diff options
author | Ashish Pandey <aspandey@redhat.com> | 2018-12-21 14:31:15 +0530 |
---|---|---|
committer | Amar Tumballi <amarts@redhat.com> | 2019-02-01 05:44:36 +0000 |
commit | 6b98735956c599ea621fa560b201fb7de6c36cac (patch) | |
tree | 0e835797f6688152fb8089118958307773f1d80f /tests/basic/afr | |
parent | 7eda6fe62d4e47937afc2ba71b4fd858399bca05 (diff) |
cluster/thin-arbiter: Consider thin-arbiter before marking new entry changelog
If a fop to create an entry fails on one of the data brick,
we mark the pending changelog on the entry on brick for which
it was successful. This is done as part of post op phase to
make sure that entry gets healed even if it gets renamed to
some other path where its parent was not marked as bad.
As it happens as part of post op, we should consider thin-arbiter
to check if the brick, which was successful, is the good brick or not.
This will avoide split brain and other issues.
Change-Id: I12686675be98f02f70a5186b3ed748c541514d53
updates: bz#1662264
Signed-off-by: Ashish Pandey <aspandey@redhat.com>
Diffstat (limited to 'tests/basic/afr')
-rw-r--r-- | tests/basic/afr/ta.t | 22 |
1 files changed, 16 insertions, 6 deletions
diff --git a/tests/basic/afr/ta.t b/tests/basic/afr/ta.t index d8beb1be461..05d48431c95 100644 --- a/tests/basic/afr/ta.t +++ b/tests/basic/afr/ta.t @@ -27,18 +27,28 @@ EXPECT_WITHIN $PROCESS_DOWN_TIMEOUT "0" afr_child_up_status_meta $M0 $V0-replica TEST touch $M0/b.txt EXPECT "000000000000000000000001" get_hex_xattr trusted.afr.$V0-client-0 $B0/brick1 EXPECT "000000010000000200000000" get_hex_xattr trusted.afr.$V0-client-0 $B0/brick1/b.txt -#TODO: New entry mark lead to pending data on the file but no such marking happened on ta -#EXPECT "000000010000000100000001" get_hex_xattr trusted.afr.$V0-client-0 $B0/ta/trusted.afr.patchy-ta-2 +#New entry mark lead to pending data on the file and on ta +EXPECT "000000010000000100000000" get_hex_xattr trusted.afr.$V0-client-0 $B0/ta/trusted.afr.patchy-ta-2 TEST ! ls $B0/brick0/b.txt TEST ls $B0/brick1/b.txt +#Try to create an entry while good brick is down and bad brick is UP. Should not create +TEST ta_start_brick_process brick0 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_meta $M0 $V0-replicate-0 0 +TEST ta_kill_brick brick1 +EXPECT_WITHIN $PROCESS_DOWN_TIMEOUT "0" afr_child_up_status_meta $M0 $V0-replicate-0 1 +TEST ! touch $M0/d.txt +EXPECT "000000010000000100000000" get_hex_xattr trusted.afr.$V0-client-0 $B0/ta/trusted.afr.patchy-ta-2 + +TEST ta_start_brick_process brick1 +EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_meta $M0 $V0-replicate-0 1 +TEST ta_kill_brick brick0 +EXPECT_WITHIN $PROCESS_DOWN_TIMEOUT "0" afr_child_up_status_meta $M0 $V0-replicate-0 0 + TEST ta_kill_brick ta -# Entry create must fail since only 1 brick is up. +# Entry create must fail if only one brick is UP, even if that is a good brick. TEST ! touch $M0/c.txt TEST ! ls $B0/brick0/c.txt TEST ! ls $B0/brick1/c.txt -TEST ta_start_brick_process brick0 -EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_meta $M0 $V0-replicate-0 0 - cleanup; |