ceph/qa/standalone/erasure-code/test-erasure-eio.sh
cuixf 3eb1679b1f osd: retry to read object attrs at EC recovery
In EC recovery read, if the object's attrs read failed or with errors, we erase the attrs we have read and
try to read it again from left shards. This will make the primary osd get the object's attrs correct and
avoid assert.

Signed-off-by: xiaofei cui <cuixiaofei@sangfor.com>
2018-06-01 06:26:56 -04:00

672 lines
18 KiB
Bash
Executable File

#!/usr/bin/env bash
#
# Copyright (C) 2015 Red Hat <contact@redhat.com>
#
#
# Author: Kefu Chai <kchai@redhat.com>
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU Library Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Library Public License for more details.
#
source $CEPH_ROOT/qa/standalone/ceph-helpers.sh
function run() {
local dir=$1
shift
export CEPH_MON="127.0.0.1:7112" # git grep '\<7112\>' : there must be only one
export CEPH_ARGS
CEPH_ARGS+="--fsid=$(uuidgen) --auth-supported=none "
CEPH_ARGS+="--mon-host=$CEPH_MON "
local funcs=${@:-$(set | sed -n -e 's/^\(TEST_[0-9a-z_]*\) .*/\1/p')}
for func in $funcs ; do
setup $dir || return 1
run_mon $dir a || return 1
run_mgr $dir x || return 1
create_rbd_pool || return 1
# check that erasure code plugins are preloaded
CEPH_ARGS='' ceph --admin-daemon $(get_asok_path mon.a) log flush || return 1
grep 'load: jerasure.*lrc' $dir/mon.a.log || return 1
$func $dir || return 1
teardown $dir || return 1
done
}
function setup_osds() {
local count=$1
shift
for id in $(seq 0 $(expr $count - 1)) ; do
run_osd $dir $id || return 1
done
wait_for_clean || return 1
# check that erasure code plugins are preloaded
CEPH_ARGS='' ceph --admin-daemon $(get_asok_path osd.0) log flush || return 1
grep 'load: jerasure.*lrc' $dir/osd.0.log || return 1
}
function get_state() {
local pgid=$1
local sname=state
ceph --format json pg dump pgs 2>/dev/null | \
jq -r ".[] | select(.pgid==\"$pgid\") | .$sname"
}
function create_erasure_coded_pool() {
local poolname=$1
shift
local k=$1
shift
local m=$1
shift
ceph osd erasure-code-profile set myprofile \
plugin=jerasure \
k=$k m=$m \
crush-failure-domain=osd || return 1
create_pool $poolname 1 1 erasure myprofile \
|| return 1
wait_for_clean || return 1
}
function delete_erasure_coded_pool() {
local poolname=$1
ceph osd pool delete $poolname $poolname --yes-i-really-really-mean-it
ceph osd erasure-code-profile rm myprofile
}
function rados_put() {
local dir=$1
local poolname=$2
local objname=${3:-SOMETHING}
for marker in AAA BBB CCCC DDDD ; do
printf "%*s" 1024 $marker
done > $dir/ORIGINAL
#
# get and put an object, compare they are equal
#
rados --pool $poolname put $objname $dir/ORIGINAL || return 1
}
function rados_get() {
local dir=$1
local poolname=$2
local objname=${3:-SOMETHING}
local expect=${4:-ok}
#
# Expect a failure to get object
#
if [ $expect = "fail" ];
then
! rados --pool $poolname get $objname $dir/COPY
return
fi
#
# get an object, compare with $dir/ORIGINAL
#
rados --pool $poolname get $objname $dir/COPY || return 1
diff $dir/ORIGINAL $dir/COPY || return 1
rm $dir/COPY
}
function inject_remove() {
local pooltype=$1
shift
local which=$1
shift
local poolname=$1
shift
local objname=$1
shift
local dir=$1
shift
local shard_id=$1
shift
local -a initial_osds=($(get_osds $poolname $objname))
local osd_id=${initial_osds[$shard_id]}
objectstore_tool $dir $osd_id $objname remove || return 1
}
# Test with an inject error
function rados_put_get_data() {
local inject=$1
shift
local dir=$1
shift
local shard_id=$1
shift
local arg=$1
# inject eio to speificied shard
#
local poolname=pool-jerasure
local objname=obj-$inject-$$-$shard_id
rados_put $dir $poolname $objname || return 1
inject_$inject ec data $poolname $objname $dir $shard_id || return 1
rados_get $dir $poolname $objname || return 1
if [ "$arg" = "recovery" ];
then
#
# take out the last OSD used to store the object,
# bring it back, and check for clean PGs which means
# recovery didn't crash the primary.
#
local -a initial_osds=($(get_osds $poolname $objname))
local last_osd=${initial_osds[-1]}
# Kill OSD
kill_daemons $dir TERM osd.${last_osd} >&2 < /dev/null || return 1
ceph osd out ${last_osd} || return 1
! get_osds $poolname $objname | grep '\<'${last_osd}'\>' || return 1
ceph osd in ${last_osd} || return 1
run_osd $dir ${last_osd} || return 1
wait_for_clean || return 1
fi
shard_id=$(expr $shard_id + 1)
inject_$inject ec data $poolname $objname $dir $shard_id || return 1
# Now 2 out of 3 shards get an error, so should fail
rados_get $dir $poolname $objname fail || return 1
rm $dir/ORIGINAL
}
# Change the size of speificied shard
#
function set_size() {
local objname=$1
shift
local dir=$1
shift
local shard_id=$1
shift
local bytes=$1
shift
local mode=${1}
local poolname=pool-jerasure
local -a initial_osds=($(get_osds $poolname $objname))
local osd_id=${initial_osds[$shard_id]}
ceph osd set noout
if [ "$mode" = "add" ];
then
objectstore_tool $dir $osd_id $objname get-bytes $dir/CORRUPT || return 1
dd if=/dev/urandom bs=$bytes count=1 >> $dir/CORRUPT
elif [ "$bytes" = "0" ];
then
touch $dir/CORRUPT
else
dd if=/dev/urandom bs=$bytes count=1 of=$dir/CORRUPT
fi
objectstore_tool $dir $osd_id $objname set-bytes $dir/CORRUPT || return 1
rm -f $dir/CORRUPT
ceph osd unset noout
}
function rados_get_data_bad_size() {
local dir=$1
shift
local shard_id=$1
shift
local bytes=$1
shift
local mode=${1:-set}
local poolname=pool-jerasure
local objname=obj-size-$$-$shard_id-$bytes
rados_put $dir $poolname $objname || return 1
# Change the size of speificied shard
#
set_size $objname $dir $shard_id $bytes $mode || return 1
rados_get $dir $poolname $objname || return 1
# Leave objname and modify another shard
shard_id=$(expr $shard_id + 1)
set_size $objname $dir $shard_id $bytes $mode || return 1
rados_get $dir $poolname $objname fail || return 1
rm $dir/ORIGINAL
}
#
# These two test cases try to validate the following behavior:
# For object on EC pool, if there is one shard having read error (
# either primary or replica), client can still read object.
#
# If 2 shards have read errors the client will get an error.
#
function TEST_rados_get_subread_eio_shard_0() {
local dir=$1
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# inject eio on primary OSD (0) and replica OSD (1)
local shard_id=0
rados_put_get_data eio $dir $shard_id || return 1
delete_erasure_coded_pool $poolname
}
function TEST_rados_get_subread_eio_shard_1() {
local dir=$1
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# inject eio into replicas OSD (1) and OSD (2)
local shard_id=1
rados_put_get_data eio $dir $shard_id || return 1
delete_erasure_coded_pool $poolname
}
# We don't remove the object from the primary because
# that just causes it to appear to be missing
function TEST_rados_get_subread_missing() {
local dir=$1
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# inject remove into replicas OSD (1) and OSD (2)
local shard_id=1
rados_put_get_data remove $dir $shard_id || return 1
delete_erasure_coded_pool $poolname
}
#
#
# These two test cases try to validate that following behavior:
# For object on EC pool, if there is one shard which an incorrect
# size this will cause an internal read error, client can still read object.
#
# If 2 shards have incorrect size the client will get an error.
#
function TEST_rados_get_bad_size_shard_0() {
local dir=$1
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# Set incorrect size into primary OSD (0) and replica OSD (1)
local shard_id=0
rados_get_data_bad_size $dir $shard_id 10 || return 1
rados_get_data_bad_size $dir $shard_id 0 || return 1
rados_get_data_bad_size $dir $shard_id 256 add || return 1
delete_erasure_coded_pool $poolname
}
function TEST_rados_get_bad_size_shard_1() {
local dir=$1
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# Set incorrect size into replicas OSD (1) and OSD (2)
local shard_id=1
rados_get_data_bad_size $dir $shard_id 10 || return 1
rados_get_data_bad_size $dir $shard_id 0 || return 1
rados_get_data_bad_size $dir $shard_id 256 add || return 1
delete_erasure_coded_pool $poolname
}
function TEST_rados_get_with_subreadall_eio_shard_0() {
local dir=$1
local shard_id=0
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# inject eio on primary OSD (0)
rados_put_get_data eio $dir $shard_id recovery || return 1
delete_erasure_coded_pool $poolname
}
function TEST_rados_get_with_subreadall_eio_shard_1() {
local dir=$1
local shard_id=1
setup_osds 4 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 2 1 || return 1
# inject eio on replica OSD (1)
rados_put_get_data eio $dir $shard_id recovery || return 1
delete_erasure_coded_pool $poolname
}
# Test recovery the object attr read error
function TEST_ec_object_attr_read_error() {
local dir=$1
local objname=myobject
setup_osds 7 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 3 2 || return 1
local primary_osd=$(get_primary $poolname $objname)
# Kill primary OSD
kill_daemons $dir TERM osd.${primary_osd} >&2 < /dev/null || return 1
# Write data
rados_put $dir $poolname $objname || return 1
# Inject eio, shard 1 is the one read attr
inject_eio ec mdata $poolname $objname $dir 1 || return 1
# Restart OSD
run_osd $dir ${primary_osd} || return 1
# Cluster should recover this object
wait_for_clean || return 1
rados_get $dir $poolname myobject || return 1
delete_erasure_coded_pool $poolname
}
# Test recovery the first k copies aren't all available
function TEST_ec_single_recovery_error() {
local dir=$1
local objname=myobject
setup_osds 7 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 3 2 || return 1
rados_put $dir $poolname $objname || return 1
inject_eio ec data $poolname $objname $dir 0 || return 1
local -a initial_osds=($(get_osds $poolname $objname))
local last_osd=${initial_osds[-1]}
# Kill OSD
kill_daemons $dir TERM osd.${last_osd} >&2 < /dev/null || return 1
ceph osd down ${last_osd} || return 1
ceph osd out ${last_osd} || return 1
# Cluster should recover this object
wait_for_clean || return 1
rados_get $dir $poolname myobject || return 1
delete_erasure_coded_pool $poolname
}
# Test recovery when repeated reads are needed due to EIO
function TEST_ec_recovery_multiple_errors() {
local dir=$1
local objname=myobject
setup_osds 9 || return 1
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 4 4 || return 1
rados_put $dir $poolname $objname || return 1
inject_eio ec data $poolname $objname $dir 0 || return 1
# first read will try shards 0,1,2 when 0 gets EIO, shard 3 gets
# tried as well. Make that fail to test multiple-EIO handling.
inject_eio ec data $poolname $objname $dir 3 || return 1
inject_eio ec data $poolname $objname $dir 4 || return 1
local -a initial_osds=($(get_osds $poolname $objname))
local last_osd=${initial_osds[-1]}
# Kill OSD
kill_daemons $dir TERM osd.${last_osd} >&2 < /dev/null || return 1
ceph osd down ${last_osd} || return 1
ceph osd out ${last_osd} || return 1
# Cluster should recover this object
wait_for_clean || return 1
rados_get $dir $poolname myobject || return 1
delete_erasure_coded_pool $poolname
}
# Test recovery when there's only one shard to recover, but multiple
# objects recovering in one RecoveryOp
function TEST_ec_recovery_multiple_objects() {
local dir=$1
local objname=myobject
ORIG_ARGS=$CEPH_ARGS
CEPH_ARGS+=' --osd-recovery-max-single-start 3 --osd-recovery-max-active 3 '
setup_osds 7 || return 1
CEPH_ARGS=$ORIG_ARGS
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 3 2 || return 1
rados_put $dir $poolname test1
rados_put $dir $poolname test2
rados_put $dir $poolname test3
ceph osd out 0 || return 1
# Cluster should recover these objects all at once
wait_for_clean || return 1
rados_get $dir $poolname test1
rados_get $dir $poolname test2
rados_get $dir $poolname test3
delete_erasure_coded_pool $poolname
}
# test multi-object recovery when the one missing shard gets EIO
function TEST_ec_recovery_multiple_objects_eio() {
local dir=$1
local objname=myobject
ORIG_ARGS=$CEPH_ARGS
CEPH_ARGS+=' --osd-recovery-max-single-start 3 --osd-recovery-max-active 3 '
setup_osds 7 || return 1
CEPH_ARGS=$ORIG_ARGS
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 3 2 || return 1
rados_put $dir $poolname test1
rados_put $dir $poolname test2
rados_put $dir $poolname test3
# can't read from this shard anymore
inject_eio ec data $poolname $objname $dir 0 || return 1
ceph osd out 0 || return 1
# Cluster should recover these objects all at once
wait_for_clean || return 1
rados_get $dir $poolname test1
rados_get $dir $poolname test2
rados_get $dir $poolname test3
delete_erasure_coded_pool $poolname
}
# Test backfill with unfound object
function TEST_ec_backfill_unfound() {
local dir=$1
local objname=myobject
local lastobj=300
# Must be between 1 and $lastobj
local testobj=obj250
ORIG_ARGS=$CEPH_ARGS
CEPH_ARGS+=' --osd_min_pg_log_entries=5 --osd_max_pg_log_entries=10'
setup_osds 5 || return 1
CEPH_ARGS=$ORIG_ARGS
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 3 2 || return 1
ceph pg dump pgs
rados_put $dir $poolname $objname || return 1
local -a initial_osds=($(get_osds $poolname $objname))
local last_osd=${initial_osds[-1]}
kill_daemons $dir TERM osd.${last_osd} 2>&2 < /dev/null || return 1
ceph osd down ${last_osd} || return 1
ceph osd out ${last_osd} || return 1
ceph pg dump pgs
dd if=/dev/urandom of=${dir}/ORIGINAL bs=1024 count=4
for i in $(seq 1 $lastobj)
do
rados --pool $poolname put obj${i} $dir/ORIGINAL || return 1
done
inject_eio ec data $poolname $testobj $dir 0 || return 1
inject_eio ec data $poolname $testobj $dir 1 || return 1
run_osd $dir ${last_osd} || return 1
ceph osd in ${last_osd} || return 1
sleep 15
for tmp in $(seq 1 100); do
state=$(get_state 2.0)
echo $state | grep backfill_unfound
if [ "$?" = "0" ]; then
break
fi
echo $state
sleep 1
done
ceph pg dump pgs
ceph pg 2.0 list_missing | grep -q $testobj || return 1
# Command should hang because object is unfound
timeout 5 rados -p $poolname get $testobj $dir/CHECK
test $? = "124" || return 1
ceph pg 2.0 mark_unfound_lost delete
wait_for_clean || return 1
for i in $(seq 1 $lastobj)
do
if [ obj${i} = "$testobj" ]; then
# Doesn't exist anymore
! rados -p $poolname get $testobj $dir/CHECK || return 1
else
rados --pool $poolname get obj${i} $dir/CHECK || return 1
diff -q $dir/ORIGINAL $dir/CHECK || return 1
fi
done
rm -f ${dir}/ORIGINAL ${dir}/CHECK
delete_erasure_coded_pool $poolname
}
# Test recovery with unfound object
function TEST_ec_recovery_unfound() {
local dir=$1
local objname=myobject
local lastobj=100
# Must be between 1 and $lastobj
local testobj=obj75
ORIG_ARGS=$CEPH_ARGS
CEPH_ARGS+=' --osd-recovery-max-single-start 3 --osd-recovery-max-active 3 '
CEPH_ARGS+=' --osd_min_pg_log_entries=5 --osd_max_pg_log_entries=10'
setup_osds 5 || return 1
CEPH_ARGS=$ORIG_ARGS
local poolname=pool-jerasure
create_erasure_coded_pool $poolname 3 2 || return 1
ceph pg dump pgs
rados_put $dir $poolname $objname || return 1
local -a initial_osds=($(get_osds $poolname $objname))
local last_osd=${initial_osds[-1]}
kill_daemons $dir TERM osd.${last_osd} 2>&2 < /dev/null || return 1
ceph osd down ${last_osd} || return 1
ceph osd out ${last_osd} || return 1
ceph pg dump pgs
dd if=/dev/urandom of=${dir}/ORIGINAL bs=1024 count=4
for i in $(seq 1 $lastobj)
do
rados --pool $poolname put obj${i} $dir/ORIGINAL || return 1
done
inject_eio ec data $poolname $testobj $dir 0 || return 1
inject_eio ec data $poolname $testobj $dir 1 || return 1
run_osd $dir ${last_osd} || return 1
ceph osd in ${last_osd} || return 1
sleep 15
for tmp in $(seq 1 100); do
state=$(get_state 2.0)
echo $state | grep recovery_unfound
if [ "$?" = "0" ]; then
break
fi
echo "$state "
sleep 1
done
ceph pg dump pgs
ceph pg 2.0 list_missing | grep -q $testobj || return 1
# Command should hang because object is unfound
timeout 5 rados -p $poolname get $testobj $dir/CHECK
test $? = "124" || return 1
ceph pg 2.0 mark_unfound_lost delete
wait_for_clean || return 1
for i in $(seq 1 $lastobj)
do
if [ obj${i} = "$testobj" ]; then
# Doesn't exist anymore
! rados -p $poolname get $testobj $dir/CHECK || return 1
else
rados --pool $poolname get obj${i} $dir/CHECK || return 1
diff -q $dir/ORIGINAL $dir/CHECK || return 1
fi
done
rm -f ${dir}/ORIGINAL ${dir}/CHECK
delete_erasure_coded_pool $poolname
}
main test-erasure-eio "$@"
# Local Variables:
# compile-command: "cd ../.. ; make -j4 && test/erasure-code/test-erasure-eio.sh"
# End: