[mirror_zfs.git] / tests / zfs-tests / tests / functional / cli_root / zpool_import / import_cachefile_device_replaced.ksh

#!/bin/ksh -p

#
# This file and its contents are supplied under the terms of the
# Common Development and Distribution License ("CDDL"), version 1.0.
# You may only use this file in accordance with the terms of version
# 1.0 of the CDDL.
#
# A full copy of the text of the CDDL should have accompanied this
# source.  A copy of the CDDL is also available via the Internet at
# http://www.illumos.org/license/CDDL.
#

#
# Copyright (c) 2016 by Delphix. All rights reserved.
#

. $STF_SUITE/tests/functional/cli_root/zpool_import/zpool_import.kshlib

#
# DESCRIPTION:
#	A pool should be importable using an outdated cachefile that is unaware
#	of a zpool replace operation at different stages in time.
#
# STRATEGY:
#	1. Create a pool with some devices and an alternate cachefile.
#	2. Backup the cachefile.
#	3. Initiate device replacement, backup cachefile again and export pool.
#	   Special care must be taken so that resilvering doesn't complete
#	   before we exported the pool.
#	4. Verify that we can import the pool using the first cachefile backup.
#	   (Test 1. cachefile: pre-replace, pool: resilvering)
#	5. Wait for the resilvering to finish and export the pool.
#	6. Verify that we can import the pool using the first cachefile backup.
#	   (Test 2. cachefile: pre-replace, pool: post-replace)
#	7. Export the pool.
#	8. Verify that we can import the pool using the second cachefile backup.
#	   (Test 3. cachefile: resilvering, pool: post-replace)
#
# STRATEGY TO SLOW DOWN RESILVERING:
#	1. Reduce zfs_txg_timeout, which controls how long can we resilver for
#	   each sync.
#	2. Add data to pool
#	3. Re-import the pool so that data isn't cached
#	4. Use zfs_scan_suspend_progress to ensure resilvers don't progress
#	5. Trigger the resilvering
#	6. Use spa freeze to stop writing to the pool.
#	7. Re-enable scan progress
#	8. Export the pool
#

verify_runnable "global"

ZFS_TXG_TIMEOUT=""

function custom_cleanup
{
	# Revert zfs_txg_timeout to defaults
	[[ -n ZFS_TXG_TIMEOUT ]] &&
	    log_must set_zfs_txg_timeout $ZFS_TXG_TIMEOUT

	log_must set_tunable32 SCAN_SUSPEND_PROGRESS 0
	cleanup
}

log_onexit custom_cleanup

function test_replacing_vdevs
{
	typeset poolcreate="$1"
	typeset replacevdev="$2"
	typeset replaceby="$3"
	typeset poolfinalstate="$4"
	typeset zinjectdevices="$5"
	typeset earlyremove="$6"
	typeset writedata="$7"

	log_note "$0: pool '$poolcreate', replace $replacevdev by $replaceby."

	log_must zpool create -o cachefile=$CPATH $TESTPOOL1 $poolcreate

	# Cachefile: pool in pre-replace state
	log_must cp $CPATH $CPATHBKP

	# Steps to insure resilvering happens very slowly.
	log_must write_some_data $TESTPOOL1 $writedata
	log_must zpool export $TESTPOOL1
	log_must cp $CPATHBKP $CPATH
	log_must zpool import -c $CPATH -o cachefile=$CPATH $TESTPOOL1
	log_must set_tunable32 SCAN_SUSPEND_PROGRESS 1
	log_must zpool replace $TESTPOOL1 $replacevdev $replaceby

	# Cachefile: pool in resilvering state
	log_must cp $CPATH $CPATHBKP2

	# Confirm pool is still replacing
	log_must pool_is_replacing $TESTPOOL1
	log_must zpool export $TESTPOOL1
	log_must set_tunable32 SCAN_SUSPEND_PROGRESS 0

	( $earlyremove ) && log_must rm $replacevdev

	############################################################
	# Test 1. Cachefile: pre-replace, pool: resilvering
	############################################################
	log_must cp $CPATHBKP $CPATH
	log_must zpool import -c $CPATH $TESTPOOL1

	# Wait for resilvering to finish
	log_must wait_for_pool_config $TESTPOOL1 "$poolfinalstate"
	log_must zpool export $TESTPOOL1

	( ! $earlyremove ) && log_must rm $replacevdev

	############################################################
	# Test 2. Cachefile: pre-replace, pool: post-replace
	############################################################
	log_must zpool import -c $CPATHBKP $TESTPOOL1
	log_must check_pool_config $TESTPOOL1 "$poolfinalstate"
	log_must zpool export $TESTPOOL1

	############################################################
	# Test 3. Cachefile: resilvering, pool: post-replace
	############################################################
	log_must zpool import -c $CPATHBKP2 $TESTPOOL1
	log_must check_pool_config $TESTPOOL1 "$poolfinalstate"

	# Cleanup
	log_must zpool destroy $TESTPOOL1
	log_must rm -f $CPATH $CPATHBKP $CPATHBKP2
	log_must mkfile $FILE_SIZE $replacevdev

	log_note ""
}

# We set zfs_txg_timeout to 1 to reduce resilvering time at each sync.
ZFS_TXG_TIMEOUT=$(get_zfs_txg_timeout)
set_zfs_txg_timeout 1

test_replacing_vdevs "$VDEV0 $VDEV1" \
    "$VDEV1" "$VDEV2" \
    "$VDEV0 $VDEV2" \
    "$VDEV0 $VDEV1" \
    false 20

test_replacing_vdevs "mirror $VDEV0 $VDEV1" \
	"$VDEV1" "$VDEV2" \
	"mirror $VDEV0 $VDEV2" \
	"$VDEV0 $VDEV1" \
	true 10

test_replacing_vdevs "raidz $VDEV0 $VDEV1 $VDEV2" \
	"$VDEV1" "$VDEV3" \
	"raidz $VDEV0 $VDEV3 $VDEV2" \
	"$VDEV0 $VDEV1 $VDEV2" \
	true 20

test_replacing_vdevs "draid:1s $VDEV0 $VDEV1 $VDEV2 $VDEV3 $VDEV4" \
	"$VDEV1" "$VDEV5" \
	"draid $VDEV0 $VDEV5 $VDEV2 $VDEV3 $VDEV4 spares draid1-0-0" \
	"$VDEV0 $VDEV1 $VDEV2 $VDEV3 $VDEV4" \
	true 30

set_zfs_txg_timeout $ZFS_TXG_TIMEOUT

log_pass "zpool import -c cachefile_unaware_of_replace passed."
Commit	Line	Data
29badadd	1	#!/bin/ksh -p
6cb8e530 PZ	2
	3	#
	4	# This file and its contents are supplied under the terms of the
	5	# Common Development and Distribution License ("CDDL"), version 1.0.
	6	# You may only use this file in accordance with the terms of version
	7	# 1.0 of the CDDL.
	8	#
	9	# A full copy of the text of the CDDL should have accompanied this
	10	# source. A copy of the CDDL is also available via the Internet at
	11	# http://www.illumos.org/license/CDDL.
	12	#
	13
	14	#
	15	# Copyright (c) 2016 by Delphix. All rights reserved.
	16	#
	17
	18	. $STF_SUITE/tests/functional/cli_root/zpool_import/zpool_import.kshlib
	19
	20	#
	21	# DESCRIPTION:
	22	# A pool should be importable using an outdated cachefile that is unaware
	23	# of a zpool replace operation at different stages in time.
	24	#
	25	# STRATEGY:
	26	# 1. Create a pool with some devices and an alternate cachefile.
	27	# 2. Backup the cachefile.
	28	# 3. Initiate device replacement, backup cachefile again and export pool.
	29	# Special care must be taken so that resilvering doesn't complete
	30	# before we exported the pool.
	31	# 4. Verify that we can import the pool using the first cachefile backup.
	32	# (Test 1. cachefile: pre-replace, pool: resilvering)
	33	# 5. Wait for the resilvering to finish and export the pool.
	34	# 6. Verify that we can import the pool using the first cachefile backup.
	35	# (Test 2. cachefile: pre-replace, pool: post-replace)
	36	# 7. Export the pool.
	37	# 8. Verify that we can import the pool using the second cachefile backup.
	38	# (Test 3. cachefile: resilvering, pool: post-replace)
	39	#
	40	# STRATEGY TO SLOW DOWN RESILVERING:
	41	# 1. Reduce zfs_txg_timeout, which controls how long can we resilver for
	42	# each sync.
	43	# 2. Add data to pool
	44	# 3. Re-import the pool so that data isn't cached
cef48f14	45	# 4. Use zfs_scan_suspend_progress to ensure resilvers don't progress
6cb8e530 PZ	46	# 5. Trigger the resilvering
6cb8e530 PZ	47	# 6. Use spa freeze to stop writing to the pool.
cef48f14	48	# 7. Re-enable scan progress
6cb8e530 PZ	49	# 8. Export the pool
	50	#
	51
	52	verify_runnable "global"
	53
	54	ZFS_TXG_TIMEOUT=""
	55
	56	function custom_cleanup
	57	{
	58	# Revert zfs_txg_timeout to defaults
	59	[[ -n ZFS_TXG_TIMEOUT ]] &&
	60	log_must set_zfs_txg_timeout $ZFS_TXG_TIMEOUT
	61
2476f103	62	log_must set_tunable32 SCAN_SUSPEND_PROGRESS 0
6cb8e530 PZ	63	cleanup
	64	}
	65
	66	log_onexit custom_cleanup
	67
	68	function test_replacing_vdevs
	69	{
	70	typeset poolcreate="$1"
	71	typeset replacevdev="$2"
	72	typeset replaceby="$3"
	73	typeset poolfinalstate="$4"
	74	typeset zinjectdevices="$5"
	75	typeset earlyremove="$6"
	76	typeset writedata="$7"
	77
	78	log_note "$0: pool '$poolcreate', replace $replacevdev by $replaceby."
	79
	80	log_must zpool create -o cachefile=$CPATH $TESTPOOL1 $poolcreate
	81
	82	# Cachefile: pool in pre-replace state
	83	log_must cp $CPATH $CPATHBKP
	84
	85	# Steps to insure resilvering happens very slowly.
	86	log_must write_some_data $TESTPOOL1 $writedata
	87	log_must zpool export $TESTPOOL1
	88	log_must cp $CPATHBKP $CPATH
	89	log_must zpool import -c $CPATH -o cachefile=$CPATH $TESTPOOL1
2476f103	90	log_must set_tunable32 SCAN_SUSPEND_PROGRESS 1
6cb8e530 PZ	91	log_must zpool replace $TESTPOOL1 $replacevdev $replaceby
	92
	93	# Cachefile: pool in resilvering state
	94	log_must cp $CPATH $CPATHBKP2
	95
6cb8e530 PZ	96	# Confirm pool is still replacing
6cb8e530 PZ	97	log_must pool_is_replacing $TESTPOOL1
6cb8e530	98	log_must zpool export $TESTPOOL1
2476f103	99	log_must set_tunable32 SCAN_SUSPEND_PROGRESS 0
6cb8e530 PZ	100
	101	( $earlyremove ) && log_must rm $replacevdev
	102
	103	############################################################
	104	# Test 1. Cachefile: pre-replace, pool: resilvering
	105	############################################################
	106	log_must cp $CPATHBKP $CPATH
	107	log_must zpool import -c $CPATH $TESTPOOL1
	108
	109	# Wait for resilvering to finish
	110	log_must wait_for_pool_config $TESTPOOL1 "$poolfinalstate"
	111	log_must zpool export $TESTPOOL1
	112
	113	( ! $earlyremove ) && log_must rm $replacevdev
	114
	115	############################################################
	116	# Test 2. Cachefile: pre-replace, pool: post-replace
	117	############################################################
	118	log_must zpool import -c $CPATHBKP $TESTPOOL1
	119	log_must check_pool_config $TESTPOOL1 "$poolfinalstate"
	120	log_must zpool export $TESTPOOL1
	121
	122	############################################################
	123	# Test 3. Cachefile: resilvering, pool: post-replace
	124	############################################################
	125	log_must zpool import -c $CPATHBKP2 $TESTPOOL1
	126	log_must check_pool_config $TESTPOOL1 "$poolfinalstate"
	127
	128	# Cleanup
	129	log_must zpool destroy $TESTPOOL1
	130	log_must rm -f $CPATH $CPATHBKP $CPATHBKP2
	131	log_must mkfile $FILE_SIZE $replacevdev
	132
	133	log_note ""
	134	}
	135
	136	# We set zfs_txg_timeout to 1 to reduce resilvering time at each sync.
	137	ZFS_TXG_TIMEOUT=$(get_zfs_txg_timeout)
	138	set_zfs_txg_timeout 1
	139
	140	test_replacing_vdevs "$VDEV0 $VDEV1" \
	141	"$VDEV1" "$VDEV2" \
	142	"$VDEV0 $VDEV2" \
	143	"$VDEV0 $VDEV1" \
	144	false 20
	145
	146	test_replacing_vdevs "mirror $VDEV0 $VDEV1" \
	147	"$VDEV1" "$VDEV2" \
	148	"mirror $VDEV0 $VDEV2" \
	149	"$VDEV0 $VDEV1" \
	150	true 10
	151
	152	test_replacing_vdevs "raidz $VDEV0 $VDEV1 $VDEV2" \
	153	"$VDEV1" "$VDEV3" \
	154	"raidz $VDEV0 $VDEV3 $VDEV2" \
	155	"$VDEV0 $VDEV1 $VDEV2" \
	156	true 20
	157
b2255edc BB	158	test_replacing_vdevs "draid:1s $VDEV0 $VDEV1 $VDEV2 $VDEV3 $VDEV4" \
	159	"$VDEV1" "$VDEV5" \
	160	"draid $VDEV0 $VDEV5 $VDEV2 $VDEV3 $VDEV4 spares draid1-0-0" \
	161	"$VDEV0 $VDEV1 $VDEV2 $VDEV3 $VDEV4" \
	162	true 30
	163
6cb8e530 PZ	164	set_zfs_txg_timeout $ZFS_TXG_TIMEOUT
	165
	166	log_pass "zpool import -c cachefile_unaware_of_replace passed."