4645     (cherry picked from commit f019ad563ce90f5aea0d8dd8b7b98688441596e0)
4646
4647 commit 91a968b8fc7b363cae351b8648259211a1e71d18
4648 Author: Kefu Chai <kchai@redhat.com>
4649 Date:   Fri Jun 17 01:17:05 2016 +0800
4650
4651     msg/async: set close on exec on server sockets
4652
4653     mds execv() when handling the "respawn" command, to avoid fd leakage,
4654     and enormous CLOSE_WAIT connections after respawning, we need to set
4655     FD_CLOEXEC flag for the socket fds.
4656
4657     Fixes: http://tracker.ceph.com/issues/16390
4658     Signed-off-by: Kefu Chai <kchai@redhat.com>
4659     (cherry picked from commit eaf68c724144d07f9506037a14d9192cb9f16d70)
4660
4661     Conflicts:
4662             src/msg/async/AsyncMessenger.cc: Processor::accept(): applied
4663                     the hunk manually (invoke set_close_on_exec on a socket
4664                     returned by accept)
4665
4666 commit 547e867628975c7144590e9332aa62b0ef82a433
4667 Author: Jason Dillaman <dillaman@redhat.com>
4668 Date:   Thu Jan 5 12:12:57 2017 -0500
4669
4670     librbd: possible deadlock with flush if refresh in-progress
4671
4672     Fixes: http://tracker.ceph.com/issues/18419
4673     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4674     (cherry picked from commit b95f92a5572d3035c20eba07e76d2c825a9853f7)
4675
4676     Conflicts:
4677             src/librbd/ImageState.h (master commit just adds a function
4678                           declaration, so just add it to jewel as well)
4679
4680 commit 07501dec6f1c70afd4e4c2a50d7f874c39f2220b
4681 Author: Sage Weil <sage@redhat.com>
4682 Date:   Wed Jan 18 17:02:54 2017 -0600
4683
4684     mon/OSDMonitor: make 'osd crush move ...' work on osds
4685
4686     Currently it only allows you to move buckets, which is annoying and much
4687     less useful.  To move an OSD you need to use create-or-move, which is
4688     harder to use.
4689
4690     Fixes: http://tracker.ceph.com/issues/18587
4691     Signed-off-by: Sage Weil <sage@redhat.com>
4692     (cherry picked from commit 47956475dea8bb8e07331dd76344a60b776b5158)
4693
4694     Conflicts:
4695         qa/workunits/mon/crush_ops.sh: adapted "ceph osd find" to jewel syntax
4696
4697 commit 7c6c3c753ccdd3baea834338e1a761f05b4e0a12
4698 Author: Vikhyat Umrao <vumrao@redhat.com>
4699 Date:   Thu Feb 16 23:51:11 2017 +0530
4700
4701     auth: 'ceph auth import -i' overwrites caps, if caps are not specified
4702     in given keyring file, should alert user and should not allow this import.
4703     Because in 'ceph auth list' we keep all the keyrings with caps and importing
4704     'client.admin' user keyring without caps locks the cluster with error[1]
4705     because admin keyring caps are missing in 'ceph auth'.
4706
4707     [1] Error connecting to cluster: PermissionDeniedError
4708
4709     Fixes: http://tracker.ceph.com/issues/18932
4710
4711     Signed-off-by: Vikhyat Umrao <vumrao@redhat.com>
4712     (cherry picked from commit 90144aa64c11a685b6a7cb3aafea75d427f569be)
4713
4714 commit 8c7a1df251e8289e7cf2df5b3096b91d8640695d
4715 Author: Sage Weil <sage@redhat.com>
4716 Date:   Tue Feb 14 15:00:09 2017 -0500
4717
4718     osd/PG: restrict want_acting to up+acting on recovery completion
4719
4720     On recovery completion we recalculate want_acting to see if we
4721     should add recently backfilled osds into acting.  However, at
4722     this point we may have gotten infos from others OSDs outside
4723     of up/acting that could be used for want_acting.  We currently
4724     assert that only up/acting osds are used in
4725     PG::RecoveryState::Active::react(const AdvMap&), so we must
4726     restrict want_acting to up/acting here.
4727
4728     We could remove this restriction, but it would mean
4729
4730     1) checking on every map change that want_acting hasn't been
4731     invalidated, and if so, recalculating want_acting and requesting
4732     a new pg_temp.  Also, presumably
4733
4734     2) on each new info, checking whether we can construct a better
4735     want_acting, and if so, doing it.
4736
4737     That would be a good thing, but is a more complicated change.  In
4738     reality this case comes up very rarely, so simply make our
4739     post-recovery want_acting calculation limit itself to up+acting.
4740
4741     See 1db67c443d84dc5d1ff53cc820fdfd4a2128b680 for the assertion.
4742
4743     Signed-off-by: Sage Weil <sage@redhat.com>
4744     (cherry picked from commit 0f2dee9aa48a00a7f2f809cd4d20e98df771da81)
4745
4746 commit cfa37d6a1674e3f6f8eef4d8519823a7af70df01
4747 Author: craigchi <craig10624@gmail.com>
4748 Date:   Thu Feb 16 19:21:48 2017 +0800
4749
4750     ceph-disk: Fix getting wrong group name when --setgroup in bluestore
4751
4752     ceph-disk prepare --setgroup <GROUP NAME> will be wrong when using with
4753     bluestore
4754
4755     Signed-off-by: craigchi <craig10624@gmail.com>
4756     (cherry picked from commit a8c0870e7370a0948e8e7fd53d3376b85bf9c649)
4757
4758 commit 2d17092fab8080f819369d74d4c76d8ae58d899b
4759 Author: Alexey Sheplyakov <asheplyakov@mirantis.com>
4760 Date:   Tue Feb 7 16:47:45 2017 +0400
4761
4762     ceph-osd: --flush-journal: sporadic segfaults on exit
4763
4764     FileStore holds a number of recources like op thread pool and work
4765     queue, key/value DB threads, etc. These should be properly stopped
4766     (released) before exiting to avoid segfaults on exit.
4767
4768     Note: more code paths (mkfs, dump_journal, etc) need similar fixes,
4769     these will be submitted as separate patches.
4770
4771     Fixes: http://tracker.ceph.com/issues/18820
4772     Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
4773     (cherry picked from commit 00184814c156f6194a6ba4b696073ca1c18a3f8f)
4774
4775     Adjustments:
4776      - release g_ceph_context in the same way as the main code path does
4777
4778 commit d012c381e8c59994ea9a40dc006d23f1bdd6a026
4779 Author: Sebastien Ponce <sebastien.ponce@cern.ch>
4780 Date:   Tue May 10 11:27:59 2016 +0200
4781
4782     radosstriper : protect aio_write API from calls with 0 bytes
4783
4784     an assertion was failing so far, while we only have to return without doing anything
4785
4786     Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch>
4787     (cherry picked from commit 7cce1e8c51640f466d8bb37a21c0d5f1b00db8ab)
4788
4789 commit 915dbace5ddea69fff29f7965f213229b6fbc0ac
4790 Author: Jason Dillaman <dillaman@redhat.com>
4791 Date:   Tue Jan 17 11:55:00 2017 -0500
4792
4793     osdc: cache should ignore error bhs during trim
4794
4795     A read error (such as injecting a timeout into an OSD op) might result
4796     in a bh in an error state. These should be trimable by the cache.
4797
4798     Fixes: http://tracker.ceph.com/issues/18436
4799     Signed-off-by: Jason Dillaman <dillaman@redhat.com>
4800     (cherry picked from commit 5910ed9de9856b5821488a1836487bbbd3d6460e)
4801
4802 commit 419c9926d9ed57cb60228bc95956a9a1471b92cb
4803 Author: Piotr Dałek <piotr.dalek@corp.ovh.com>
4804 Date:   Tue Jan 31 16:07:18 2017 +0100
4805
4806     OSD: allow client throttler to be adjusted on-fly, without restart
4807
4808     This patch allows the osd_client_message_cap and
4809     osd_client_message_size_cap to be adjusted on-fly, using admin socket
4810     functionality.
4811
4812     Fixes: http://tracker.ceph.com/issues/18791
4813     Signed-off-by: Piotr Dałek <piotr.dalek@corp.ovh.com>
4814     (cherry picked from commit 64c309d7e18a975931b526e6f5d6f610c3a0d632)
4815
4816     Conflicts:
4817             src/osd/OSD.cc (suppressed post-jewel option)
4818
4819 commit 957c19b844fb44cde78ad59f872815f82bbf23b8
4820 Author: Ali Maredia <amaredia@redhat.com>
4821 Date:   Thu Nov 10 13:58:35 2016 -0500
4822
4823     swift: added "--cluster" to rgw-admin command for multisite support
4824
4825     Signed-off-by: Ali Maredia <amaredia@redhat.com>
4826
4827 commit 8423bc1eefa45366bdd215a17c61701c9b05dfdd
4828 Merge: 6dc30c4140 c078534376
4829 Author: Andrew Schoen <andrew.schoen@gmail.com>
4830 Date:   Thu Apr 23 14:49:30 2015 -0500
4831
4832     Merge pull request #470 from ceph/wip-remote
4833
4834     Add timeouts to Remote connection functions
4835
4836 commit 6dc30c4140a833fd6cd126f8b5c1eceebad90510
4837 Merge: d55484f9e5 c078534376
4838 Author: Andrew Schoen <andrew.schoen@gmail.com>
4839 Date:   Mon Apr 20 12:38:29 2015 -0500
4840
4841     Merge pull request #466 from ceph/wip-11426
4842
4843     Log stderr in get_latest_image_version_deb()
4844
4845 commit d55484f9e562779e7d47a8f63ff029337dc01eef
4846 Merge: c078534376 cd72cf2b31
4847 Author: Dan Mick <dan.mick@redhat.com>
4848 Date:   Wed Apr 15 10:30:46 2015 -0700
4849
4850     Merge pull request #462 from ceph/wip-ssh-keys
4851
4852     When modifying authorized_keys, store a backup
4853
4854 commit c078534376d594aa3bf70d1d2e5dfc09ca8ae248
4855 Merge: cd72cf2b31 fce2ed683f
4856 Author: Zack Cerza <zack@cerza.org>
4857 Date:   Tue Apr 14 11:38:15 2015 -0600
4858
4859     Merge pull request #460 from zhouyuan/mkdir_p
4860
4861     Make parent directories as needed
4862
4863 commit cd72cf2b3132e7c6371aa4dbfe7564ad3ad0509e
4864 Author: Yehuda Sadeh <yehuda@redhat.com>
4865 Date:   Mon Nov 24 09:59:30 2014 -0800
4866
4867     swift: set full access to subusers creation
4868
4869     Default subuser permissions are 'none'.
4870
4871     Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
4872
4873 commit fce2ed683fcd3798db968a40200f8e8f215595fa
4874 Author: Zack Cerza <zack.cerza@inktank.com>
4875 Date:   Wed Aug 6 10:06:34 2014 -0600
4876
4877     Remove most ceph-specific tasks. They are in ceph-qa-suite now.
4878
4879     Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
4880
4881 commit fac452ae55594aea482db5c13a0bd0207b6ecff6
4882 Author: Zack Cerza <zack@cerza.org>
4883 Date:   Thu Mar 27 11:35:28 2014 -0500
4884
4885     Revert "Lines formerly of the form '(remote,) = ctx.cluster.only(role).remotes.keys()'"
4886
4887     This reverts commit d693b3f8950ffd1f2492a4db0f8234fee31f00f0.
4888
4889 commit e98b107302e06fa5c3c628a7ab0e7455de9ab568
4890 Author: Warren Usui <warren.usui@inktank.com>
4891 Date:   Fri Feb 28 19:13:40 2014 -0800
4892
4893     Lines formerly of the form '(remote,) = ctx.cluster.only(role).remotes.keys()'
4894     and '(remote,) = ctx.cluster.only(role).remotes.iterkeys()' would fail with
4895     ValueError and no message if there were less than 0 or more than 1 key.
4896     Now a new function, get_single_remote_value() is called which prints out
4897     more understandable messages.
4898
4899     Fixes: 7510
4900     Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
4901     Signed-off-by: Warren Usui <warren.usui@inktank.com>
4902
4903 commit e5fe884edfec66cc7e520938bd520aa8f1344f85
4904 Merge: 8e2cdbf5ed f8bf53c4fe
4905 Author: Alfredo Deza <alfredo@deza.pe>
4906 Date:   Wed Feb 19 16:40:17 2014 -0500
4907
4908     Merge pull request #186 from ceph/wip-7369
4909
4910     Fix #7369: "sed expression must be raw string"
4911
4912 commit 8e2cdbf5ede871ebde260e6bdaec13daae03cfc5
4913 Merge: 132f3e8ae6 f8bf53c4fe
4914 Author: Zack Cerza <zack@cerza.org>
4915 Date:   Fri Feb 14 14:59:04 2014 -0600
4916
4917     Merge pull request #188 from ceph/wip-calamari-onefile
4918
4919     Add Calamari test tasks, test script
4920
4921 commit 132f3e8ae664a35cad89896b7300be13b5c604ec
4922 Merge: 24363351c3 7b63876676
4923 Author: Zack Cerza <zack@cerza.org>
4924 Date:   Fri Feb 14 11:50:55 2014 -0600
4925
4926     Merge pull request #192 from ceph/wip-6537-wusui
4927
4928     Readjust the indentation of mon_clock_skew_check.py and mon_thrash.py.
4929
4930 commit 24363351c3b1d23a8d6ef4e3f952855415f97ba8
4931 Merge: 4eb147291f 7b63876676
4932 Author: Zack Cerza <zack@cerza.org>
4933 Date:   Fri Feb 14 11:45:51 2014 -0600
4934
4935     Merge pull request #194 from ceph/wip-6534-wusui
4936
4937     Add docstrings to internal.py
4938
4939 commit 4eb147291fccdb2164a46e977ee43dbf50894e5a
4940 Merge: ad9aaf8fa3 f4284b520a
4941 Author: Zack Cerza <zack@cerza.org>
4942 Date:   Fri Feb 14 11:45:19 2014 -0600
4943
4944     Merge pull request #193 from ceph/wip-6538-wusui
4945
4946     Add doc strings to Swift tests
4947
4948 commit ad9aaf8fa35d49c33373fa69df7b38d3aca6abc5
4949 Merge: 7b63876676 f8bf53c4fe
4950 Author: Zack Cerza <zack@cerza.org>
4951 Date:   Fri Feb 14 11:43:15 2014 -0600
4952
4953     Merge pull request #187 from ceph/wip-better-debug
4954
4955     Debug output improvements
4956
4957 commit f4284b520a554b1cbe130731741e53a7fcf4e35f
4958 Author: Warren Usui <warren.usui@inktank.com>
4959 Date:   Thu Feb 13 21:11:34 2014 -0800
4960
4961     Add doc strings to Swift tests
4962
4963     Fixes: 6538
4964     Signed-off-by: Warren Usui <warren.usui@inktank.com>
4965
4966 commit 7b63876676f1a6845ba3b9147cf7bb2348ef2468
4967 Author: Warren Usui <warren.usui@inktank.com>
4968 Date:   Tue Feb 11 20:21:06 2014 -0800
4969
4970     Add docstrings to s3 related tasks.
4971
4972     Fixes: 6536
4973     Signed-off-by: Warren Usui <warren.usui@inktank.com>
4974
4975 commit f8bf53c4fe52009abdf730b05e2cb2ddbb412dea
4976 Author: Zack Cerza <zack@cerza.org>
4977 Date:   Tue Sep 24 14:19:24 2013 -0500
4978
4979     Fix namespace collision
4980
4981 commit db6efe3e0ba7446ca42baf2a50eef18a10cf4a10
4982 Merge: 611733c8b5 66555a4039
4983 Author: Alfredo Deza <alfredo@deza.pe>
4984 Date:   Tue Sep 24 08:17:22 2013 -0700
4985
4986     Merge pull request #106 from ceph/wip-mirror
4987
4988     Remove lots of ceph.com hardcoding; default to upstream sources
4989
4990 commit 66555a4039c61db9b96d6eecf8d2f298c98b6bad
4991 Author: Zack Cerza <zack@cerza.org>
4992 Date:   Fri Sep 20 15:53:58 2013 -0500
4993
4994     Don't hardcode the git://ceph.com/git/ mirror
4995
4996     Default to https://github.com/ceph/ but add a ~/teuthology.yaml option
4997
4998 commit 611733c8b5ea55206df50c122efac612cb146c2f
4999 Merge: 2346f1d735 6e8a3807c7
5000 Author: Sage Weil <sage@inktank.com>
5001 Date:   Fri Sep 6 13:24:34 2013 -0700
5002
5003     Merge pull request #78 from ceph/wip-6247
5004
5005     Move helper scripts to /usr/local/bin to clean up logs.
5006
5007 commit 6e8a3807c766f728027c3099eebfa24cdc645bd1
5008 Author: Zack Cerza <zack@cerza.org>
5009 Date:   Fri Sep 6 15:08:01 2013 -0500
5010
5011     Helper scripts live in /usr/local/bin now!
5012
5013 commit 2346f1d735ccb40d00b8ff61d4acb446f684b3b2
5014 Author: Joe Buck <jbbuck@gmail.com>
5015 Date:   Fri Aug 23 19:54:53 2013 -0700
5016
5017     s3tests: extend for multi-region tests
5018
5019     Added code to the s3tests task to extract
5020     multi-region info so that that data
5021     can be added to the S3TEST_CONF file
5022     used to run S3 tests.
5023
5024     Signed-off-by: Joe Buck <jbbuck@gmail.com>
5025     Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
5026
5027 commit d3b6d633e161bc422f4781c7be5011f8a7ed8a32
5028 Merge: 442a36c57c 09b01b27a3
5029 Author: Alfredo Deza <alfredo@deza.pe>
5030 Date:   Mon Aug 19 14:10:18 2013 -0700
5031
5032     Merge pull request #41 from ceph/wip-3791
5033
5034     Various usability and documentation fixes
5035
5036 commit 442a36c57c3173b36ce17e5f85a49abf7e80ea93
5037 Merge: 09b01b27a3 9b2c4fa4ad
5038 Author: wusui <warren.usui@inktank.com>
5039 Date:   Fri Aug 16 14:47:59 2013 -0700
5040
5041     Merge pull request #40 from ceph/wip-teutharm-wusui
5042
5043     Wip teutharm wusui
5044
5045 commit 09b01b27a3a1310d4257133def60896ad37fb575
5046 Author: Zack Cerza <zack@cerza.org>
5047 Date:   Thu Aug 15 08:49:35 2013 -0500
5048
5049     Fix some instances where print is being used instead of log
5050
5051 commit 9b2c4fa4ad4c4258b26526afb0c16c71ce47f593
5052 Author: Josh Durgin <josh.durgin@inktank.com>
5053 Date:   Wed Jul 31 13:32:58 2013 -0700
5054
5055     s3/swift tests: call radosgw-admin as the right client
5056
5057     This allows the right region and zone info to be read from ceph.conf
5058
5059     Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
5060
5061 commit 3b3816df58a3ba1f2f850faf8969ad070aa0046e
5062 Author: Yehuda Sadeh <yehuda@inktank.com>
5063 Date:   Thu Jul 25 16:47:34 2013 -0700
5064
5065     s3tests: clone correct branch
5066
5067     Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
5068
5069 commit f5170fb460a2183aed33122b4e6d0117a4220fec
5070 Merge: 7207a31e58 bd56af707a
5071 Author: Sandon Van Ness <sandon@inktank.com>
5072 Date:   Thu Jul 25 19:50:39 2013 -0700
5073
5074     Merge branch 'master' of github.com:ceph/teuthology
5075
5076 commit 7207a31e5812dd0b29f2d6378360015622ddf4aa
5077 Merge: bd56af707a 3da945512e
5078 Author: Sandon Van Ness <sandon@inktank.com>
5079 Date:   Thu Jul 25 19:50:02 2013 -0700
5080
5081     Merge remote-tracking branch 'origin/wip-sandon-vm'
5082
5083     Conflicts:
5084             teuthology/lock.py
5085             teuthology/misc.py
5086             teuthology/task/install.py
5087
5088 commit bd56af707ae50c98ec46344cf57eb333061847b1
5089 Merge: 3da945512e 343a42c0d8
5090 Author: Josh Durgin <josh.durgin@inktank.com>
5091 Date:   Fri Jul 19 14:44:51 2013 -0700
5092
5093     Merge branch 'wip-centos-rgw'
5094
5095 commit 343a42c0d86af5b8630a30716c03fc84ba22f944
5096 Author: Josh Durgin <josh.durgin@inktank.com>
5097 Date:   Tue Jul 9 18:50:52 2013 -0700
5098
5099     s3tests: fix client configurations that aren't dictionaries
5100
5101     They're always used as dictionaries later on.
5102
5103     Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
5104
5105 commit 3da945512ed78081be5dbe9ba59c836a311e1973
5106 Merge: 2c34d1971f 253cc98d98
5107 Author: Sage Weil <sage@inktank.com>
5108 Date:   Mon Jun 24 16:18:36 2013 -0700
5109
5110     Merge pull request #15 from ceph/wip-ulimits
5111
5112     Reviewed-by: Warren Usui <warren.usui@inktank.com>
5113
5114 commit 253cc98d98855d65be7ebcdd46a39aa1004f8e67
5115 Author: Sage Weil <sage@inktank.com>
5116 Date:   Sun Jun 23 09:15:28 2013 -0700
5117
5118     enable-coredump -> adjust-ulimits
5119
5120     and set max_files to be big, too!
5121
5122 commit 2c34d1971f1e82311b364bf8efe60b223158d676
5123 Merge: 61dba20d1f b366ad334a
5124 Author: Warren Usui <warren.usui@inktank.com>
5125 Date:   Tue May 7 19:27:51 2013 -0700
5126
5127     Merge branch 'wip-teuth4768a-wusui'
5128
5129     Conflicts:
5130             teuthology/task/install.py
5131
5132 commit 61dba20d1fce774eac5b56f0d61d4229460875c0
5133 Merge: a9f3eb6310 5a7267f85c
5134 Author: Sage Weil <sage@inktank.com>
5135 Date:   Mon May 6 21:31:36 2013 -0700
5136
5137     Merge branch 'next'
5138
5139 commit a9f3eb631064931cbdde7ef218c16e154bdb9991
5140 Author: Sage Weil <sage@inktank.com>
5141 Date:   Thu May 2 13:47:46 2013 -0700
5142
5143     s3tests: add force-branch with higher precdence than 'branch'
5144
5145     This way we can force a branch despite something in overrides.
5146
5147     Signed-off-by: Sage Weil <sage@inktank.com>
5148
5149 commit b366ad334af55867ba781e22c8f87b6ac7775bf2
5150 Merge: 2a51e32891 5a7267f85c
5151 Author: Josh Durgin <josh.durgin@inktank.com>
5152 Date:   Wed May 1 09:52:02 2013 -0700
5153
5154     Merge remote branch 'origin/next'
5155
5156 commit 5a7267f85c80f88aca1b0081b07de1de3909f2e7
5157 Author: Josh Durgin <josh.durgin@inktank.com>
5158 Date:   Tue Apr 30 17:07:53 2013 -0700
5159
5160     fix some errors found by pyflakes
5161
5162     Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
5163
5164 commit f866037f045887ccc5da15404935ce3361a74a08
5165 Author: Josh Durgin <josh.durgin@inktank.com>
5166 Date:   Tue Apr 30 13:23:22 2013 -0700
5167
5168     s3tests: revert useless portion of 1c50db6a4630d07e72144dafd985c397f8a42dc5
5169
5170     Perhaps it was attempting to debug something, but it shouldn't have been committed.
5171
5172     Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
5173
5174 commit 2dcce57c00264b85cfb906223dfb89db9cc61ba5
5175 Author: Josh Durgin <josh.durgin@inktank.com>
5176 Date:   Tue Apr 30 16:49:04 2013 -0700
5177
5178     rgw tests: remove users after each test
5179
5180     These should all be cleanup up at some point. They're
5181     almost all the same code.
5182
5183     Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
5184
5185 commit 3c604251d9014dba45f3adbb2008bac16ff6346d
5186 Author: Josh Durgin <josh.durgin@inktank.com>
5187 Date:   Tue Apr 30 16:47:34 2013 -0700
5188
5189     rgw tests: clean up immediately after the test
5190
5191     There's no need for an explicit cleanup function, so move it back
5192     to where it came from (except in s3roundtrip, which did not have it).
5193
5194     Instead, since these use a nested contextmanager, pass through
5195     and yield to the top-level run_tasks after the nested
5196     contextmanager has finished (and thus run all the cleanup steps
5197     in the subtasks for this test).
5198
5199     Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
5200
5201 commit 022bd4aa42312b80917169282cbfba655bbad6f1
5202 Author: Yehuda Sadeh <yehuda@inktank.com>
5203 Date:   Tue Apr 30 07:06:03 2013 -0700
5204
5205     swift, s3readwrite: add missing yield
5206
5207     Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
5208
5209 commit 820c72b8d0177b01f69887ec64d98702db37077c
5210 Author: Yehuda Sadeh <yehuda@inktank.com>
5211 Date:   Mon Apr 29 11:24:04 2013 -0700
5212
5213     s3tests, s3readwrite, swift: cleanup explicitly
5214
5215     Cleaning up test dir explicitly after run, so that
5216     consecutive runs don't fail.
5217
5218     Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
5219
5220 commit 2a51e328913f16917a881129b475a4aeeab24ed0
5221 Merge: cccadb9b03 617534e769
5222 Author: Yehuda Sadeh <yehuda@inktank.com>
5223 Date:   Wed Feb 20 14:10:50 2013 -0800
5224
5225     Merge remote-tracking branch 'origin/wip-3634'
5226
5227 commit cccadb9b03ca4421a5fd841a61bf252c329e3649
5228 Merge: 3eb19c8107 fa1f89478a
5229 Author: Sage Weil <sage@inktank.com>
5230 Date:   Tue Feb 19 21:04:24 2013 -0800
5231
5232     Merge branch 'unstable'
5233
5234     Conflicts:
5235             teuthology/task/workunit.py
5236
5237 commit fa1f89478a76373cb33cf2524e2ebf68b3cd622c
5238 Author: Sander Pool <sander.pool@inktank.com>
5239 Date:   Wed Feb 6 19:16:52 2013 +0000
5240
5241     Install ceph debs and use installed debs
5242
5243     The ceph task installs ceph using the debian
5244     packages now, and all invocations of binaries installed
5245     in {tmpdir}/binary/usr/local/bin/ are replace with
5246     the use of the binaries installed in standard locations
5247     by the debs.
5248
5249     Author:    Sander Pool <sander.pool@inktank.com>
5250     Signed-off-by: Sam Lang <sam.lang@inktank.com>
5251
5252 commit 3eb19c810725b011baacdb8a6f5b172f4720a39a
5253 Author: Sam Lang <sam.lang@inktank.com>
5254 Date:   Wed Jan 23 14:37:39 2013 -0600
5255
5256     Replace /tmp/cephtest/ with configurable path
5257
5258     Teuthology uses /tmp/cephtest/ as the scratch test directory for
5259     a run.  This patch replaces /tmp/cephtest/ everywhere with a
5260     per-run directory: {basedir}/{rundir} where {basedir} is a directory
5261     configured in .teuthology.yaml (/tmp/cephtest if not specified),
5262     and {rundir} is the name of the run, as given in --name.  If no name
5263     is specified, {user}-{timestamp} is used.
5264
5265     To get the old behavior (/tmp/cephtest), set test_path: /tmp/cephtest
5266     in .teuthology.yaml.
5267
5268     This change was modivated by #3782, which requires a test dir that
5269     survives across reboots, but also resolves #3767.
5270
5271     Signed-off-by: Sam Lang <sam.lang@inktank.com>
5272     Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
5273
5274 commit 2f829870e140c87b30e5b7aa3ad237a90dcb2179
5275 Author: Yehuda Sadeh <yehuda@inktank.com>
5276 Date:   Fri Dec 21 10:20:02 2012 -0800
5277
5278     task/swift: change upstream repository url
5279
5280     Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
5281
5282 commit 334d6386753510c312898552c6f92313942786ef
5283 Merge: b8e6ce4db9 26df886d82
5284 Author: Joao Eduardo Luis <jecluis@gmail.com>
5285 Date:   Thu Nov 29 00:53:59 2012 +0000
5286
5287     Merge branch 'wip-mon-thrasher'
5288
5289 commit b8e6ce4db9a603ce3523b1759c65eeadee55daa7
5290 Author: Sage Weil <sage@inktank.com>
5291 Date:   Thu Nov 22 13:59:58 2012 -0800
5292
5293     s3tests: fix typo
5294
5295 commit 26df886d825e28c25b630887b8dcc1c8c6d687d8
5296 Author: Yehuda Sadeh <yehuda@inktank.com>
5297 Date:   Mon Nov 19 16:19:06 2012 -0800
5298
5299     rgw-logsocket: a task to verify opslog socket works
5300
5301     Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
5302
5303 commit 617534e76978acb09a9f925f18bba475a65a7dd2
5304 Author: Sage Weil <sage@inktank.com>
5305 Date:   Mon Sep 10 11:08:57 2012 -0700
5306
5307     s3tests: run against arbitrary branch/sha1 of s3-tests.git
5308
5309 commit 7d5c7ee8c6f2bfedd193a8d3b7102b4cfe0bf74f
5310 Author: Sage Weil <sage@newdream.net>
5311 Date:   Wed Jun 6 16:00:55 2012 -0700
5312
5313     pull s3-tests.git using git, not http
5314
5315 commit ce951cf4caffd5d6883cc8dcd24372bcdf03690a
5316 Author: Sage Weil <sage@newdream.net>
5317 Date:   Sat May 5 09:30:41 2012 -0700
5318
5319     ceph.newdream.net -> ceph.com
5320
5321 commit 2b879905fcfd660e242ed1a804d1c8301d17ab84
5322 Merge: 1ac4bb10fc 1970713a2f
5323 Author: Mark Nelson <nhm@clusterfaq.org>
5324 Date:   Wed Mar 14 15:32:23 2012 -0500
5325
5326     Merge branch 'master' of github.com:ceph/teuthology
5327
5328 commit 1970713a2fc43e3afae376712356ca93a65d9e1f
5329 Author: Sage Weil <sage@newdream.net>
5330 Date:   Fri Mar 2 10:55:19 2012 -0800
5331
5332     github.com/NewDreamNetwork -> github.com/ceph
5333
5334 commit 1ac4bb10fc4b7d8d07c44b0e92b1627c721ab925
5335 Author: Josh Durgin <josh.durgin@dreamhost.com>
5336 Date:   Tue Feb 21 14:54:33 2012 -0800
5337
5338     Add necessary imports for s3 tasks, and keep them alphabetical.
5339
5340 commit 92110e5a4460281139233dcea3f629d01182d398
5341 Author: Yehuda Sadeh <yehuda.sadeh@dreamhost.com>
5342 Date:   Tue Feb 21 12:12:03 2012 -0800
5343
5344     rgw: access key uses url safe chars
5345
5346     Signed-off-by: Yehuda Sadeh <yehuda.sadeh@dreamhost.com>
5347
5348 commit 709d9441127fec93da74c7702cafa54a47e10e8f
5349 Author: Sage Weil <sage@newdream.net>
5350 Date:   Sun Jan 15 22:48:33 2012 -0800
5351
5352     use local mirrors for (most) github urls
5353
5354     A cronjob on ceph.newdream.net updates these every 15 minutes.  Sigh.
5355
5356 commit 9598e47949ba65030c722947dc433e38875b1bd6
5357 Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
5358 Date:   Mon Dec 5 10:07:25 2011 -0800
5359
5360     Rename "testrados" and "testswift" tasks to not begin with "test".
5361
5362     Anything "test*" looks like a unit test, and shouldn't be used for
5363     actual code.
5364
5365 commit 6236e7db22edac7b51fc6329188b6afa74f3fc78
5366 Author: Yehuda Sadeh <yehuda.sadeh@dreamhost.com>
5367 Date:   Thu Nov 17 16:53:21 2011 -0800
5368
5369     testswift: fix config
5370
5371 commit 1dd607cabb07126769b4beb1ba6677e21c448719
5372 Author: Yehuda Sadeh <yehuda.sadeh@dreamhost.com>
5373 Date:   Wed Nov 16 16:00:01 2011 -0800
5374
5375     rgw: add swift task
5376
5377     still not completely working (for some reason it skips all the tests)
5378
5379 commit cb425c158085568cd92c239a071e282c74eddf1a
5380 Author: Greg Farnum <gregory.farnum@dreamhost.com>
5381 Date:   Fri Sep 30 09:26:42 2011 -0700
5382
5383     s3-tests: use radosgw-admin instead of radosgw_admin
5384
5385     Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
5386
5387 commit 37d7d515345ab04c333d6fada722e432e5816eb3
5388 Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
5389 Date:   Fri Sep 16 11:09:45 2011 -0700
5390
5391     s3tests: Clone repository from github.
5392
5393     Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
5394
5395 commit 29a242d97dd4a5a9110710027cdddb244b8b0e29
5396 Author: Tommi Virtanen <tv@eagain.net>
5397 Date:   Tue Sep 13 14:53:02 2011 -0700
5398
5399     Move orchestra to teuthology.orchestra so there's just one top-level package.
5400
5401 commit ec49a5f263f71aa473257e3fd49d86e475fe9456
5402 Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
5403 Date:   Fri Sep 9 13:22:03 2011 -0700
5404
5405     Callers of task s3tests.create_users don't need to provide dummy "fixtures" dict.
5406
5407 commit d7d995e82b45e6077040b467c8ef9a82a573faf7
5408 Author: Stephon Striplin <stephon.striplin@dreamhost.com>
5409 Date:   Tue Aug 9 13:43:46 2011 -0700
5410
5411     allow s3tests.create_users defaults be overridden
5412
5413 commit 0086109767d5bfbbc370ca13d3fe91895b207821
5414 Author: Josh Durgin <josh.durgin@dreamhost.com>
5415 Date:   Thu Jul 14 16:47:29 2011 -0700
5416
5417     Make targets a dictionary mapping hosts to ssh host keys.
5418
5419 commit 1b2c96416f554c0890b2690291b9a2dc8a6dc17a
5420 Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
5421 Date:   Wed Jul 6 14:17:24 2011 -0700
5422
5423     Skip s3-tests marked fails_on_rgw, they will fail anyway.
5424
5425 commit 06fb9b95e39985630d89e1635dcd12510686d9cd
5426 Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
5427 Date:   Tue Jul 5 09:27:28 2011 -0700
5428
5429     The shell exits after the command, hence there is no need for pushd/popd.
5430
5431 commit cd524a6904bf8254edc73a9148308f642638e33d
5432 Author: Josh Durgin <josh.durgin@dreamhost.com>
5433 Date:   Fri Jun 24 17:09:47 2011 -0700
5434
5435     Add s3tests task.