From 657d78dbad118e511e1fca8b1badb9f8ae7a6f60 Mon Sep 17 00:00:00 2001 From: karthik-us Date: Wed, 7 Jun 2017 15:56:13 +0530 Subject: cluster/afr: GFID split-brain resolution with existing CLI Problem: Currently there is no way for the admin from CLI to resolve gfid split-brain based on some policy like choice of the brick, mtime or size. Fix: With the existing CLI options based on size, mtime, and choice of brick, we do lookup on the parent for the specified file. As part of the lookup, if we find gfid mismatch, we resolve them based on the policy and return. If the file is not in gfid split- brain, then we check for the data and metadata split-brain in the getxattr code path, and resolve if any. This will work provided absolute path to the file with the CLI and not with gfid of the file. Hence the source-brick policy without any file path will also not resolve the gfid split-brain since it uses the gfid of the files. But it can resolve any other type of split-brains and skip the gfid mismatch resolution with the usual error message. Reverting the change https://review.gluster.org/17290. This patch resolves the issue. Fixes gluster/glusterfs#135 Change-Id: Iaeba6fc32f184a34255d03be87cda02773130a09 BUG: 1459530 Signed-off-by: karthik-us Reviewed-on: https://review.gluster.org/17485 Reviewed-by: Ravishankar N Reviewed-by: Pranith Kumar Karampuri CentOS-regression: Gluster Build System Smoke: Gluster Build System --- .../basic/afr/gfid-mismatch-resolution-with-cli.t | 168 +++++++++++++++++++++ ...fid-mismatch-resolution-with-fav-child-policy.t | 3 - 2 files changed, 168 insertions(+), 3 deletions(-) create mode 100644 tests/basic/afr/gfid-mismatch-resolution-with-cli.t (limited to 'tests') diff --git a/tests/basic/afr/gfid-mismatch-resolution-with-cli.t b/tests/basic/afr/gfid-mismatch-resolution-with-cli.t new file mode 100644 index 00000000000..b739ddc49cc --- /dev/null +++ b/tests/basic/afr/gfid-mismatch-resolution-with-cli.t @@ -0,0 +1,168 @@ +#!/bin/bash + +. $(dirname $0)/../../include.rc +. $(dirname $0)/../../volume.rc + +cleanup; + +TEST glusterd +TEST pidof glusterd +TEST $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1} +TEST $CLI volume start $V0 +TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0 +TEST $CLI volume set $V0 self-heal-daemon off +TEST $CLI volume set $V0 cluster.entry-self-heal off +TEST $CLI volume set $V0 cluster.metadata-self-heal off +TEST $CLI volume set $V0 cluster.data-self-heal off +cd $M0 + +##### Healing from latest mtime ###### + +TEST kill_brick $V0 $H0 $B0/${V0}0 +echo "Sink based on mtime" > f1 +TEST $CLI volume start $V0 force +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 0 +TEST kill_brick $V0 $H0 $B0/${V0}1 +echo "Source based on mtime" > f1 + +gfid_0=$(gf_get_gfid_xattr $B0/${V0}0/f1) +gfid_1=$(gf_get_gfid_xattr $B0/${V0}1/f1) +TEST [ "$gfid_0" != "$gfid_1" ] + +TEST $CLI volume start $V0 force +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 1 + +#We know that first brick has the latest mtime +LATEST_MTIME_MD5=$(md5sum $B0/${V0}0/f1 | awk '{print $1}') + +TEST $CLI volume heal $V0 split-brain latest-mtime /f1 + +#gfid split-brain should be resolved +gfid_1=$(gf_get_gfid_xattr $B0/${V0}1/f1) +TEST [ "$gfid_0" == "$gfid_1" ] + +#Heal the data and check the md5sum +TEST $CLI volume set $V0 self-heal-daemon on +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Y" glustershd_up_status +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 1 +TEST $CLI volume heal $V0 +EXPECT_WITHIN $HEAL_TIMEOUT "^0$" get_pending_heal_count $V0 +HEALED_MD5=$(md5sum $B0/${V0}1/f1 | awk '{print $1}') +TEST [ "$LATEST_MTIME_MD5" == "$HEALED_MD5" ] + + +##### Healing from bigger file ###### + +TEST mkdir test +TEST $CLI volume set $V0 self-heal-daemon off +TEST kill_brick $V0 $H0 $B0/${V0}0 +echo "Bigger file" > test/f2 +TEST $CLI volume start $V0 force +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 0 +TEST kill_brick $V0 $H0 $B0/${V0}1 +echo "Small file" > test/f2 + +gfid_0=$(gf_get_gfid_xattr $B0/${V0}0/test/f2) +gfid_1=$(gf_get_gfid_xattr $B0/${V0}1/test/f2) +TEST [ "$gfid_0" != "$gfid_1" ] + +TEST $CLI volume start $V0 force +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 1 + +#We know that second brick has the bigger file +BIGGER_FILE_MD5=$(md5sum $B0/${V0}1/test/f2 | awk '{print $1}') + +TEST $CLI volume heal $V0 split-brain bigger-file /test/f2 + +#gfid split-brain should be resolved +gfid_0=$(gf_get_gfid_xattr $B0/${V0}0/test/f2) +TEST [ "$gfid_0" == "$gfid_1" ] + +#Heal the data and check the md5sum +TEST $CLI volume set $V0 self-heal-daemon on +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Y" glustershd_up_status +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 1 +TEST $CLI volume heal $V0 +EXPECT_WITHIN $HEAL_TIMEOUT "^0$" get_pending_heal_count $V0 +HEALED_MD5=$(md5sum $B0/${V0}0/test/f2 | awk '{print $1}') +TEST [ "$BIGGER_FILE_MD5" == "$HEALED_MD5" ] + + +#Add one more brick, and heal. +TEST $CLI volume add-brick $V0 replica 3 $H0:$B0/${V0}2 +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Y" glustershd_up_status +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0 +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1 +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}2 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 2 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 2 + +TEST $CLI volume heal $V0 +EXPECT_WITHIN $HEAL_TIMEOUT "^0$" get_pending_heal_count $V0 + + +##### Healing from source brick ###### + +TEST $CLI volume set $V0 self-heal-daemon off +TEST $CLI volume set $V0 cluster.quorum-type none +TEST kill_brick $V0 $H0 $B0/${V0}0 +echo "We will consider these as sinks" > test/f3 +TEST $CLI volume start $V0 force +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 0 +TEST kill_brick $V0 $H0 $B0/${V0}1 +TEST kill_brick $V0 $H0 $B0/${V0}2 +echo "We will take this as source" > test/f3 + +gfid_0=$(gf_get_gfid_xattr $B0/${V0}0/test/f3) +gfid_1=$(gf_get_gfid_xattr $B0/${V0}1/test/f3) +gfid_2=$(gf_get_gfid_xattr $B0/${V0}2/test/f3) +TEST [ "$gfid_0" != "$gfid_1" ] +TEST [ "$gfid_1" == "$gfid_2" ] + +TEST $CLI volume start $V0 force +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1 +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}2 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status $V0 2 + +#We will try to heal the split-brain with bigger file option. +#It should fail, since we have same file size in bricks 1 & 2. +EXPECT "No bigger file for file /test/f3" $CLI volume heal $V0 split-brain bigger-file /test/f3 + +#Now heal from taking the brick 0 as the source +SOURCE_MD5=$(md5sum $B0/${V0}0/test/f3 | awk '{print $1}') + +TEST $CLI volume heal $V0 split-brain source-brick $H0:$B0/${V0}0 /test/f3 + +#gfid split-brain should be resolved +gfid_1=$(gf_get_gfid_xattr $B0/${V0}1/test/f3) +gfid_2=$(gf_get_gfid_xattr $B0/${V0}2/test/f3) +TEST [ "$gfid_0" == "$gfid_1" ] +TEST [ "$gfid_0" == "$gfid_2" ] + +#Heal the data and check the md5sum +TEST $CLI volume set $V0 self-heal-daemon on +EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Y" glustershd_up_status +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 2 +TEST $CLI volume heal $V0 +EXPECT_WITHIN $HEAL_TIMEOUT "^0$" get_pending_heal_count $V0 +HEALED_MD5_1=$(md5sum $B0/${V0}1/test/f3 | awk '{print $1}') +HEALED_MD5_2=$(md5sum $B0/${V0}2/test/f3 | awk '{print $1}') +TEST [ "$SOURCE_MD5" == "$HEALED_MD5_1" ] +TEST [ "$SOURCE_MD5" == "$HEALED_MD5_2" ] + +cd - +cleanup; diff --git a/tests/basic/afr/gfid-mismatch-resolution-with-fav-child-policy.t b/tests/basic/afr/gfid-mismatch-resolution-with-fav-child-policy.t index 9f1347bbb44..2f14f838e49 100644 --- a/tests/basic/afr/gfid-mismatch-resolution-with-fav-child-policy.t +++ b/tests/basic/afr/gfid-mismatch-resolution-with-fav-child-policy.t @@ -226,6 +226,3 @@ HEALED_MD5=$(md5sum $B0/${V0}2/f4 | cut -d\ -f1) TEST [ "$MAJORITY_MD5" == "$HEALED_MD5" ] cleanup; - -#G_TESTDEF_TEST_STATUS_CENTOS6=BAD_TEST,BUG=1450730 -#G_TESTDEF_TEST_STATUS_NETBSD7=BAD_TEST,BUG=1450730 -- cgit