mirror of
https://git.proxmox.com/git/mirror_zfs.git
synced 2026-05-22 10:37:35 +03:00
Prevent unnecessary resilver restarts
If a device is participating in an active resilver, then it will have a
non-empty DTL. Operations like vdev_{open,reopen,probe}() can cause the
resilver to be restarted (or deferred to be restarted later), which is
unnecessary if the DTL is still covered by the current scan range. This
is similar to the logic in vdev_dtl_should_excise() where the DTL can
only be excised if it's max txg is in the resilvered range.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: John Gallagher <john.gallagher@delphix.com>
Reviewed-by: Kjeld Schouten <kjeld@schouten-lebbing.nl>
Signed-off-by: John Poduska <jpoduska@datto.com>
Issue #840
Closes #9155
Closes #9378
Closes #9551
Closes #9588
This commit is contained in:
jwpoduska
committed by
Brian Behlendorf
Brian Behlendorf
parent
894f6696b4
commit
3c819a2c7d
@@ -761,6 +761,10 @@ tests = ['reservation_001_pos', 'reservation_002_pos', 'reservation_003_pos',
|
||||
'reservation_022_pos']
|
||||
tags = ['functional', 'reservation']
|
||||
|
||||
[tests/functional/resilver]
|
||||
tests = ['resilver_restart_001']
|
||||
tags = ['functional', 'resilver']
|
||||
|
||||
[tests/functional/rootpool]
|
||||
tests = ['rootpool_002_neg', 'rootpool_003_neg', 'rootpool_007_pos']
|
||||
tags = ['functional', 'rootpool']
|
||||
|
||||
@@ -61,6 +61,7 @@ SUBDIRS = \
|
||||
rename_dirs \
|
||||
replacement \
|
||||
reservation \
|
||||
resilver \
|
||||
rootpool \
|
||||
rsend \
|
||||
scrub_mirror \
|
||||
|
||||
@@ -0,0 +1,8 @@
|
||||
pkgdatadir = $(datadir)/@PACKAGE@/zfs-tests/tests/functional/resilver
|
||||
dist_pkgdata_SCRIPTS = \
|
||||
setup.ksh \
|
||||
cleanup.ksh \
|
||||
resilver_restart_001.ksh
|
||||
|
||||
dist_pkgdata_DATA = \
|
||||
resilver.cfg
|
||||
+31
@@ -0,0 +1,31 @@
|
||||
#!/bin/ksh -p
|
||||
#
|
||||
# CDDL HEADER START
|
||||
#
|
||||
# The contents of this file are subject to the terms of the
|
||||
# Common Development and Distribution License (the "License").
|
||||
# You may not use this file except in compliance with the License.
|
||||
#
|
||||
# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
|
||||
# or http://www.opensolaris.org/os/licensing.
|
||||
# See the License for the specific language governing permissions
|
||||
# and limitations under the License.
|
||||
#
|
||||
# When distributing Covered Code, include this CDDL HEADER in each
|
||||
# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
|
||||
# If applicable, add the following below this CDDL HEADER, with the
|
||||
# fields enclosed by brackets "[]" replaced with your own identifying
|
||||
# information: Portions Copyright [yyyy] [name of copyright owner]
|
||||
#
|
||||
# CDDL HEADER END
|
||||
|
||||
#
|
||||
# Copyright (c) 2019, Datto Inc. All rights reserved.
|
||||
#
|
||||
|
||||
. $STF_SUITE/include/libtest.shlib
|
||||
. $STF_SUITE/tests/functional/resilver/resilver.cfg
|
||||
|
||||
verify_runnable "global"
|
||||
|
||||
log_pass
|
||||
@@ -0,0 +1,32 @@
|
||||
#
|
||||
# CDDL HEADER START
|
||||
#
|
||||
# The contents of this file are subject to the terms of the
|
||||
# Common Development and Distribution License (the "License").
|
||||
# You may not use this file except in compliance with the License.
|
||||
#
|
||||
# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
|
||||
# or http://www.opensolaris.org/os/licensing.
|
||||
# See the License for the specific language governing permissions
|
||||
# and limitations under the License.
|
||||
#
|
||||
# When distributing Covered Code, include this CDDL HEADER in each
|
||||
# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
|
||||
# If applicable, add the following below this CDDL HEADER, with the
|
||||
# fields enclosed by brackets "[]" replaced with your own identifying
|
||||
# information: Portions Copyright [yyyy] [name of copyright owner]
|
||||
#
|
||||
# CDDL HEADER END
|
||||
|
||||
#
|
||||
# Copyright (c) 2019, Datto Inc. All rights reserved.
|
||||
#
|
||||
|
||||
. $STF_SUITE/include/libtest.shlib
|
||||
|
||||
verify_runnable "global"
|
||||
|
||||
set -A VDEV_FILES $TEST_BASE_DIR/file-{1..4}
|
||||
SPARE_VDEV_FILE=$TEST_BASE_DIR/spare-1
|
||||
|
||||
VDEV_FILE_SIZE=$(( $SPA_MINDEVSIZE * 2 ))
|
||||
@@ -0,0 +1,185 @@
|
||||
#!/bin/ksh -p
|
||||
|
||||
#
|
||||
# CDDL HEADER START
|
||||
#
|
||||
# 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.
|
||||
#
|
||||
# CDDL HEADER END
|
||||
#
|
||||
|
||||
#
|
||||
# Copyright (c) 2019, Datto Inc. All rights reserved.
|
||||
#
|
||||
|
||||
. $STF_SUITE/include/libtest.shlib
|
||||
. $STF_SUITE/tests/functional/resilver/resilver.cfg
|
||||
|
||||
#
|
||||
# DESCRIPTION:
|
||||
# Testing resilver restart logic both with and without the deferred resilver
|
||||
# feature enabled, verifying that resilver is not restarted when it is
|
||||
# unecessary.
|
||||
#
|
||||
# STRATEGY:
|
||||
# 1. Create a pool
|
||||
# 2. Create four filesystems with the primary cache disable to force reads
|
||||
# 3. Write four files simultaneously, one to each filesystem
|
||||
# 4. Do with and without deferred resilvers enabled
|
||||
# a. Replace a vdev with a spare & suspend resilver immediately
|
||||
# b. Verify resilver starts properly
|
||||
# c. Offline / online another vdev to introduce a new DTL range
|
||||
# d. Verify resilver restart restart or defer
|
||||
# e. Inject read errors on vdev that was offlined / onlned
|
||||
# f. Verify that resilver did not restart
|
||||
# g. Unsuspend resilver and wait for it to finish
|
||||
# h. Verify that there are two resilvers and nothing is deferred
|
||||
#
|
||||
|
||||
function cleanup
|
||||
{
|
||||
echo $ORIG_RESILVER_MIN_TIME > $ZFS_PARAMS/zfs_resilver_min_time_ms
|
||||
echo $ORIG_SCAN_SUSPEND_PROGRESS > $ZFS_PARAMS/zfs_scan_suspend_progress
|
||||
log_must zinject -c all
|
||||
destroy_pool $TESTPOOL
|
||||
rm -f ${VDEV_FILES[@]} $SPARE_VDEV_FILE
|
||||
}
|
||||
|
||||
# Count resilver events in zpool and number of deferred rsilvers on vdevs
|
||||
function verify_restarts # <msg> <cnt> <defer>
|
||||
{
|
||||
msg=$1
|
||||
cnt=$2
|
||||
defer=$3
|
||||
|
||||
# check the number of resilver start in events log
|
||||
RESILVERS=$(zpool events | grep -c sysevent.fs.zfs.resilver_start)
|
||||
log_note "expected $cnt resilver start(s)$msg, found $RESILVERS"
|
||||
[[ "$RESILVERS" -ne "$cnt" ]] &&
|
||||
log_fail "expected $cnt resilver start(s)$msg, found $RESILVERS"
|
||||
|
||||
[[ -z "$defer" ]] && return
|
||||
|
||||
# use zdb to find which vdevs have the resilver defer flag
|
||||
VDEV_DEFERS=$(zdb -C $TESTPOOL | \
|
||||
sed -n -e '/^ *children\[[0-9]\].*$/{h}' \
|
||||
-e '/ *com.datto:resilver_defer$/{g;p}')
|
||||
|
||||
if [[ "$defer" == "-" ]]
|
||||
then
|
||||
[[ -n $VDEV_DEFERS ]] &&
|
||||
log_fail "didn't expect any vdevs to have resilver deferred"
|
||||
return
|
||||
fi
|
||||
|
||||
[[ "x${VDEV_DEFERS}x" =~ "x +children[$defer]:x" ]] ||
|
||||
log_fail "resilver deferred set on unexpected vdev: $VDEV_DEFERS"
|
||||
}
|
||||
|
||||
log_assert "Check for unnecessary resilver restarts"
|
||||
|
||||
ZFS_PARAMS=/sys/module/zfs/parameters
|
||||
ORIG_RESILVER_MIN_TIME=$(cat $ZFS_PARAMS/zfs_resilver_min_time_ms)
|
||||
ORIG_SCAN_SUSPEND_PROGRESS=$(cat $ZFS_PARAMS/zfs_scan_suspend_progress)
|
||||
|
||||
set -A RESTARTS -- '1' '2' '2' '2'
|
||||
set -A VDEVS -- '' '' '' ''
|
||||
set -A DEFER_RESTARTS -- '1' '1' '1' '2'
|
||||
set -A DEFER_VDEVS -- '-' '2' '2' '-'
|
||||
|
||||
VDEV_REPLACE="${VDEV_FILES[1]} $SPARE_VDEV_FILE"
|
||||
|
||||
log_onexit cleanup
|
||||
|
||||
log_must truncate -s $VDEV_FILE_SIZE ${VDEV_FILES[@]} $SPARE_VDEV_FILE
|
||||
|
||||
log_must zpool create -f -o feature@resilver_defer=disabled $TESTPOOL \
|
||||
raidz ${VDEV_FILES[@]}
|
||||
|
||||
# Create 4 filesystems
|
||||
for fs in fs{0..3}
|
||||
do
|
||||
log_must zfs create -o primarycache=none -o recordsize=1k $TESTPOOL/$fs
|
||||
done
|
||||
|
||||
# simultaneously write 16M to each of them
|
||||
set -A DATAPATHS /$TESTPOOL/fs{0..3}/dat.0
|
||||
log_note "Writing data files"
|
||||
for path in ${DATAPATHS[@]}
|
||||
do
|
||||
dd if=/dev/urandom of=$path bs=1M count=16 > /dev/null 2>&1 &
|
||||
done
|
||||
wait
|
||||
|
||||
# Test without and with deferred resilve feature enabled
|
||||
for test in "without" "with"
|
||||
do
|
||||
log_note "Testing $test deferred resilvers"
|
||||
|
||||
if [[ $test == "with" ]]
|
||||
then
|
||||
log_must zpool set feature@resilver_defer=enabled $TESTPOOL
|
||||
RESTARTS=( "${DEFER_RESTARTS[@]}" )
|
||||
VDEVS=( "${DEFER_VDEVS[@]}" )
|
||||
VDEV_REPLACE="$SPARE_VDEV_FILE ${VDEV_FILES[1]}"
|
||||
fi
|
||||
|
||||
# clear the events
|
||||
log_must zpool events -c
|
||||
|
||||
# limit scanning time
|
||||
echo 50 > $ZFS_PARAMS/zfs_resilver_min_time_ms
|
||||
|
||||
# initiate a resilver and suspend the scan as soon as possible
|
||||
log_must zpool replace $TESTPOOL $VDEV_REPLACE
|
||||
echo 1 > $ZFS_PARAMS/zfs_scan_suspend_progress
|
||||
|
||||
# there should only be 1 resilver start
|
||||
verify_restarts '' "${RESTARTS[0]}" "${VDEVS[0]}"
|
||||
|
||||
# offline then online a vdev to introduce a new DTL range after current
|
||||
# scan, which should restart (or defer) the resilver
|
||||
log_must zpool offline $TESTPOOL ${VDEV_FILES[2]}
|
||||
log_must zpool sync $TESTPOOL
|
||||
log_must zpool online $TESTPOOL ${VDEV_FILES[2]}
|
||||
log_must zpool sync $TESTPOOL
|
||||
|
||||
# there should now be 2 resilver starts w/o defer, 1 with defer
|
||||
verify_restarts ' after offline/online' "${RESTARTS[1]}" "${VDEVS[1]}"
|
||||
|
||||
# inject read io errors on vdev and verify resilver does not restart
|
||||
log_must zinject -a -d ${VDEV_FILES[2]} -e io -T read -f 0.25 $TESTPOOL
|
||||
log_must cat ${DATAPATHS[1]} > /dev/null
|
||||
log_must zinject -c all
|
||||
|
||||
# there should still be 2 resilver starts w/o defer, 1 with defer
|
||||
verify_restarts ' after zinject' "${RESTARTS[2]}" "${VDEVS[2]}"
|
||||
|
||||
# unsuspend resilver
|
||||
echo 0 > $ZFS_PARAMS/zfs_scan_suspend_progress
|
||||
echo 3000 > $ZFS_PARAMS/zfs_resilver_min_time_ms
|
||||
|
||||
# wait for resilver to finish
|
||||
for iter in {0..59}
|
||||
do
|
||||
is_pool_resilvered $TESTPOOL && break
|
||||
sleep 1
|
||||
done
|
||||
is_pool_resilvered $TESTPOOL ||
|
||||
log_fail "resilver timed out"
|
||||
|
||||
# wait for a few txg's to see if a resilver happens
|
||||
log_must zpool sync $TESTPOOL
|
||||
|
||||
# there should now be 2 resilver starts
|
||||
verify_restarts ' after resilver' "${RESTARTS[3]}" "${VDEVS[3]}"
|
||||
done
|
||||
|
||||
log_pass "Resilver did not restart unnecessarily"
|
||||
+31
@@ -0,0 +1,31 @@
|
||||
#!/bin/ksh -p
|
||||
#
|
||||
# CDDL HEADER START
|
||||
#
|
||||
# The contents of this file are subject to the terms of the
|
||||
# Common Development and Distribution License (the "License").
|
||||
# You may not use this file except in compliance with the License.
|
||||
#
|
||||
# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
|
||||
# or http://www.opensolaris.org/os/licensing.
|
||||
# See the License for the specific language governing permissions
|
||||
# and limitations under the License.
|
||||
#
|
||||
# When distributing Covered Code, include this CDDL HEADER in each
|
||||
# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
|
||||
# If applicable, add the following below this CDDL HEADER, with the
|
||||
# fields enclosed by brackets "[]" replaced with your own identifying
|
||||
# information: Portions Copyright [yyyy] [name of copyright owner]
|
||||
#
|
||||
# CDDL HEADER END
|
||||
|
||||
#
|
||||
# Copyright (c) 2019, Datto Inc. All rights reserved.
|
||||
#
|
||||
|
||||
. $STF_SUITE/include/libtest.shlib
|
||||
. $STF_SUITE/tests/functional/resilver/resilver.cfg
|
||||
|
||||
verify_runnable "global"
|
||||
|
||||
log_pass
|
||||
Reference in New Issue
Block a user