]> git.proxmox.com Git - ceph.git/blob - ceph/doc/changelog/v0.94.8.txt
add subtree-ish sources for 12.0.3
[ceph.git] / ceph / doc / changelog / v0.94.8.txt
1 commit 838cd35201e4fe1339e16d987cc33e873524af90 (tag: refs/tags/v0.94.8, refs/remotes/gh/hammer)
2 Author: Jenkins Build Slave User <jenkins-build@trusty-huge--349f4788-0128-42d9-aca5-78c0aad008e6.localdomain>
3 Date: Fri Aug 26 00:02:49 2016 +0000
4
5 0.94.8
6
7 commit eb27865f268eaa4abc4e4b46354f1ea1556d1a5c
8 Merge: 08277b7 5248929
9 Author: Loic Dachary <loic@dachary.org>
10 Date: Wed Aug 17 15:26:49 2016 +0200
11
12 Merge pull request #10750 from ceph/wip-14799-hammer
13
14 hammer: rocksdb do not link against tcmalloc if it's disabled
15
16 Reviewed-by: Loic Dachary <ldachary@redhat.com>
17
18 commit 5248929c296d6130ee99fe94fbe69151d241efb8
19 Author: Kefu Chai <kchai@redhat.com>
20 Date: Tue Aug 16 23:45:18 2016 -0700
21
22 rocksdb: disable tcmalloc if disabled
23
24 the rocksdb/configure.ac does not support --with-tcmalloc before,
25 and the updated rocksdb commit has the change for '--with-tcmalloc'
26 option, so let's pick it up.
27
28 Fixes: http://tracker.ceph.com/issues/14799
29 Signed-off-by: Kefu Chai <kchai@redhat.com>
30
31 commit fdfcd9b66cda1c637fbaad1c15d2ea04e426210d
32 Author: Sage Weil <sage@redhat.com>
33 Date: Sat Sep 12 14:12:20 2015 -0400
34
35 ceph.spec: respect CEPH_EXTRA_CONFIGURE_ARGS
36
37 Signed-off-by: Sage Weil <sage@redhat.com>
38 (cherry picked from commit 51abff11688f0201b8f4076ac515e4515929d4cb)
39 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
40
41 commit 08277b7bc7c0e533c3fd56a0040dc0ddc74637d6
42 Merge: bcab369 d5274a3
43 Author: Orit Wasserman <owasserm@redhat.com>
44 Date: Sat Aug 13 19:19:37 2016 +0200
45
46 Merge pull request #10718 from oritwas/wip-14328-hammer
47
48 hammer: rgw: the swift key remains after removing a subuser
49
50 commit bcab36992580e5666afbf8e7bada4e850dfc84c6
51 Merge: da2960f 699b7c8
52 Author: Orit Wasserman <owasserm@redhat.com>
53 Date: Sat Aug 13 18:22:19 2016 +0200
54
55 Merge pull request #10509 from yehudasa/wip-15557-hammer
56
57 rgw: Have a flavor of bucket deletion to bypass GC and to trigger
58
59 commit d5274a3aeb7cc40f5ac04ac1ffcd6e3fc3019be3
60 Author: Orit Wasserman <owasserm@redhat.com>
61 Date: Wed Jan 20 17:24:38 2016 +0100
62
63 rgw: fix subuser rm command failure
64
65 Fixes: 14375
66 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
67 (cherry picked from commit bbc06e70b49b2917ee63985727cb92724fb99209)
68
69 commit f9637743efc7ba3164e3bb9ee7a768801e1599da
70 Author: Sangdi Xu <xu.sangdi@h3c.com>
71 Date: Wed Dec 16 09:09:16 2015 +0800
72
73 rgw: add a method to purge all associate keys when removing a subuser
74
75 Fixes: #12890
76
77 When removing a subuser, make sure all of its keys, including the swift key and possible s3 keys, are also deleted.
78
79 Signed-off-by: Sangdi Xu <xu.sangdi@h3c.com>
80 (cherry picked from commit e7b7e1afc7a81c3f97976f7442fbdc5118b532b5)
81
82 commit da2960f974b9a64a683ac8b77faf08fda3a5d0ba
83 Merge: 0f67539 3a30ffc
84 Author: Nathan Cutler <presnypreklad@gmail.com>
85 Date: Fri Aug 12 12:23:41 2016 +0200
86
87 Merge pull request #10291 from badone/wip-13829-hammer
88
89 hammer: config set with negative value results in "error setting 'filestore_merge_threshold' to '-40': (22) Invalid argument"
90
91 Reviewed-by: Samuel Just <sjust@redhat.com>
92 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
93 Reviewed-by: Kefu Chai <kchai@redhat.com>
94 Reviewed-by: Nathan Cutler <ncutler@suse.com>
95
96 commit 0f67539cf5efc46a92d7ea3ac031e0ce1c573a7a
97 Merge: 1e01986 0d4b601
98 Author: Loic Dachary <loic@dachary.org>
99 Date: Fri Aug 12 09:46:30 2016 +0200
100
101 Merge pull request #10699 from ktdreyer/wip-15867-nroff
102
103 jewel: doc: fix by-parttypeuuid in ceph-disk(8) nroff
104
105 Reviewed-by: Loic Dachary <ldachary@redhat.com>
106
107 commit 0d4b6013f0f130c93e2e488d1cd83527ad91274f
108 Author: Ken Dreyer <kdreyer@redhat.com>
109 Date: Thu Aug 11 17:11:41 2016 -0600
110
111 doc: fix by-parttypeuuid in ceph-disk(8) nroff
112
113 Commit 221efb0b893adbfd7a19df171cf967fee87afcc7 altered the rST source
114 for the ceph-disk man page. In Hammer, we also have to modify the nroff
115 sources, because static copies of the generated man pages are stored in
116 Git.
117
118 Fixes: http://tracker.ceph.com/issues/15867
119 Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
120
121 commit 1e01986e9550700a11adfc1b175eca042c9cb188
122 Merge: de28029 187d308
123 Author: Nathan Cutler <presnypreklad@gmail.com>
124 Date: Thu Aug 11 22:58:33 2016 +0200
125
126 Merge pull request #9741 from SUSE/wip-16343-hammer
127
128 hammer: boost uuid makes valgrind complain
129
130 Reviewed-by: Samuel Just <sjust@redhat.com>
131 Reviewed-by: Sage Weil <sage@redhat.com>
132 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
133 Reviewed-by: Nathan Cutler <ncutler@suse.com>
134
135 commit de28029c25e4ff527c6021488381022538331473
136 Merge: cb0488c 64211fa
137 Author: Orit Wasserman <owasserm@redhat.com>
138 Date: Thu Aug 11 15:58:51 2016 +0200
139
140 Merge pull request #10554 from Vicente-Cheng/wip-16815-hammer
141
142 hammer: radosgw-admin region-map set is not reporting the bucket quota correctly
143 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
144
145 commit cb0488c181a122de13d63686b2a64e1aa00791ca
146 Merge: 0bc2d66 a3003f6
147 Author: Orit Wasserman <owasserm@redhat.com>
148 Date: Thu Aug 11 15:56:39 2016 +0200
149
150 Merge pull request #10555 from Vicente-Cheng/wip-16317-hammer
151
152 hammer: rgw: bucket listing following object delete is partial
153 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
154
155 commit 0bc2d66db5feba6cd887d7c297839b231012b880
156 Merge: 5866fef 81aef60
157 Author: Orit Wasserman <owasserm@redhat.com>
158 Date: Thu Aug 11 15:55:01 2016 +0200
159
160 Merge pull request #10530 from yehudasa/wip-16412-hammer
161
162 rgw: remove bucket index objects when deleting the bucket
163 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
164
165 commit 5866fefb32cc443549891754a97eb8a6b8dd2af0
166 Merge: da4f735 300c111
167 Author: Orit Wasserman <owasserm@redhat.com>
168 Date: Thu Aug 11 11:14:53 2016 +0200
169
170 Merge pull request #8755 from kmroz/wip-hammer-convert-plain-object-to-versioned
171
172 hammer: rgw: convert plain object to versioned (with null version) when removing
173 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
174
175 commit da4f735d7937aa623dfef41f921ad9381b1a07d7
176 Merge: 4a12a88 d3eae0a
177 Author: Nathan Cutler <presnypreklad@gmail.com>
178 Date: Thu Aug 4 21:41:15 2016 +0200
179
180 Merge pull request #9271 from XinzeChi/wip-fix-digest-hammer
181
182 osd: fix omap digest compare when scrub
183
184 Reviewed-by: Samuel Just <sjust@redhat.com>
185 Reviewed-by: David Zafman <dzafman@redhat.com>
186 Reviewed-by: Nathan Cutler <ncutler@suse.com>
187
188 commit 4a12a88d7098c1f67365a82f3cf9a174dd5ada32
189 Merge: 6f107f1 ec02d8b
190 Author: Nathan Cutler <presnypreklad@gmail.com>
191 Date: Thu Aug 4 21:38:57 2016 +0200
192
193 Merge pull request #8805 from SUSE/wip-15482-hammer
194
195 hammer: is_split crash in handle_pg_create
196
197 Reviewed-by: Kefu Chai <kchai@redhat.com>
198
199 commit 6f107f1a7a42c070ef2da7f2b0b07d8bd2826429
200 Merge: 3b9e73f 3cb72dd
201 Author: Nathan Cutler <presnypreklad@gmail.com>
202 Date: Thu Aug 4 21:37:24 2016 +0200
203
204 Merge pull request #8593 from renhwsky/wip-15478-hammer
205
206 hammer: Pool quota alarm is not in effect
207
208 Reviewed-by: Kefu Chai <kchai@redhat.com>
209 Reviewed-by: Nathan Cutler <ncutler@suse.com>
210
211 commit 3b9e73f319b28c0403871c7d842304aca74f4c6a
212 Merge: d34ac2a b9a4ad9
213 Author: John Spray <jspray@redhat.com>
214 Date: Wed Aug 3 13:31:41 2016 +0100
215
216 Merge pull request #7542 from SUSE/wip-14668-hammer
217
218 hammer: Wrong ceph get mdsmap assertion
219
220 Reviewed-by: John Spray <john.spray@redhat.com>
221
222 commit d34ac2a4d76eda4a371cc16dbc6bfdc0223c81d8
223 Merge: 8f51f14 3ecdedd
224 Author: John Spray <jspray@redhat.com>
225 Date: Wed Aug 3 13:31:30 2016 +0100
226
227 Merge pull request #8042 from ukernel/hammer-15050
228
229 hammer: mds: fix stray purging in 'stripe_count > 1' case
230
231 Reviewed-by: John Spray <john.spray@redhat.com>
232
233 commit 8f51f146c98e281f065debdfdc4399624e0cce97
234 Merge: 4773e9c 049bc8a
235 Author: John Spray <jspray@redhat.com>
236 Date: Wed Aug 3 13:31:22 2016 +0100
237
238 Merge pull request #8804 from SUSE/wip-15512-hammer
239
240 hammer: Double decreased the count to trim caps which will cause failing to respond to cache pressure
241
242 Reviewed-by: John Spray <john.spray@redhat.com>
243
244 commit 4773e9c40849c2796aa16b7cacddbfe99536919c
245 Merge: 85a25e4 f6076dd
246 Author: John Spray <jspray@redhat.com>
247 Date: Wed Aug 3 13:31:09 2016 +0100
248
249 Merge pull request #9404 from SUSE/wip-16082-hammer
250
251 hammer: mds: wrongly treat symlink inode as normal file/dir when symlink inode is stale on kcephfs
252
253 Reviewed-by: John Spray <john.spray@redhat.com>
254
255 commit 85a25e43d4451b999ad517f089593d45895a7c59
256 Merge: 3c4d14f 2633ec3
257 Author: John Spray <jspray@redhat.com>
258 Date: Wed Aug 3 13:30:58 2016 +0100
259
260 Merge pull request #10198 from SUSE/wip-16626-hammer
261
262 hammer: Failing file operations on kernel based cephfs mount point leaves unaccessible file behind on hammer 0.94.7
263
264 Reviewed-by: John Spray <john.spray@redhat.com>
265
266 commit a3003f6f4f02e37cf96b6e243203d86ad2e39549
267 Author: Orit Wasserman <owasserm@redhat.com>
268 Date: Thu Feb 25 16:02:15 2016 +0100
269
270 rgw: reset return code in when iterating over the bucket the objects
271
272 Fixes: #14826
273
274 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
275 (cherry picked from commit e23fdcbbce47aaf041b9f8a0812c8a729aa5014b)
276
277 commit 64211fa78f872648aad2b2cec2b4b3cb8f2b9953
278 Author: Yehuda Sadeh <yehuda@redhat.com>
279 Date: Tue Oct 13 17:04:57 2015 -0700
280
281 rgw: fix compilation
282
283 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
284 (cherry picked from commit 9dfbae6d65bf6f8ab7cdf8597d556505b81cedd9)
285
286 commit 3e45c6bb431f7763a12a48daac30fbe7aa3047cd
287 Author: Orit Wasserman <owasserm@redhat.com>
288 Date: Mon Oct 12 11:14:19 2015 +0200
289
290 rgw: add bucket_quota to RGWRegionMap::encode_json
291
292 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
293 (cherry picked from commit e18b72f7e97a65af681e368d3706a465603df89d)
294
295 commit 699b7c8d5f51d895f5ff0578de5bf610436ee33b
296 Author: root <root@ceph-node1.homeoffice.wal-mart.com>
297 Date: Wed Apr 20 14:15:55 2016 +0530
298
299 rgw: Have a flavor of bucket deletion to bypass GC and to trigger
300 object deletions async.
301
302 Fixes: http://tracker.ceph.com/issues/15557
303 Signed-off-by: Pavan Rallabhandi <PRallabhandi@walmartlabs.com>
304 (cherry picked from commit b7a69fca248afeef1de1278890076693b16cf6d3)
305
306 Conflicts:
307 src/rgw/rgw_admin.cc
308 src/rgw/rgw_bucket.cc
309 src/rgw/rgw_bucket.h
310 src/test/cli/radosgw-admin/help.t
311
312 commit 81aef600671f0fe523206cea16000e79bc6bb75c
313 Author: Orit Wasserman <owasserm@redhat.com>
314 Date: Mon Jul 4 15:01:51 2016 +0200
315
316 rgw: remove bucket index objects when deleting the bucket
317
318 Fixes: http://tracker.ceph.com/issues/16412
319 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
320 (cherry picked from commit 3ae276390641ad5fc4fef0c03971db95948880b4)
321
322 commit 3c4d14fe73a97c52345369e732e445c78fb656c8
323 Merge: 79753ed 8a39e5e
324 Author: Nathan Cutler <presnypreklad@gmail.com>
325 Date: Fri Jul 29 23:51:47 2016 +0200
326
327 Merge pull request #10182 from gaurav36/wip-configure-xfs-issue-hammer
328
329 hammer: configure: Add -D_LARGEFILE64_SOURCE to Linux build.
330
331 Reviewed-by: Samuel Just <sjust@redhat.com>
332 Reviewed-by: Casey Bodley <cbodley@redhat.com>
333 Reviewed-by: Nathan Cutler <ncutler@suse.com>
334
335 commit 79753edf9c4763ae15d82502b53c6f3c9cd24349
336 Merge: 75271a2 7dbace5
337 Author: Nathan Cutler <presnypreklad@gmail.com>
338 Date: Fri Jul 29 09:40:46 2016 +0200
339
340 Merge pull request #7961 from dzafman/wip-15002
341
342 hammer: objects unfound after repair (fixed by repeering the pg)
343
344 Reviewed-by: Samuel Just <sjust@redhat.com>
345 Reviewed-by: Nathan Cutler <ncutler@suse.com>
346
347 commit 75271a2529b1504d249b528fe3fafd4c5e90a9bc
348 Merge: 76fa3b1 43d1b92
349 Author: Nathan Cutler <presnypreklad@gmail.com>
350 Date: Fri Jul 29 09:38:15 2016 +0200
351
352 Merge pull request #9125 from dzafman/wip-15635
353
354 hammer: [MON] Fixed calculation of %USED
355
356 Reviewed-by: Samuel Just <sjust@redhat.com>
357 Reviewed-by: Nathan Cutler <ncutler@suse.com>
358
359 commit 76fa3b1097b3711ef75bd36aa2a4074e7a8fa9e6
360 Merge: cf5f111 ca0beef
361 Author: Nathan Cutler <presnypreklad@gmail.com>
362 Date: Fri Jul 29 09:36:05 2016 +0200
363
364 Merge pull request #8464 from dzafman/wip-13039-15315
365
366 hammer: "FAILED assert(pg->info.history.same_interval_since == p.same_interval_since)" in upgrade:hammer-x-jewel-distro-basic-openstack
367
368 Reviewed-by: Samuel Just <sjust@redhat.com>
369 Reviewed-by: Nathan Cutler <ncutler@suse.com>
370
371 commit cf5f1117afe9e03c683fd40e5af964eaf5975e2b
372 Merge: 76ce775 488a787
373 Author: Nathan Cutler <presnypreklad@gmail.com>
374 Date: Fri Jul 29 09:32:47 2016 +0200
375
376 Merge pull request #8573 from Vicente-Cheng/wip-15293-hammer
377
378 hammer: ECBackend: send write and read sub ops on behalf of client ops at normal priority
379
380 Reviewed-by: Samuel Just <sjust@redhat.com>
381 Reviewed-by: Nathan Cutler <ncutler@suse.com>
382
383 commit 76ce775761f755bd3085ba22cf088c3a46b78d3a
384 Merge: 8629d37 e20df80
385 Author: Nathan Cutler <presnypreklad@gmail.com>
386 Date: Fri Jul 29 09:31:16 2016 +0200
387
388 Merge pull request #8665 from Vicente-Cheng/wip-15525-hammer
389
390 hammer: ReplicatedPG.cc: 430: FAILED assert(!actingbackfill.empty()): old watch timeout tries to queue repop on replica
391
392 Reviewed-by: Samuel Just <sjust@redhat.com>
393 Reviewed-by: Nathan Cutler <ncutler@suse.com>
394
395 commit 8629d378b05d4f0374436373a0c972360bae7273
396 Merge: 79fc401 7b3f1da
397 Author: Nathan Cutler <presnypreklad@gmail.com>
398 Date: Fri Jul 29 09:29:58 2016 +0200
399
400 Merge pull request #8806 from SUSE/wip-15320-hammer
401
402 hammer: LibRadosMiscConnectFailure.ConnectFailure (not so intermittent) failure in upgrade/hammer-x
403
404 Reviewed-by: Samuel Just <sjust@redhat.com>
405 Reviewed-by: Nathan Cutler <ncutler@suse.com>
406
407 commit 79fc401844c935283f43fbea83e113e298a830ba
408 Merge: f4d56b0 cbd5aaf
409 Author: Nathan Cutler <presnypreklad@gmail.com>
410 Date: Fri Jul 29 09:28:32 2016 +0200
411
412 Merge pull request #9001 from Vicente-Cheng/wip-15730-hammer
413
414 hammer: osd: acting_primary not updated on split
415
416 Reviewed-by: Samuel Just <sjust@redhat.com>
417 Reviewed-by: Nathan Cutler <ncutler@suse.com>
418
419 commit f4d56b03b1decf83bebbf6996909ee89783068f0
420 Merge: f71c9e6 410ff15
421 Author: Nathan Cutler <presnypreklad@gmail.com>
422 Date: Fri Jul 29 09:26:55 2016 +0200
423
424 Merge pull request #9400 from Vicente-Cheng/wip-15796-hammer
425
426 hammer: LibRadosWatchNotifyPPTests/LibRadosWatchNotifyPP.WatchNotify2Timeout/1 segv
427
428 Reviewed-by: Samuel Just <sjust@redhat.com>
429 Reviewed-by: Nathan Cutler <ncutler@suse.com>
430
431 commit f71c9e658d6f5af3cfa48a0dfb9a063f4965e7ce
432 Merge: fdb730e 23498a9
433 Author: Nathan Cutler <presnypreklad@gmail.com>
434 Date: Fri Jul 29 09:24:42 2016 +0200
435
436 Merge pull request #9416 from chardan/wip-15770-hammer
437
438 hammer: mon/OSDMonistor: improve reweight_by_utilization() logic
439
440 Reviewed-by: Samuel Just <sjust@redhat.com>
441 Reviewed-by: Nathan Cutler <ncutler@suse.com>
442
443 commit fdb730e169d8885ccd72e140c31797179c0a2c75
444 Merge: f9e8dc8 2c18015
445 Author: Nathan Cutler <presnypreklad@gmail.com>
446 Date: Fri Jul 29 09:21:25 2016 +0200
447
448 Merge pull request #10189 from dreamhost/wip-16618-hammer
449
450 rgw: fix multi-delete query param parsing.
451
452 Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
453 Reviewed-by: Nathan Cutler <ncutler@suse.com>
454
455 commit f9e8dc824a5aa89e24ef12fc8fcfdbdd4f926aef
456 Merge: f3cbf5f 24cc4f9
457 Author: Nathan Cutler <presnypreklad@gmail.com>
458 Date: Fri Jul 29 09:17:57 2016 +0200
459
460 Merge pull request #8379 from Vicente-Cheng/wip-15283-hammer
461
462 hammer: RGW shouldn't send Content-Type nor Content-Length for 304 responses
463
464 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
465 Reviewed-by: Nathan Cutler <ncutler@suse.com>
466
467 commit f3cbf5f891029975f4f5a915366fce28fcabd196
468 Merge: ba8de3a f28477c
469 Author: Nathan Cutler <presnypreklad@gmail.com>
470 Date: Fri Jul 29 09:14:39 2016 +0200
471
472 Merge pull request #8313 from SUSE/wip-15257-hammer
473
474 hammer: rgw: radosgw server abort when user passed bad parameters to set quota
475
476 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
477
478 commit ba8de3ab31567c8179b72fb1dc8dbf7ccd1f5f38
479 Merge: 9562198 ed4ca7c
480 Author: Nathan Cutler <presnypreklad@gmail.com>
481 Date: Fri Jul 29 09:13:09 2016 +0200
482
483 Merge pull request #8528 from wido/hammer-issue-15348
484
485 rgw: Set Access-Control-Allow-Origin to a Asterisk if allowed in a rule
486
487 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
488 Reviewed-by: Nathan Cutler <ncutler@suse.com>
489
490 commit 95621986d893029bc0a4a686c7da5cd6e0fa6f2b
491 Merge: 28107d3 1ac920b
492 Author: Nathan Cutler <presnypreklad@gmail.com>
493 Date: Fri Jul 29 09:10:57 2016 +0200
494
495 Merge pull request #9671 from ceph/hammer-rgw-copy-if-newer
496
497 hammer: rgw: backport rgwx-copy-if-newer for radosgw-agent
498
499 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
500 Reviewed-by: Nathan Cutler <ncutler@suse.com>
501
502 commit 28107d3ebfded3cd9046e5a23ab83d4e0252cce2
503 Merge: 5e1a57a 850881c
504 Author: Nathan Cutler <presnypreklad@gmail.com>
505 Date: Fri Jul 29 09:09:14 2016 +0200
506
507 Merge pull request #9961 from SUSE/wip-16190-hammer
508
509 hammer: subuser rm fails with status 125
510
511 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
512
513 commit 5e1a57a6f9c2afbdfc6d3e73d438551202c68229
514 Merge: 91889db ce313cd
515 Author: Nathan Cutler <presnypreklad@gmail.com>
516 Date: Fri Jul 29 09:07:10 2016 +0200
517
518 Merge pull request #9402 from SUSE/wip-15966-hammer
519
520 hammer: No Last-Modified, Content-Size and X-Object-Manifest headers if no segments in DLO manifest
521
522 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
523 Reviewed-by: Nathan Cutler <ncutler@suse.com>
524
525 commit 23498a9620f792cd099dba028c5bdf96b1a625be
526 Author: xie xingguo <xie.xingguo@zte.com.cn>
527 Date: Tue May 31 15:40:05 2016 -0700
528
529 mon/OSDMonitor: avoid potential expensive grace calculation
530
531 The grace calculation during check_failure() is now very complicated
532 and time-consuming. Therefore we shall skip this when it is possible.
533
534 Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
535 (cherry picked from commit 3557903d5d57642179b2ae137bedc389974b1956)
536
537 Conflicts:
538 src/mon/OSDMonitor.cc
539
540 Resolved by choosing the move-to-top implementation. Removed unused vars.
541
542 commit 1b6f6f27b77803727a523b4337cbad411e8321ed
543 Author: xie xingguo <xie.xingguo@zte.com.cn>
544 Date: Tue Apr 26 11:13:32 2016 +0800
545
546 mon/OSDMonitor: improve reweight_by_utilization() logic
547
548 By calling reweight_by_utilization() method, we are aiming at an evener result
549 of utilization among all osds. To achieve this, we shall decrease weights of
550 osds which are currently overloaded, and try to increase weights of osds which
551 are currently underloaded when it is possible.
552 However, we can't do this all at a time in order to avoid a massive pg migrations
553 between osds. Thus we introduce a max_osds limit to smooth the progress.
554
555 The problem here is that we have sorted the utilization of all osds in a descending
556 manner and we always try to decrease the weights of the most overloaded osds
557 since they are most likely to encounter a nearfull/full transition soon, but
558 we won't increase the weights from the most underloaded(least utilized by contrast)
559 at the same time, which I think is not quite reasonable.
560
561 Actually, the best thing would probably be to iterate over teh low and high osds
562 in parallel, and do the ones that are furthest from the average first.
563
564 Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
565 (cherry picked from commit e7a32534ebc9e27f955ff2d7a8d1db511383301e)
566
567 Conflicts:
568 src/mon/OSDMonitor.cc
569
570 Resolved by picking the lambda implemenation.
571 NOTE: Because hammer does not support C++11, the lambda functionality from the
572 current master has been moved into the "Sorter" function object.
573
574 commit 91889db73ab097502615b208c6e9f3b40411e682
575 Merge: f3c4f7b b78a1be
576 Author: Nathan Cutler <presnypreklad@gmail.com>
577 Date: Wed Jul 27 08:28:38 2016 +0200
578
579 Merge pull request #10038 from tchaikov/wip-16297-hammer
580
581 mon: Monitor: validate prefix on handle_command()
582
583 Reviewed-by: Samuel Just <sjust@redhat.com>
584 Reviewed-by: Nathan Cutler <ncutler@suse.com>
585
586 commit f3c4f7b61ab519c532a2218a11c5472daaaf5fa9
587 Merge: 2fe8edb b31ac2d
588 Author: Nathan Cutler <presnypreklad@gmail.com>
589 Date: Tue Jul 26 19:08:23 2016 +0200
590
591 Merge pull request #9090 from tchaikov/wip-13990-hammer
592
593 hammer: osd: remove all stale osdmaps in handle_osd_map()
594
595 Reviewed-by: Samuel Just <sjust@redhat.com>
596 Reviewed-by: Nathan Cutler <ncutler@suse.com>
597
598 commit 2fe8edb87e7384f206414db8a49e909e51f48de3
599 Merge: 67b7f11 0498969
600 Author: Nathan Cutler <presnypreklad@gmail.com>
601 Date: Mon Jul 25 21:44:14 2016 +0200
602
603 Merge pull request #10227 from SUSE/wip-15081-hammer
604
605 hammer: When logging to a file fails, ceph logs excessively to stderr
606
607 Reviewed-by: Samuel Just <sjust@redhat.com>
608
609 commit 67b7f11c63313a5dcdf34b904137ef7a4673f4ec
610 Merge: bd59c96 474abb8
611 Author: Nathan Cutler <presnypreklad@gmail.com>
612 Date: Mon Jul 25 21:43:49 2016 +0200
613
614 Merge pull request #9893 from Vicente-Cheng/wip-16430-hammer
615
616 hammer: OSDMonitor: drop pg temps from not the current primary
617
618 Reviewed-by: Samuel Just <sjust@redhat.com>
619 Reviewed-by: Nathan Cutler <ncutler@suse.com>
620
621 commit bd59c9695fadedef89b77166f60f5a8cf85825ff
622 Merge: 3794e3a 055427c
623 Author: Nathan Cutler <presnypreklad@gmail.com>
624 Date: Mon Jul 25 21:42:54 2016 +0200
625
626 Merge pull request #9238 from SUSE/wip-15962-hammer
627
628 hammer: ceph-osd valgrind invalid reads/writes
629
630 Reviewed-by: Samuel Just <sjust@redhat.com>
631
632 commit 3794e3ad44d1e429f110abe71fb2d2319567bcd1
633 Merge: 89233e3 d96086a
634 Author: Nathan Cutler <presnypreklad@gmail.com>
635 Date: Mon Jul 25 21:42:06 2016 +0200
636
637 Merge pull request #8849 from SUSE/wip-14326-hammer
638
639 hammer: osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to == head)
640
641 Reviewed-by: Samuel Just <sjust@redhat.com>
642
643 commit 89233e32ffcdd7f795873771859caba1ce3ad23f
644 Merge: 253143b fb1b40f
645 Author: Nathan Cutler <presnypreklad@gmail.com>
646 Date: Mon Jul 25 21:40:57 2016 +0200
647
648 Merge pull request #8845 from SUSE/wip-15647-hammer
649
650 hammer: osd: rados cppool omap to ec pool crashes osd
651
652 Reviewed-by: Samuel Just <sjust@redhat.com>
653 Reviewed-by: Nathan Cutler <ncutler@suse.com>
654
655 commit 253143b559001eadb73ed9945abe2d0b8d946b65
656 Merge: 81133dd 365f21b
657 Author: Loic Dachary <loic@dachary.org>
658 Date: Mon Jul 25 15:05:54 2016 +0200
659
660 Merge pull request #8470 from dachary/wip-15403-api-test-timeout-hammer
661
662 hammer: tests: be more generous with test timeout
663
664 Reviewed-by: Loic Dachary <ldachary@redhat.com>
665
666 commit 81133dd6e2724ee78cc69bda8e89ca35b20b6b23
667 Merge: d71845c 3bb248b
668 Author: Nathan Cutler <presnypreklad@gmail.com>
669 Date: Mon Jul 25 15:05:31 2016 +0200
670
671 Merge pull request #9977 from shun-s/hammer
672
673 replcatedBackend: delete one of the repeated op->mark_started in ReplicatedBackend::sub_op_modify_impl
674
675 Reviewed-by: Kefu Chai <kchai@redhat.com>
676 Reviewed-by: Nathan Cutler <ncutler@suse.com>
677
678 commit d71845cec169c660c180a4c0313e0cdbc25329dc
679 Merge: 387d5c1 5057c34
680 Author: Nathan Cutler <presnypreklad@gmail.com>
681 Date: Mon Jul 25 12:51:45 2016 +0200
682
683 Merge pull request #8960 from badone/wip-hammer-rados-bench-units
684
685 hammer: Add units to rados bench output and clean up formatting
686
687 Reviewed-by: Nathan Cutler <ncutler@suse.com>
688
689 commit 474abb8275cd7507d7c7eae0d0e5647612fc177e
690 Author: Samuel Just <sjust@redhat.com>
691 Date: Thu Jun 2 10:43:17 2016 -0700
692
693 OSDMonitor: drop pg temp from sources other than the current primary
694
695 Fixes: http://tracker.ceph.com/issues/16127
696 Signed-off-by: Samuel Just <sjust@redhat.com>
697 (cherry picked from commit 1a07123c38e3fecb3fb2e43bbbae962d8411d287)
698
699 change:
700 use NULL to replace the nullptr because we don't have C++11
701
702 commit b31ac2d37ecb68ed9661923fbaeaedeac241ba4a
703 Author: Kefu Chai <kchai@redhat.com>
704 Date: Thu May 12 20:28:11 2016 +0800
705
706 osd: reset session->osdmap if session is not waiting for a map anymore
707
708 we should release the osdmap reference once we are done with it,
709 otherwise we might need to wait very long to update that reference with
710 a newer osdmap ref. this appears to be an OSDMap leak: it is held by an
711 quiet OSD::Session forever.
712
713 the osdmap is not reset in OSD::session_notify_pg_create(), because its
714 only caller is wake_pg_waiters(), which will call
715 dispatch_session_waiting() later. and dispatch_session_waiting() will
716 check the session->osdmap, and will also reset the osdmap if
717 session->waiting_for_pg.empty().
718
719 Fixes: http://tracker.ceph.com/issues/13990
720 Signed-off-by: Kefu Chai <kchai@redhat.com>
721 (cherry picked from commit 82b0af7cedc3071cd83ee53479f834c23c62b7d0)
722
723 commit 3a30ffc21a5edd52726867302da6b358f93aa8dc
724 Author: Brad Hubbard <bhubbard@redhat.com>
725 Date: Wed Jan 27 11:34:47 2016 +1000
726
727 qa: Add test for #13829
728
729 qa/workunits/cephtool/test.sh: add test for setting negative int options
730 src/test/daemon_config.cc: remove tests for failed negative values
731
732 Fixes: #13829
733
734 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
735 (cherry picked from commit 994ac2942d04584d1617e6d4bbd5b880b1ea0448)
736
737 commit f70e4adf8682ed82b267e21e5e63039048893243
738 Author: Brad Hubbard <bhubbard@redhat.com>
739 Date: Wed Jan 27 11:18:16 2016 +1000
740
741 common: Allow config set with negative value
742
743 A recent commit disabled negative values but they are required for variables
744 such as filestore_merge_threshold.
745
746 Modified patch to remove C++11 specific elements so it will build for hammer
747
748 Fixes: #13829
749
750 Signed-off-by: Kefu Chai <kchai@redhat.com>
751 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
752 (cherry picked from commit 8b777a0c346bc70fd10d07e89368b3785b58f10e)
753
754 commit 04989695915fd7ee5ef85ae34ede1fd680514f9d
755 Author: Sage Weil <sage@redhat.com>
756 Date: Thu Mar 10 09:50:07 2016 -0500
757
758 log: do not repeat errors to stderr
759
760 If we get an error writing to the log, log it only once to stderr.
761 This avoids generating, say, 72 GB of ENOSPC errors in
762 teuthology.log when /var/log fills up.
763
764 Fixes: #14616
765 Signed-off-by: Sage Weil <sage@redhat.com>
766 (cherry picked from commit d9ac0474b864afda58a44b9012cca4bbc6aaf509)
767
768 Conflicts:
769 src/log/Log.cc (drop m_uid and m_gid which are not used in hammer;
770 order of do_stderr, do_syslog, do_fd conditional blocks is reversed in
771 hammer; drop irrelevant speed optimization code from
772 5bfe05aebfefdff9022f0eb990805758e0edb1dc)
773
774 commit 2633ec3934ad8f7629c55b4345f426cacfb7b140
775 Author: Yan, Zheng <zyan@redhat.com>
776 Date: Fri Apr 15 20:15:14 2016 +0800
777
778 mds: disallow 'open truncate' non-regular inode
779
780 Signed-off-by: Yan, Zheng <zyan@redhat.com>
781 (cherry picked from commit 0e4b6f2332bb4822cf324587a94144f1c98e4b97)
782
783 Conflicts:
784 src/mds/Server.cc (hammer has cur->inode.inline_version - in master
785 this has been changed to cur->inode.inline_data.version)
786
787 commit 3f0fb20ed08dfde614cff6c056d41b59af200c2c
788 Author: Yan, Zheng <zyan@redhat.com>
789 Date: Fri Apr 15 19:45:23 2016 +0800
790
791 mds: only open non-regular inode with mode FILE_MODE_PIN
792
793 ceph_atomic_open() in kernel client does lookup and open at the same
794 time. So it can open a symlink inode with mode CEPH_FILE_MODE_WR.
795 Open a symlink inode with mode CEPH_FILE_MODE_WR triggers assertion
796 in Locker::check_inode_max_size();
797
798 Signed-off-by: Yan, Zheng <zyan@redhat.com>
799 (cherry picked from commit 4d15eb12298e007744486e28924a6f0ae071bd06)
800
801 commit 2c180159c00aa184e1fe8b051b1aa1e9f0f332fc
802 Author: Robin H. Johnson <robin.johnson@dreamhost.com>
803 Date: Thu Jul 7 14:14:36 2016 -0700
804
805 rgw: fix multi-delete query param parsing.
806
807 Multi-delete is triggered by a query parameter on POST, but there are
808 multiple valid ways of representing it, and Ceph should accept ANY way
809 that has the query parameter set, regardless of what value or absence of
810 value.
811
812 This caused the RubyGem aws-sdk-v1 to break, and has been present since
813 multi-delete was first added in commit 0a1f4a97da, for the bobtail
814 release.
815
816 Fixes: http://tracker.ceph.com/issues/16618
817 Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
818 (cherry picked from commit a7016e1b67e82641f0702fda4eae799e953063e6)
819
820 commit 387d5c1ba836833a0cf11ddf9a4fb8690a532878
821 Merge: 2fd75cc 4eea92b
822 Author: Yehuda Sadeh <yehuda@redhat.com>
823 Date: Thu Jul 7 10:16:31 2016 -0700
824
825 Merge pull request #9407 from SUSE/wip-15833-hammer
826
827 hammer: RGW :: Subset of uploaded objects via radosgw are unretrievable when using erasure coded pool
828
829 commit 8a39e5e67bc9a93cadb6c13f25c7ccd57421af6b
830 Author: Ira Cooper <ira@redhat.com>
831 Date: Thu Mar 31 12:30:45 2016 -0400
832
833 configure: Add -D_LARGEFILE64_SOURCE to Linux build.
834
835 Fixes: http://tracker.ceph.com/issues/16612
836
837 Without -D_LARGEFILE64_SOURCE we can't build against libxfs,
838 because off64_t must be defined.
839
840 Signed-off-by: Ira Cooper <ira@redhat.com>
841 (cherry picked from commit 602425a)
842
843 commit 2fd75cced9dbcfdfdcba979eae90a6f4fb384758
844 Merge: dac65d0 2c97cb3
845 Author: Orit Wasserman <owasserm@redhat.com>
846 Date: Wed Jul 6 20:25:20 2016 +0200
847
848 Merge pull request #8475 from Vicente-Cheng/wip-15343-hammer
849
850 hammer: Multipart ListPartsResult has missing quotes on ETag
851 Reviewed-bu: Orit Wasserman <owasserm@redhat.com>
852
853 commit 3bb248b7d530403b347e3ff68a8996b624fc1c98
854 Author: shun-s <song.shun3@zte.com.cn>
855 Date: Tue Jun 28 15:30:16 2016 +0800
856
857 replcatedBackend: delete one useless op->mark_started as there are two in ReplicatedBackend::sub_op_modify_impl
858 delete one mark_start event as there are two same op->mark_started in ReplicatedBackend::sub_op_modify_impl
859 Fixes: http://tracker.ceph.com/issues/16572
860
861 Signed-off-by: shun-s <song.shun3@zte.com.cn>
862
863 commit ed4ca7c92cdf7404754dab6f9d58c64b647136c5
864 Author: Wido den Hollander <wido@42on.com>
865 Date: Tue Apr 5 11:14:16 2016 +0200
866
867 rgw: Set Access-Control-Allow-Origin to a Asterisk if allowed in a rule
868
869 Before this patch the RGW would respond with the Origin send by the client in the request
870 if a wildcard/asterisk was specified as a valid Origin.
871
872 This patch makes sure we respond with a header like this:
873
874 Access-Control-Allow-Origin: *
875
876 This way a resource can be used on different Origins by the same browser and that browser
877 will use the content as the asterisk.
878
879 We also keep in mind that when Authorization is send by the client different rules apply.
880 In the case of Authorization we may not respond with an Asterisk, but we do have to
881 add the Vary header with 'Origin' as a value to let the browser know that for different
882 Origins it has to perform a new request.
883
884 More information: https://developer.mozilla.org/en-US/docs/Web/HTTP/Access_control_CORS
885
886 Fixes: #15348
887
888 Signed-off-by: Wido den Hollander <wido@42on.com>
889 (cherry picked from commit 0021e224480c7164330eaa7cc1078bb8795169bf)
890
891 Conflicts:
892 src/rgw/rgw_rest.cc
893 hammer still uses s->cio->print() where master uses STREAM_IO(s)->print()
894
895 commit b78a1be835706e7dabc505be343945d0ac05697d
896 Author: Kefu Chai <kchai@redhat.com>
897 Date: Thu Jun 30 13:24:22 2016 +0800
898
899 mon: Monitor: validate prefix on handle_command()
900
901 Fixes: http://tracker.ceph.com/issues/16297
902
903 Signed-off-by: You Ji <youji@ebay.com>
904 (cherry picked from commit 7cb3434fed03a5497abfd00bcec7276b70df0654)
905
906 Conflicts:
907 src/mon/Monitor.cc (the signature of Monitor::reply_command()
908 changed a little bit in master, so adapt the
909 commit to work with the old method)
910
911 commit dac65d048919f701877de96d3271131853e532ed (refs/remotes/gh/wip-optracker-4354)
912 Merge: 2e156d7 a2e7ca1
913 Author: Nathan Cutler <presnypreklad@gmail.com>
914 Date: Mon Jun 27 17:51:19 2016 +0200
915
916 Merge pull request #8538 from Vicente-Cheng/wip-15360-hammer
917
918 hammer: Linking to -lrbd causes process startup times to balloon
919
920 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
921
922 commit 850881cf787ccb23eca989861a6688f3a4de1a7c
923 Author: Orit Wasserman <owasserm@redhat.com>
924 Date: Wed Jan 20 17:24:38 2016 +0100
925
926 rgw: fix subuser rm command failure
927
928 Fixes: 14375
929 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
930 (cherry picked from commit bbc06e70b49b2917ee63985727cb92724fb99209)
931
932 Conflicts:
933 src/rgw/rgw_user.cc The "if (op_state.will_purge_keys())" block was
934 later changed to "always purge all associated keys" by
935 e7b7e1afc7a81c3f97976f7442fbdc5118b532b5 - keep the hammer version
936
937 commit 055427cf58857eeadda11ed1b3381e84e384bc85
938 Author: Samuel Just <sjust@redhat.com>
939 Date: Thu May 12 16:57:49 2016 -0700
940
941 Pipe: take a ref to existing while we are waiting
942
943 Otherwise, if it is reaped while we are waiting, it'll be a
944 use-after-free.
945
946 Fixes: http://tracker.ceph.com/issues/15870
947 Signed-off-by: Samuel Just <sjust@redhat.com>
948 (cherry picked from commit b224912d249453d754fc0478d3680f8cfa1a5c22)
949
950 Conflicts:
951 src/msg/simple/Pipe.cc: nullptr changed to 0 because hammer is not compiled
952 with -std=c++11
953
954 commit 24cc4f9e67d9c61f68ccc8f7d80ac227d67aeda1
955 Author: Nathan Cutler <ncutler@suse.com>
956 Date: Wed Jun 15 16:53:16 2016 +0200
957
958 rgw: check for -ERR_NOT_MODIFIED in rgw_rest_s3.cc
959
960 Fixes: http://tracker.ceph.com/issues/16327
961 Signed-off-by: Nathan Cutler <ncutler@suse.com>
962 (cherry picked from commit fc38346c596d8b0bc156183970d6a327943cb577)
963
964 Conflicts:
965 src/rgw/rgw_rest_s3.cc
966 use ret instead of op_ret to check op result
967
968 commit 7dbace59b83a059fcb364b78fc78d4e38164df6c
969 Author: Loic Dachary <ldachary@redhat.com>
970 Date: Thu May 28 22:35:08 2015 +0200
971
972 erasure-code: s/unreocvery/unfound/
973
974 Signed-off-by: Loic Dachary <ldachary@redhat.com>
975 (cherry picked from commit 28820d1cd630ec466ed1069a437d740a025fb9aa)
976
977 commit e726f211107cd84cc722744c38213b96dc59cdc7
978 Author: Jianpeng Ma <jianpeng.ma@intel.com>
979 Date: Wed Apr 22 09:45:04 2015 +0800
980
981 test: add test-case for repair unrecovery-ec pg.
982
983 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
984 (cherry picked from commit 8f30db85219a6c57cb2fc03f4049a3f57446ca55)
985
986 commit 40b1c2b231b62471d4f3cfba3ba2c3d596c9e170
987 Author: Jianpeng Ma <jianpeng.ma@intel.com>
988 Date: Tue Apr 14 13:29:54 2015 +0800
989
990 osd: Remove the duplicated func MissingLoc::get_all_missing.
991
992 It is the same as MissingLoc::get_needs_recovery.
993
994 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
995 (cherry picked from commit bdd6205408cf20e039496728052a36c1a7d758c6)
996
997 commit 47d5dfc0323e8d7614ad5793d85c1476b63c18e6
998 Author: Jianpeng Ma <jianpeng.ma@intel.com>
999 Date: Tue Apr 14 09:11:58 2015 +0800
1000
1001 osd: Fix ec pg repair endless when met unrecover object.
1002
1003 In repair_object, if bad_peer is replica, it don't add soid in
1004 MissingLoc for ec pool. If there are more bad replica for ec pool
1005 which cause object can't recover, the later recoverying will endless.
1006
1007 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1008 Signed-off-by: Kefu Chai <kchai@redhat.com>
1009 (cherry picked from commit d51806f5b330d5f112281fbb95ea6addf994324e)
1010
1011 commit 187d308b5b0cb0512b442bbad1bddfda8ef9203f
1012 Author: Sage Weil <sage@redhat.com>
1013 Date: Mon Aug 24 14:51:47 2015 -0400
1014
1015 uuid: use boost::random:random_device
1016
1017 The boost mt code uses uninitialized memory for extra randomness,
1018 which is a bad idea in general but more importantly makes valgrind
1019 unhappy. Use /dev/urandom instead.
1020
1021 Unfortunately this introduces a link time dependency.. meh!
1022
1023 Fixes: #12736
1024 Signed-off-by: Sage Weil <sage@redhat.com>
1025 (cherry picked from commit dbcaa544856fcc99ab912a101c4a28e1eb3fb94e)
1026
1027 Conflicts:
1028 ceph.spec.in (trivial resolution)
1029
1030 commit 174de7fce8080df6e02b363e7821e8cca6f8157f
1031 Author: Rohan Mars <code@rohanmars.com>
1032 Date: Wed Aug 12 21:19:31 2015 -0700
1033
1034 moved to use boost uuid implementation, based on commit 4fe89a7b14c97b2ed7f357132901beb2bdcec551
1035
1036 Signed-off-by: Rohan Mars <code@rohanmars.com>
1037 Reviewed-by: Casey Bodley <cbodley@redhat.com>
1038 (cherry picked from commit 62bfc7a1ab1587e81ed3bff0ddfbb1aa69d1c299)
1039
1040 Conflicts:
1041 debian/control (trivial resolution)
1042 src/common/Makefile.am (trivial resolution)
1043 src/common/blkdev.cc (no get_device_by_uuid() function in hammer)
1044
1045 commit 2e156d7ad4b9f4ffd6028df3a460b50b30c8b0d3
1046 Merge: 054a90e 5cd922c
1047 Author: Mykola Golub <to.my.trociny@gmail.com>
1048 Date: Thu Jun 16 08:26:00 2016 +0300
1049
1050 Merge pull request #9722 from trociny/wip-16289-hammer
1051
1052 hammer: qa/workunits/rbd: respect RBD_CREATE_ARGS environment variable
1053
1054 Reviewed-by: Jason Dillaman <dillaman@redhat.com>
1055
1056 commit 5cd922ca31d2eb987992b2afcbd45d5d92aeac69
1057 Author: Mykola Golub <mgolub@mirantis.com>
1058 Date: Tue Jun 14 15:28:06 2016 +0300
1059
1060 qa/workunits/rbd: respect RBD_CREATE_ARGS environment variable
1061
1062 Fixes: http://tracker.ceph.com/issues/16289
1063 Signed-off-by: Mykola Golub <mgolub@mirantis.com>
1064 (cherry picked from commit 3048d3c7ce800a6174c54946b838e4e4079ec609)
1065
1066 commit 1ac920bf291a58cf680108f8dc738197e9bae44d
1067 Author: Yehuda Sadeh <yehuda@redhat.com>
1068 Date: Fri Sep 18 17:34:51 2015 -0700
1069
1070 rgw: fix identification of canceled operation
1071
1072 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1073 (cherry picked from commit 9f63b15b3a7c4cc633ff90f529934f071aea7c62)
1074
1075 commit a38f157a09097d1eaef64813fc947aac19e3844a
1076 Author: Yehuda Sadeh <yehuda@redhat.com>
1077 Date: Wed Mar 25 17:35:40 2015 -0700
1078
1079 rgw: identify racing writes when using copy-if-newer
1080
1081 When copying an object from a different zone, and copy-if-newer is
1082 specified, if the final meta write is canceled check whether the
1083 destinatioin that was created is actually newer than our mtime,
1084 otherwise retry.
1085
1086 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1087 (cherry picked from commit fe9c64b75caf3f53ce2dfdea5e39d794528dfc33)
1088
1089 commit 02f6d8aaa86f4fdb9912763883d079d02589e5be
1090 Author: Yehuda Sadeh <yehuda@redhat.com>
1091 Date: Tue Mar 24 22:48:03 2015 -0700
1092
1093 rgw: translate http error 304 to appropriate status
1094
1095 So that we can return the appropriate response on a conditioinal multi
1096 zone copy operation.
1097
1098 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1099 (cherry picked from commit 20d24e965884af440f94c7e3093587c7b8643cb9)
1100
1101 commit 7319d76b269d541249124f46890f1b04279ba2ea
1102 Author: Yehuda Sadeh <yehuda@redhat.com>
1103 Date: Tue Mar 24 20:24:55 2015 -0700
1104
1105 rgw: fix if-modified-since boundary
1106
1107 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1108 (cherry picked from commit 27946e8ba4bc5dbfec2da319fa23266d5885d655)
1109
1110 commit 5e4de5ad66dc24b059ef19664d1eef22be6890f7
1111 Author: Yehuda Sadeh <yehuda@redhat.com>
1112 Date: Tue Mar 24 20:16:02 2015 -0700
1113
1114 rgw: add rgwx-copy-if-newer http header
1115
1116 Similar to IF_MOD_SINCE, but does not take a time argument, and should
1117 protected against races (not there yet).
1118
1119 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1120 (cherry picked from commit 8813a4401198636e0f4d3a220a0f726625a5c5d4)
1121
1122 commit 054a90edb2812f78426b8cb1dac2e768b2e7fc51
1123 Merge: 785ffd1 c722d00
1124 Author: Orit Wasserman <owasserm@redhat.com>
1125 Date: Tue Jun 7 15:48:00 2016 +0200
1126
1127 Merge pull request #8113 from Vicente-Cheng/wip-15023-hammer
1128
1129 hammer: rgw: user quota may not adjust on bucket removal
1130 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
1131
1132 commit 785ffd1f616864909b24314c803d12566c885d14
1133 Merge: e0a3589 139691c
1134 Author: Samuel Just <sjust@redhat.com>
1135 Date: Mon Jun 6 12:42:14 2016 -0700
1136
1137 Merge pull request #9277 from ceph/wip-split-lfn-hammer
1138
1139 Hammer LFN collection split fix: backport of http://tracker.ceph.com/issues/14766
1140
1141 Reviewed-by: Sage Weil <sage@redhat.com>
1142
1143 commit e0a358972644d3718c80970fd7236a4df8e61afe
1144 Merge: 735c524 006ea56
1145 Author: Josh Durgin <jdurgin@redhat.com>
1146 Date: Mon Jun 6 08:19:23 2016 -0700
1147
1148 Merge pull request #9521 from ceph/revert-9496-wip-16149-hammer
1149
1150 Revert "hammer: Scrub error: 0/1 pinned"
1151
1152 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1153
1154 commit 006ea56f603e160c83c8b1fe35ee8022321bf3b2
1155 Author: Sage Weil <sage@redhat.com>
1156 Date: Mon Jun 6 11:15:54 2016 -0400
1157
1158 Revert "hammer: Scrub error: 0/1 pinned"
1159
1160 commit 735c52423305d4ad548464e986ff88e57333b8ca
1161 Merge: a1a5f57 c294bd3
1162 Author: Sage Weil <sage@redhat.com>
1163 Date: Mon Jun 6 08:46:45 2016 -0400
1164
1165 Merge pull request #9496 from Vicente-Cheng/wip-16149-hammer
1166
1167 hammer: Scrub error: 0/1 pinned
1168
1169 commit a1a5f57558e30fc0fd581756a75da5abee3e897b
1170 Merge: a336291 d1ab71f
1171 Author: Orit Wasserman <owasserm@redhat.com>
1172 Date: Mon Jun 6 11:39:34 2016 +0200
1173
1174 Merge pull request #8987 from dreamhost/hammer-wip-pr7712
1175
1176 hammer: rgw: S3: set EncodingType in ListBucketResult
1177 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
1178
1179 commit c294bd34e7e29f2c96e3277491c02ea26bf6da5c
1180 Author: Samuel Just <sjust@redhat.com>
1181 Date: Fri May 20 13:59:10 2016 -0700
1182
1183 ReplicatedPG: adjust num_pinned in _delete_oid
1184
1185 Fixes: http://tracker.ceph.com/issues/15952
1186 Signed-off-by: Samuel Just <sjust@redhat.com>
1187 (cherry picked from commit 907d4e25c676fd3e1c2be90ce8ab5b64b362b0bc)
1188
1189 commit 43d1b92b97e0bb1c419341ba3c740fa632cafef0
1190 Author: David Zafman <dzafman@redhat.com>
1191 Date: Tue May 17 17:14:07 2016 -0700
1192
1193 test: Fix grouping of mon_cephdf_commands by moving to MON_TESTS
1194
1195 Signed-off-by: David Zafman <dzafman@redhat.com>
1196 (cherry picked from commit 2fb0530a5650bcdb9c89d8bdd2504a7dd1642678)
1197
1198 Conflicts:
1199 qa/workunits/cephtool/test.sh (trivial)
1200
1201 commit 300c11179290d9b119ff6f310efd1cf5b559b3e1
1202 Author: Yehuda Sadeh <yehuda@redhat.com>
1203 Date: Tue Mar 22 18:14:57 2016 -0700
1204
1205 rgw: convert plain object to versioned (with null version) when removing
1206
1207 Fixes #15243
1208
1209 When removing a plain null versioned object (was created prior to bucket versioning
1210 enabled), need to convert the bucket index representation to a versioned one. This
1211 is needed so that all the versioning mechanics play together.
1212
1213 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1214 (cherry picked from commit c6334d430b17739fed2df2b4481ae395ef6568d9)
1215
1216 Conflicts:
1217 src/rgw/rgw_rados.cc
1218 - hammer is missing get_zone() API from which log_data can be
1219 obtained. Needed to fall back to zone_public_config
1220 structure in bucket_index_unlink_instance() definition.
1221 - olh_tag string parameter added to
1222 bucket_index_unlink_instance() definition.
1223 src/rgw/rgw_rados.h
1224 - olh_tag string parameter added to
1225 bucket_index_unlink_instance() declaration.
1226
1227 commit 4eea92b4a50e06d850c39009d122f8bbbd48c857
1228 Author: Yehuda Sadeh <yehuda@redhat.com>
1229 Date: Thu May 5 14:02:25 2016 -0700
1230
1231 rgw: handle stripe transition when flushing final pending_data_bl
1232
1233 Fixes: http://tracker.ceph.com/issues/15745
1234
1235 When complete_writing_data() is called, if pending_data_bl is not empty
1236 we still need to handle stripe transition correctly. If pending_data_bl
1237 has more data that we can allow in current stripe, move to the next one.
1238
1239 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1240 (cherry picked from commit b7a7d41839558b1a8786a66e27d9d80e606f61c7)
1241
1242 Conflicts:
1243 src/rgw/rgw_rados.cc (hammer write_data() takes rgw_obj reference as
1244 4th argument, hammer throttle_data() takes rgw_obj as 2nd argument)
1245
1246 commit f6076dd0416dd4464955f208d8964ef136b6142b
1247 Author: Zhi Zhang <willzzhang@tencent.com>
1248 Date: Tue May 3 17:12:26 2016 +0800
1249
1250 mds: wrongly treat symlink inode as normal file/dir when symlink inode is stale on kcephfs
1251
1252 Signed-off-by: Zhi Zhang <zhangz.david@outlook.com>
1253 (cherry picked from commit 88b6d669e37924536152a552db46ef43a7353562)
1254
1255 Conflicts:
1256 src/mds/Server.cc (hammer has req->head.args.open.flags instead of just flags)
1257
1258 commit ce313cd89e29c0cfc6b7c600a82bdea7b351cdbd
1259 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1260 Date: Tue May 10 17:45:20 2016 +0200
1261
1262 rgw: handle errors properly during GET on Swift's DLO.
1263
1264 Backport: Jewel, Hammer
1265 Fixes: http://tracker.ceph.com/issues/15812
1266 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1267 (cherry picked from commit 5ef0202a86a1e4cb56bd29d926555f114a1acdd8)
1268
1269 Conflicts:
1270 src/rgw/rgw_op.cc (hammer uses ret instead of op_ret; enclosing if
1271 block in hammer does not check skip_manifest)
1272
1273 commit 410ff1528caafaec0cfd9ea6ee4b86ed6788b09d
1274 Author: Sage Weil <sage@redhat.com>
1275 Date: Fri May 6 09:09:43 2016 -0400
1276
1277 osdc/Objecter: upper bound watch_check result
1278
1279 This way we always return a safe upper bound on the amount of time
1280 since we did a check. Among other things, this prevents us from
1281 returning a value of 0, which is confusing.
1282
1283 Fixes: http://tracker.ceph.com/issues/15760
1284 Signed-off-by: Sage Weil <sage@redhat.com>
1285 (cherry picked from commit 2e2ce365866a7569b55aff338a82c132aa927089)
1286
1287 Conflicts:
1288 src/osdc/Objecter.cc
1289 By https://github.com/ceph/ceph/pull/8967, simply add 1 to return vaule.
1290
1291 commit d3eae0a01ae8d33988ffccee1f0af145aa10fa33
1292 Author: Xinze Chi <xinze@xsky.com>
1293 Date: Mon May 23 21:38:39 2016 +0800
1294
1295 osd: fix omap digest compare when scrub
1296
1297 Introduce by fe1c28dea4e5607a0c72eab1f046074616cd55a7.
1298
1299 Fixes: http://tracker.ceph.com/issues/16000
1300 Signed-off-by: Xinze Chi <xinze@xsky.com>
1301
1302 commit a33629106ec64acdaba45bf9db18d2402a926b8e
1303 Merge: b9b5f96 dd29310
1304 Author: Sage Weil <sage@redhat.com>
1305 Date: Sat May 21 14:18:25 2016 -0400
1306
1307 Merge pull request #9240 from yehudasa/wip-15886-hammer
1308
1309 rgw: keep track of written_objs correctly
1310
1311 commit dd29310bd6273ec3643bf90e51f7cae4801629c7
1312 Author: Yehuda Sadeh <yehuda@redhat.com>
1313 Date: Mon May 16 14:35:12 2016 -0700
1314
1315 rgw: keep track of written_objs correctly
1316
1317 Fixes: http://tracker.ceph.com/issues/15886
1318
1319 Only add a rados object to the written_objs list if the write
1320 was successful. Otherwise if the write will be canceled for some
1321 reason, we'd remove an object that we didn't write to. This was
1322 a problem in a case where there's multiple writes that went to
1323 the same part. The second writer should fail the write, since
1324 we do an exclusive write. However, we added the object's name
1325 to the written_objs list anyway, which was a real problem when
1326 the old processor was disposed (as it was clearing the objects).
1327
1328 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1329 (cherry picked from commit 8fd74d11682f9d0c9085d2dc445fc3eb5631f6e0)
1330
1331 commit c2ea6dbd652782bb227b7787c486660871b2d950
1332 Author: Kefu Chai <kchai@redhat.com>
1333 Date: Mon May 9 15:01:46 2016 +0800
1334
1335 osd: remove all stale osdmaps in handle_osd_map()
1336
1337 in a large cluster, there are better chances that the OSD fails to trim
1338 the cached osdmap in a timely manner. and sometimes, it is just unable
1339 to keep up with the incoming osdmap if skip_maps, so the osdmap cache
1340 can keep building up to over 250GB in size. in this change
1341
1342 * publish_superblock() before trimming the osdmaps, so other osdmap
1343 consumers of OSDService.superblock won't access the osdmaps being
1344 removed.
1345 * trim all stale osdmaps in batch of conf->osd_target_transaction_size
1346 if skip_maps is true. in my test, it happens when the osd only
1347 receives the osdmap from monitor occasionally because the osd happens
1348 to be chosen when monitor wants to share a new osdmap with a random
1349 osd.
1350 * always use dedicated transaction(s) for trimming osdmaps. so even in
1351 the normal case where we are able to trim all stale osdmaps in a
1352 single batch, a separated transaction is used. we can piggy back
1353 the commits for removing maps, but we keep it this way for simplicity.
1354 * use std::min() instead MIN() for type safety
1355
1356 Fixes: http://tracker.ceph.com/issues/13990
1357 Signed-off-by: Kefu Chai <kchai@redhat.com>
1358 (cherry picked from commit 369db9930887d75b498927da9c97733bff4472b6)
1359
1360 Conflicts:
1361 remove the C++11ism
1362
1363 commit ac0340ad30329df7919ce82938ad49d0a2e2d160
1364 Author: Kefu Chai <kchai@redhat.com>
1365 Date: Wed Mar 16 21:15:35 2016 +0800
1366
1367 osd: populate the trim_thru epoch using MOSDMap.oldest_map
1368
1369 instead of filling MOSDMap with the local oldest_map, we share
1370 the maximum MOSDMap.oldest_map received so far with peers. That
1371 way one OSD's failure to trim won't prevent it from sharing with
1372 others that they are allowed to trim.
1373
1374 Fixes: #13990
1375 Signed-off-by: Kefu Chai <kchai@redhat.com>
1376 (cherry picked from commit 9789c29fe0f82b236703f7ae4d356251fde3a46f)
1377
1378 Conflict: use atomic_t instead of std::atomic<>
1379
1380 commit b9b5f966ddfe7e706c038db204271ee662e0d825
1381 Merge: b90c097 bb5e015
1382 Author: Sage Weil <sage@redhat.com>
1383 Date: Wed May 18 17:02:26 2016 -0400
1384
1385 Merge pull request #9164 from liewegas/wip-osdmapcrc-hammer
1386
1387 osd: dump full map bl at 20 when crc doesn't match
1388
1389 commit bb5e015d50cb50361dd6ce258c720fe798b3835a
1390 Author: Sage Weil <sage@redhat.com>
1391 Date: Sun Sep 6 13:56:38 2015 -0400
1392
1393 osd: dump full map bl at 20 when crc doesn't match
1394
1395 This will help us debug cases where the encoding doesn't match due
1396 to a bug.
1397
1398 Signed-off-by: Sage Weil <sage@redhat.com>
1399 (cherry picked from commit bfe359af0b80f44ca04847f74d5a2d81097ce4e6)
1400
1401 commit 5057c34ebe4f93792f058ea5f2e3c7e0a29b3d3d
1402 Author: Dmitry Yatsushkevich <dyatsushkevich@mirantis.com>
1403 Date: Mon Mar 23 11:05:44 2015 -0700
1404
1405 obj_bencher: cosmetic display fixes
1406
1407 fix indents in output
1408 change `snprintf' to `setprecision'
1409
1410 Signed-off-by: Dmitry Yatsushkevich <dyatsushkevich@mirantis.com>
1411 (cherry picked from commit ddb422f40394ec2f4fc4a6a6cff49e43538121a9)
1412 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
1413
1414 commit 6d8ad0edf6bae57a6e8ca5cc1a7add1659a92691
1415 Author: Brad Hubbard <bhubbard@redhat.com>
1416 Date: Fri May 6 15:05:42 2016 +1000
1417
1418 common: Add space between timestamp and "min lat:" in bench output
1419
1420 This change is taken from 069d95eaf49cadaa9a8fa1fa186455944a50ec7d
1421 but I did not want to cherry-pick that patch since the rest of it
1422 is purely cosmetic and would be unlikely to apply cleanly.
1423
1424 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
1425
1426 commit b90c097ddd3e172df4534509309a554cb3bc3917
1427 Merge: aefe2cb aab3a40
1428 Author: Loic Dachary <loic@dachary.org>
1429 Date: Mon May 16 13:02:44 2016 +0200
1430
1431 Merge pull request #8535 from asheplyakov/bug-14512-hammer
1432
1433 hammer: rbd snap rollback: restore the link to parent
1434
1435 Reviewed-by: Jason Dillaman <dillaman@redhat.com>
1436
1437 commit 31849986896ba6b5fd2b0b13023b50f67842fd28
1438 Author: Adam Kupczyk <akupczyk@mirantis.com>
1439 Date: Wed Mar 2 12:31:01 2016 +0100
1440
1441 [MON] Fixed calculation of %USED. Now it is shows (space used by all replicas)/(raw space available on OSDs). Before it was (size of pool)/(raw space available on OSDs).
1442
1443 Signed-off-by: Adam Kupczyk akupczyk@mirantis.com
1444 (cherry picked from commit 71c4e525f27b8efd2aa4f3b5e95f4a13f123d41a)
1445
1446 commit fed256e40f5465959384735be957dec88733aeb7
1447 Author: Ruifeng Yang <yangruifeng.09209@h3c.com>
1448 Date: Tue Jan 12 22:08:41 2016 +0800
1449
1450 mon: add a column 'RAW USED' for ceph df detail
1451
1452 Signed-off-by: Ruifeng Yang <yangruifeng.09209@h3c.com>
1453 (cherry picked from commit 2b597ae6d4d53974f28a2da090232577daf09979)
1454
1455 Conflicts:
1456 qa/workunits/cephtool/test.sh (trivial)
1457 src/mon/PGMonitor.cc (trivial)
1458 src/mon/PGMonitor.h (trivial)
1459
1460 commit aefe2cb4f262112f0a028797733f868b83da7bd2
1461 Merge: d56bdf9 221efb0
1462 Author: Loic Dachary <loic@dachary.org>
1463 Date: Thu May 12 23:17:51 2016 +0200
1464
1465 Merge pull request #9107 from SUSE/wip-15867-hammer
1466
1467 hammer: ceph-disk s/by-parttype-uuid/by-parttypeuuid/
1468
1469 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1470
1471 commit 139691c646809f6c25d72c8de064f26b3c036197
1472 Author: Samuel Just <sjust@redhat.com>
1473 Date: Thu May 12 13:50:58 2016 -0700
1474
1475 src/test/objectstore/store_test.cc: fix shards for new tests
1476
1477 This didn't matter on master/jewel, but in hammer, keyvaluestore
1478 asserts that gen != NO_GEN => shard != NO_SHARD.
1479
1480 Signed-off-by: Samuel Just <sjust@redhat.com>
1481
1482 commit 221efb0b893adbfd7a19df171cf967fee87afcc7
1483 Author: Nathan Cutler <ncutler@suse.com>
1484 Date: Thu May 12 20:06:06 2016 +0200
1485
1486 doc: s/by-parttype-uuid/by-parttypeuuid/ in ceph-disk
1487
1488 Fixes: http://tracker.ceph.com/issues/15867
1489 Signed-off-by: Nathan Cutler <ncutler@suse.com>
1490
1491 commit 62f4fbe66f8af79fb874125743df0433aa92b731
1492 Author: Samuel Just <sjust@redhat.com>
1493 Date: Fri Mar 11 17:02:15 2016 -0800
1494
1495 store_test: improve synthetic coverage of the ec stash case
1496
1497 Signed-off-by: Samuel Just <sjust@redhat.com>
1498 (cherry picked from commit f3a78e046d54260de01a4d1db8e104130e35cf42)
1499
1500 Conflicts:
1501 src/test/objectstore/store_test.cc
1502
1503 commit b6bc9cbaef99670bc0fc55936ab02a3656ec8bfd
1504 Author: Samuel Just <sjust@redhat.com>
1505 Date: Fri Mar 11 14:05:37 2016 -0800
1506
1507 store_test: improve synthetic test coverage of long object names
1508
1509 1/2 of objects will now have long names, and we'll shift the region
1510 of the hash space where new objects are created every 1024 objects
1511 created.
1512
1513 Signed-off-by: Samuel Just <sjust@redhat.com>
1514 (cherry picked from commit 4b44720f752ebac63454ecd338e2b0c77c8fe501)
1515
1516 Conflicts:
1517 src/test/objectstore/store_test.cc
1518
1519 commit ec74c1281e641a45de039f2c168b50902e514729
1520 Author: Samuel Just <sjust@redhat.com>
1521 Date: Thu Mar 10 15:38:26 2016 -0800
1522
1523 TestRados: make half the objects have long names
1524
1525 Signed-off-by: Samuel Just <sjust@redhat.com>
1526 (cherry picked from commit 8b64432d07a14df9dfd2ed1314783de16530b1d1)
1527
1528 commit 9d1ee7cf390154c5939a01ff439e33d303a952c2
1529 Author: Samuel Just <sjust@redhat.com>
1530 Date: Thu Mar 10 15:19:15 2016 -0800
1531
1532 LFNIndex::lfn_translate: consider alt attr as well
1533
1534 If the file has an alt attr, there are two possible matching
1535 ghobjects. We want to make sure we choose the right one for
1536 the short name we have. If we don't, a split while there are
1537 two objects linking to the same inode will result in one of
1538 the links being orphaned in the source directory, resulting
1539 in #14766.
1540
1541 Signed-off-by: Samuel Just <sjust@redhat.com>
1542 (cherry picked from commit 850ce11ea3c4dacc6ff1a3ce7ff37000818267c2)
1543
1544 commit 6b821cc8f394e59760e3d58d27f382976bb69e13
1545 Author: Samuel Just <sjust@redhat.com>
1546 Date: Thu Mar 10 11:06:06 2016 -0800
1547
1548 LFNIndex::created: fix return value with failure injection
1549
1550 See comment for explanation.
1551
1552 Signed-off-by: Samuel Just <sjust@redhat.com>
1553 (cherry picked from commit aec5560887a1c1dbbb5aba822280e2c1e05d5d8f)
1554
1555 commit f500435da3ddf750cf0770c0f63e9cff0cebd7a7
1556 Author: Samuel Just <sjust@redhat.com>
1557 Date: Thu Mar 10 11:03:13 2016 -0800
1558
1559 store_test: add reproducer for #14766
1560
1561 Signed-off-by: Samuel Just <sjust@redhat.com>
1562 (cherry picked from commit 0fb67c772561aaca9261ac9094b74d7524498f6f)
1563
1564 Conflicts:
1565 src/test/objectstore/store_test.cc
1566
1567 commit cbd5aafe93a5c85392a460a183862f62af2836b8
1568 Author: Sage Weil <sage@redhat.com>
1569 Date: Fri Apr 29 15:24:09 2016 -0400
1570
1571 osd/PG: update info.stats.* mappings on split
1572
1573 These are updated in the init and start_peering_interval paths, but not
1574 on split.
1575
1576 Fixes: http://tracker.ceph.com/issues/15523
1577 Signed-off-by: Sage Weil <sage@redhat.com>
1578 (cherry picked from commit 8b42294d98dd70715184991da5ba495fbe93d009)
1579
1580 commit d1ab71ff6084719e86e07a447f3054af73e40868
1581 Author: Vitja Makarov <vitja.makarov@gmail.com>
1582 Date: Wed Feb 17 13:46:18 2016 +0300
1583
1584 hammer: rgw: S3: set EncodingType in ListBucketResult
1585
1586 Signed-off-by: Victor Makarov <vitja.makarov@gmail.com>
1587 (cherry picked from commit d2e281d2beb0a49aae0fd939f9387cb2af2692c8)
1588 X-Github-PR: 7712
1589 Backport: hammer
1590 Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
1591
1592 commit df4eadc2376bb4a53210e2fe176520887b840f2f
1593 Author: Gu Zhongyan <guzhongyan@360.cn>
1594 Date: Wed Dec 30 16:42:07 2015 +0800
1595
1596 rados: Add units to rados bench output
1597
1598 Fixes: #12248
1599 Signed-off-by: Gu Zhongyan <guzhongyan@360.com>
1600 Reviewed-by: Sage Weil <sage@redhat.com>
1601 (cherry picked from commit 669b932934d60bbfa751cebd9340f11bd95e2853)
1602 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
1603
1604 commit d96086af5d54a7450f3823210323b9fe08fc075a
1605 Author: Samuel Just <sjust@redhat.com>
1606 Date: Wed Dec 2 17:20:19 2015 -0800
1607
1608 PGLog::rewind_divergent_log: fix rollback_info_trimmed_to before index()
1609
1610 Fixes: #13965
1611 Signed-off-by: Samuel Just <sjust@redhat.com>
1612 (cherry picked from commit 053ee919059b2671d40110e8dff8615dc13d34e0)
1613
1614 commit e79162de4cb418576ca633d5c7ea6fb766706962
1615 Author: Samuel Just <sjust@redhat.com>
1616 Date: Wed Dec 2 17:17:57 2015 -0800
1617
1618 TestPGLog: add test for 13965
1619
1620 Signed-off-by: Samuel Just <sjust@redhat.com>
1621 (cherry picked from commit 66c7246491f4b6e5b95d80cc93c9157725a1a778)
1622
1623 commit fb1b40f5e2da1b70d9ff026a75196c79d09cf0c6
1624 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1625 Date: Sun Mar 22 22:07:24 2015 +0800
1626
1627 osd/Replicated: For CEPH_OSD_OP_WRITE, set data digest.
1628
1629 Add two cases which can add data digest for OP_WRITE:
1630 a: offset = 0, and length > original size
1631 b: offset = original size, and original has data_digest.
1632
1633 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1634 (cherry picked from commit c7702bf85d3617b3e1c6619b8ebeff34932fc3e4)
1635
1636 commit f0242598ea7c990aef2d00f3dc6a0cd65656332c
1637 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1638 Date: Mon Mar 23 09:54:52 2015 +0800
1639
1640 osd/ReplicatedPG: For obj has omap, it mean have omap data or omap header or have both.
1641
1642 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1643 (cherry picked from commit fc51ce2a837e8e878d46d8ca54531aa7bd5d01b2)
1644
1645 commit 7b3f1da0ab7f60705b6bc802ddb1bfbce10c8738
1646 Author: Sage Weil <sage@redhat.com>
1647 Date: Mon Mar 28 10:33:53 2016 -0400
1648
1649 mon/MonClient: fix shutdown race
1650
1651 While we are shutting down, we should drop all subsequent mon
1652 messages. Currently it is possible to finish the authentication
1653 sequence *while* we are shutting down, such that a subsequent
1654 attempt to authenticate succeeds.
1655
1656 Fix this by resetting cur_con early in the sequence, as this
1657 makes us drop all incoming messages.
1658
1659 Fixes: http://tracker.ceph.com/issues/13992
1660 Signed-off-by: Sage Weil <sage@redhat.com>
1661 (cherry picked from commit 5b44a348bfe4978446bbc7acf58eb8f60a47c965)
1662
1663 commit ec02d8b7b1610bd7fd00c201fb01ef80d3b9f05b
1664 Author: Kefu Chai <kchai@redhat.com>
1665 Date: Tue Apr 12 17:31:27 2016 +0800
1666
1667 PG: set epoch_created and parent_split_bits for child pg
1668
1669 if we have a wrong "created", OSD receiving an osd_pg_create message
1670 could fail to project its history and crash. a bad split_bits could
1671 mislead the monitor and osd, and hence cause problems.
1672 * set child.info.history.epoch_created to osdmap.get_epoch() instead of
1673 using the parent's epoch_created
1674 * set child.info.stats.parent_split_bits to the calculated split_bits
1675 instead using the parent's split_bits
1676
1677 Fixes: http://tracker.ceph.com/issues/15426
1678 Signed-off-by: Kefu Chai <kchai@redhat.com>
1679 (cherry picked from commit 8b985561ba08d4ed2d7751a8e09f9d9067c5fbd3)
1680
1681 commit 049bc8a498d4ab516c2daded8ac44699aa72c982
1682 Author: Zhi Zhang <willzzhang@tencent.com>
1683 Date: Thu Jan 14 11:04:42 2016 +0800
1684
1685 ceph-fuse: double decreased the count to trim caps
1686
1687 Signed-off-by: Zhi Zhang zhangz.david@outlook.com
1688 (cherry picked from commit d8a3f6d839649243b12cd6f73e6cb8aec1f955c1)
1689
1690 commit e20df8020458a7b7679d0105183f870220ed0ce7
1691 Author: Sage Weil <sage@redhat.com>
1692 Date: Fri Apr 8 10:35:29 2016 -0400
1693
1694 osd/ReplicatedPG: make handle_watch_timeout no-op if !active
1695
1696 During on_change, we clean up old events on the obcs. This can
1697 include a queued watch timeout:
1698
1699 3: (ReplicatedPG::handle_watch_timeout(std::shared_ptr<Watch>)+0x125) [0x7f1fc21fe375]
1700 4: (HandleDelayedWatchTimeout::finish(int)+0xd3) [0x7f1fc213e2e3]
1701 5: (Context::complete(int)+0x9) [0x7f1fc20ead29]
1702 6: (ReplicatedPG::finish_degraded_object(hobject_t const&)+0x354) [0x7f1fc22429e4]
1703 7: (ReplicatedPG::on_change(ObjectStore::Transaction*)+0x2ba) [0x7f1fc224353a]
1704 8: (PG::start_peering_interval(std::shared_ptr<OSDMap const>, std::vector<int, std::allocator<int> > const&, int, std::vector<int, std::allocator<int> > const&, int, ObjectStore::Transaction*)+0x7bd) [0x7f1fc219a0bd]
1705
1706 In this case, handle_watch_timeout should not assume that we are
1707 active and primary.
1708
1709 Fixes: http://tracker.ceph.com/issues/15391
1710 Signed-off-by: Sage Weil <sage@redhat.com>
1711 (cherry picked from commit 4b0e39ee11e7bd5079ff6704bc74627d3ba8ba44)
1712
1713 commit 3cb72dd57f901f400d19022269fb098c06318d6b
1714 Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
1715 Date: Tue Jun 9 14:57:54 2015 +0200
1716
1717 mon/OSDMonitor.cc: fix UNINTENDED_INTEGER_DIVISION
1718
1719 Fix for:
1720
1721 CID 1297885 (#1 of 2): Result is not floating-point (UNINTENDED_INTEGER_DIVISION)
1722 integer_division: Dividing integer expressions g_conf->mon_pool_quota_warn_threshold
1723 and 100, and then converting the integer quotient to type float. Any remainder,
1724 or fractional part of the quotient, is ignored.
1725
1726 CID 1297885 (#2 of 2): Result is not floating-point (UNINTENDED_INTEGER_DIVISION)
1727 integer_division: Dividing integer expressions g_conf->mon_pool_quota_crit_threshold
1728 and 100, and then converting the integer quotient to type float. Any remainder,
1729 or fractional part of the quotient, is ignored.
1730
1731 Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
1732 (cherry picked from commit be7e07a52871e16e8a46accde6171077cca8db5a)
1733
1734 commit aab3a40f95dafab34a7eadc2159d142a5f0f88f1
1735 Author: Alexey Sheplyakov <asheplyakov@mirantis.com>
1736 Date: Mon Apr 11 14:16:48 2016 +0300
1737
1738 hammer: rbd snap rollback: restore the link to parent
1739
1740 So snapshot, flatten, rollback of a cloned image does not loose any data
1741
1742 Fixes: #14512
1743 Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
1744
1745 commit 488a78794d800312f79f7a8cd75dfda3bf3d6ed0
1746 Author: Samuel Just <sjust@redhat.com>
1747 Date: Fri Jan 8 13:26:39 2016 -0800
1748
1749 ECBackend: send subop writes and reads at normal priority
1750
1751 We want these to be processed ahead of new client ops since
1752 there are resources being held.
1753
1754 Fixes: 14313
1755 Backport: hammer
1756 Signed-off-by: Samuel Just <sjust@redhat.com>
1757 (cherry picked from commit b1151c4b8b7f22c7ac067da059d087007b0bc8ea)
1758
1759 commit a2e7ca15d6fd0326526b77f443c15067eb1840ac
1760 Author: Richard W.M. Jones <rjones@redhat.com>
1761 Date: Mon Mar 21 14:24:35 2016 +0000
1762
1763 common/Cycles: Do not initialize Cycles globally.
1764
1765 Move initialization to the few tests that actually use it.
1766
1767 Fixes: http://tracker.ceph.com/issues/15225
1768 Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
1769 (cherry picked from commit 1c2831a2c1277c69f9649200d74a75c04a4b0296)
1770
1771 Conflicts:
1772 src/test/msgr/perf_msgr_client.cc
1773 src/test/msgr/perf_msgr_server.cc
1774 src/test/perf_local.cc
1775 These three files were not introduced on hammer, just remove
1776
1777 commit ca0beef84812b09f1ab224e4796fd44c8fe37989
1778 Author: David Zafman <dzafman@redhat.com>
1779 Date: Fri Mar 18 10:42:40 2016 -0700
1780
1781 osd/OSD: fix build_past_intervals_parallel
1782
1783 We may be only calculating older past intervals and have a valid
1784 history.same_interval_since value, in which case the local
1785 same_interval_since value will end at the newest old interval we had to
1786 generate.
1787
1788 Same as 0830275b39afc408573c2c468fa3b0e7b1d5cc4e change in generate_past_intervals()
1789 This was introduced by 70316541bbb115d9a35954bfba373cf1dc084b7e.
1790
1791 Signed-off-by: David Zafman <dzafman@redhat.com>
1792 (cherry picked from commit 0fd674bbf0c17a673be40123645adee3d64375a0)
1793
1794 commit fce79027034bef0cfedbecd9d2601589495f86e6
1795 Author: David Zafman <dzafman@redhat.com>
1796 Date: Mon Jul 20 17:48:15 2015 -0700
1797
1798 osd: When generating past intervals due to an import end at pg epoch
1799
1800 Add assert() to make sure same_interval_since isn't too far forward
1801
1802 Fixes: #12387
1803
1804 Signed-off-by: David Zafman <dzafman@redhat.com>
1805 (cherry picked from commit 65dcc2da76750d0b6dd2cf0031c44f32749f33e5)
1806
1807 commit 2c97cb33c8aa3f53af8a331075a3f79e70113cea
1808 Author: xie xingguo <xie.xingguo@zte.com.cn>
1809 Date: Thu Mar 31 19:48:38 2016 +0800
1810
1811 rgw: fix compiling error
1812
1813 Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
1814 (cherry picked from commit 535eb0e608f3bf6b7c561f277e9977b91ab97b48)
1815
1816 commit 2aa1ea6b505c768323abb99dcc8c8f24e594a898
1817 Author: Robin H. Johnson <robin.johnson@dreamhost.com>
1818 Date: Thu Mar 31 06:24:40 2016 +0000
1819
1820 rgw: Multipart ListPartsResult ETag quotes
1821
1822 ListPartsResult output has always missed quotes on the ETag since it was
1823 first committed.
1824
1825 Fixes: #15334
1826 Backports: hammer, infernalis
1827 Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
1828 (cherry picked from commit a58b774e72cc1613d62e10b25322d6d15e9d2899)
1829
1830 commit 365f21b1181b08e0bb89b916833301228e7c9efc
1831 Author: Loic Dachary <ldachary@redhat.com>
1832 Date: Wed Apr 6 15:39:23 2016 +0200
1833
1834 tests: be more generous with test timeout
1835
1836 When the thrasher is in action together with a validater (lockdep or
1837 valgrind), a single test may hang for more than 360 seconds. Increase to
1838 1200: it does not matter if the value is large, only that it prevents
1839 the test from hanging forever.
1840
1841 Fixes: http://tracker.ceph.com/issues/15403
1842
1843 Signed-off-by: Loic Dachary <loic@dachary.org>
1844 (cherry picked from commit af89474b3fb2c4aa63680aa6b30d71fad2fdd373)
1845
1846 commit c722d00af5286785ad23c574359220e47fd15076
1847 Author: Vicente Cheng <freeze.bilsted@gmail.com>
1848 Date: Tue Feb 9 12:03:24 2016 -0800
1849
1850 rgw: user quota may not adjust on bucket removal
1851
1852 Description:
1853 If the user/admin removes a bucket using --force/--purge-objects options with s3cmd/radosgw-admin respectively, the user stats will continue to reflect the deleted objects for quota purposes, and there seems to be no way to reset them. User stats need to be sync'ed prior to bucket removal.
1854
1855 Solution:
1856 Sync user stats before removing a bucket.
1857
1858 Fixes: #14507
1859 Signed-off-by: Edward Yang <eyang@us.fujitsu.com>
1860 (cherry picked from commit 06507992e3fad837c4667f467d59fc026a2ca0e8)
1861
1862 Conflicts:
1863
1864 src/rgw/rgw_op.cc
1865 reordering the check seqence and replace some op_ret to ret
1866
1867 Backport Change:
1868 We remove the `tenant` parameter because this feature is not introduced on hammer version.
1869 The rgw multi-tenant feature is introduced on pr#6784 (https://github.com/ceph/ceph/pull/6784)
1870 This feature is supported from v10.0.2 and later version.
1871
1872 commit 20f300e2900ff123974312143c10dca59ee1a64a
1873 Author: Wido den Hollander <wido@42on.com>
1874 Date: Tue Mar 22 11:35:05 2016 +0100
1875
1876 rgw: Do not send a Content-Length header on a 304 response
1877
1878 We tell the client that the content has not changed. If we
1879 send a Content-Length header RFC2616 describes that the client
1880 MUST use that new value:
1881
1882 "If a cache uses a received 304 response to update a cache entry,
1883 the cache MUST update the entry to reflect any new field values
1884 given in the response."
1885
1886 Therefor we should not send a Content-Length header
1887
1888 Fixes: #15119
1889
1890 Signed-off-by: Wido den Hollander <wido@42on.com>
1891 (cherry picked from commit fb4e5cc5d6adc82423d294496855f56e43e8c11d)
1892
1893 commit e53751d0e52f6fb611d9bd81fe2115cd7ce533b3
1894 Author: Wido den Hollander <wido@42on.com>
1895 Date: Mon Mar 21 16:50:49 2016 +0100
1896
1897 rgw: Do not send a Content-Type on a '304 Not Modified' response
1898
1899 When we say the Content has not changed we should not respond
1900 with a content type which defaults to binary/octet stream.
1901
1902 Fixes: #15119
1903
1904 Signed-off-by: Wido den Hollander <wido@42on.com>
1905 (cherry picked from commit 471fa29e643fcfd5e1383fb987f99b5b21038ed7)
1906
1907 commit 19dbc2598d15d0676017abe4fb75f7ddc3248119
1908 Author: Yehuda Sadeh <yehuda@redhat.com>
1909 Date: Fri Dec 18 10:18:22 2015 -0800
1910
1911 rgw: dump_status() uses integer
1912
1913 so that we can avoid atoi() later
1914
1915 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1916 (cherry picked from commit 744a29a34b72d2a709a58a5ea5fb6223a39c676a)
1917
1918 Conflicts:
1919 src/rgw/rgw_rest.cc
1920 rename the input parameter from `err` to `http_ret`
1921
1922 commit c79b481311b3cd992794abc6e2f194f350d43531
1923 Author: Yehuda Sadeh <yehuda@redhat.com>
1924 Date: Fri Dec 18 10:07:27 2015 -0800
1925
1926 rgw: move status_num initialization into constructor
1927
1928 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1929 (cherry picked from commit 2a12ffc3d7f98504ece8906a6fcb5dac8fd81aba)
1930
1931 commit ceb8e19096f7d4ef35ed3ad8fad508aecbfa78ef
1932 Author: Wido den Hollander <wido@42on.com>
1933 Date: Wed Nov 11 19:01:15 2015 +0100
1934
1935 rgw: Do not send a Content-Length header on status 204
1936
1937 RFC7230 says:
1938 A server MUST NOT send a Content-Length header field in any response with a
1939 status code of 1xx (Informational) or 204 (No Content).
1940
1941 Fixes: #13582
1942 Signed-off-by: Wido den Hollander <wido@42on.com>
1943 (cherry picked from commit 4e5921dbc7d1c51feb4cc5c03aa59a432742765e)
1944
1945 commit 3ecdedd5cfe67d85530123ed6684688b43910aa6
1946 Author: Yan, Zheng <zyan@redhat.com>
1947 Date: Fri Mar 11 11:13:43 2016 +0800
1948
1949 mds: fix stray purging in 'stripe_count > 1' case
1950
1951 Fixes: #15050
1952 Signed-off-by: Yan, Zheng <zyan@redhat.com>
1953 (cherry picked from commit 99ec183973e0f4686673d93df02e5381543be51f)
1954
1955 commit f28477c9cf904e9b27e6ad6bb38fbeeb80db5017
1956 Author: Dunrong Huang <riegamaths@gmail.com>
1957 Date: Mon Dec 28 18:55:37 2015 +0800
1958
1959 rgw: do not abort when user passed bad parameters to set quota
1960
1961 when user passes bad parameter(e.g. pass "null" to bool parameter), the ceph_json.cc::decode_json_obj()
1962 will raise a exception with type of JSONDecoder::err(). If we do not catch this exception, the radosgw
1963 server will abort.
1964
1965 Fixes: #14191
1966
1967 $ curl -X PUT http://ceph1:9090/admin/user?quota&uid=test5&quota-type=user -d '{"enabled": "null"}'
1968
1969 0> 2015-12-28 11:25:20.864091 7f0d46fb5700 -1 *** Caught signal (Aborted) **
1970 in thread 7f0d46fb5700
1971
1972 ceph version 10.0.1-787-gc485d29 (c485d29a53f6a2c1a3b561c54632dbac1c4e9473)
1973 1: (ceph::BackTrace::BackTrace(int)+0x2d) [0x7f0e13ef88c5]
1974 2: (()+0x8fbfb7) [0x7f0e13ef7fb7]
1975 3: (()+0xf100) [0x7f0e082c3100]
1976 4: (gsignal()+0x37) [0x7f0e072fd5f7]
1977 5: (abort()+0x148) [0x7f0e072fece8]
1978 6: (__gnu_cxx::__verbose_terminate_handler()+0x165) [0x7f0e07c019b5]
1979 7: (()+0x5e926) [0x7f0e07bff926]
1980 8: (()+0x5e953) [0x7f0e07bff953]
1981 9: (()+0x5eb73) [0x7f0e07bffb73]
1982 10: (bool JSONDecoder::decode_json<bool>(char const*, bool&, JSONObj*, bool)+0x288) [0x7f0e13d74c2c]
1983 11: (RGWQuotaInfo::decode_json(JSONObj*)+0x6c) [0x7f0e13d71128]
1984 12: (void decode_json_obj<RGWQuotaInfo>(RGWQuotaInfo&, JSONObj*)+0x23) [0x7f0e13d21f16]
1985 13: (int rgw_rest_get_json_input<RGWQuotaInfo>(CephContext*, req_state*, RGWQuotaInfo&, int, bool*)+0x11e) [0x7f0e13d21e48]
1986 14: (RGWOp_Quota_Set::execute()+0x8c9) [0x7f0e13d1ed6d]
1987 15: (()+0x7532d2) [0x7f0e13d4f2d2
1988
1989 Reported-by: Coffee Chou <coffee.zyr@gmail.com>
1990 Signed-off-by: Dunrong Huang <riegamaths@gmail.com>
1991 (cherry picked from commit 59f9058467480d21bd98e0e90aa97313a92478a1)
1992
1993 Conflicts:
1994 src/rgw/rgw_rest.h
1995 hammer version lacks rgw_rest_get_json_input_keep_data()
1996 template
1997
1998 commit 9786394b2f75d883505125bc45c95fb685b96531
1999 Author: Dunrong Huang <riegamaths@gmail.com>
2000 Date: Mon Dec 28 18:14:22 2015 +0800
2001
2002 rgw: do not abort when user passed bad parameters to set metadata
2003
2004 Fixes: #14190
2005
2006 when user passes bad parameter(e.g. pass "null" to bool parameter), the ceph_json.cc::decode_json_obj()
2007 will raise a exception with type of JSONDecoder::err(). If we do not catch this exception, the radosgw
2008 server will abort.
2009
2010 $ curl -X PUT http://ceph1:9090/admin/metadata?user&format=json -d 'data = { "mtime": "null"}'
2011
2012 terminate called after throwing an instance of 'JSONDecoder::err'
2013 *** Caught signal (Aborted) **
2014 in thread 7ff982f55700
2015 ceph version 10.0.1-787-gc485d29 (c485d29a53f6a2c1a3b561c54632dbac1c4e9473)
2016 1: (ceph::BackTrace::BackTrace(int)+0x2d) [0x7ffa814db8c5]
2017 2: (()+0x8fbfb7) [0x7ffa814dafb7]
2018 3: (()+0xf100) [0x7ffa758a6100]
2019 4: (gsignal()+0x37) [0x7ffa748e05f7]
2020 5: (abort()+0x148) [0x7ffa748e1ce8]
2021 6: (__gnu_cxx::__verbose_terminate_handler()+0x165) [0x7ffa751e49b5]
2022 7: (()+0x5e926) [0x7ffa751e2926]
2023 8: (()+0x5e953) [0x7ffa751e2953]
2024 9: (()+0x5eb73) [0x7ffa751e2b73]
2025 10: (bool JSONDecoder::decode_json<long>(char const*, long&, JSONObj*, bool)+0x28c) [0x7ffa8135920a]
2026 11: (RGWMetadataManager::put(std::string&, ceph::buffer::list&, RGWMetadataHandler::sync_type_t, obj_version*)+0x173) [0x7ffa81468029]
2027 12: (RGWOp_Metadata_Put::execute()+0x172) [0x7ffa81308dcc]
2028 13: (()+0x7532d2) [0x7ffa813322d2]
2029 14: (RGWFCGXProcess::handle_request(RGWRequest*)+0x90) [0x7ffa813328a4]
2030
2031 Reported-by: Coffee Chou <coffee.zyr@gmail.com>
2032 Signed-off-by: Dunrong Huang <riegamaths@gmail.com>
2033 (cherry picked from commit 83503239c81573232429a5fe98381f1a8f1357e9)
2034
2035 Conflicts:
2036 src/rgw/rgw_user.cc
2037 hammer version uses RGWUserInfo instead of RGWUserCompleteInfo
2038
2039 commit b9a4ad9da99d82bd247e5d39108840f19535fa5e
2040 Author: Vicente Cheng <freeze.bilsted@gmail.com>
2041 Date: Fri Mar 27 18:49:28 2015 +0800
2042
2043 Fixed the ceph get mdsmap assertion.
2044
2045 When we want to get mdsmap, we try to get_version()
2046 and the return value err = 0 means success.
2047
2048 The assert verified r == 0. r would not change in this flow.
2049 It always meet assert and lead mon failure.
2050
2051 I think this verify should be:
2052 assert(err == 0)
2053 It will help to check return value of get_version().
2054
2055 If you have any questions, feel free to let me know.
2056 Thanks!
2057
2058 Signed-off-by: Vicente Cheng <freeze.bilsted@gmail.com>
2059 (cherry picked from commit f4398d2e6c245e3f81a6038425e1b8372b265b8c)