+++ This bug was initially created as a clone of Bug #1578823 +++
Description of problem:
commit a9ccd0c8ea6989c72 included EIO in the list of errnos that triggered a migration check task. This is therefore triggered in genuine cases of split brains as well.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
--- Additional comment from Worker Ant on 2018-05-16 08:47:42 EDT ---
REVIEW: https://review.gluster.org/20028 (cluster/dht: Remove EIO from dht_inode_missing) posted (#1) for review on master by N Balachandran
--- Additional comment from Worker Ant on 2018-05-17 02:49:36 EDT ---
COMMIT: https://review.gluster.org/20028 committed in master by "Raghavendra G" <firstname.lastname@example.org> with a commit message- cluster/dht: Remove EIO from dht_inode_missing
Removed EIO from the list of errnos that triggered
a migrate check task.
Signed-off-by: N Balachandran <email@example.com>
Verified this BZ on glusterfs version: 3.12.2-15.el7.x86_64.
1) Created a distributed-replicate volume 4x3 and start it.
2) FUSE mount it on multiple clients.
3) Create a file "f1"
4) set cluster.self-heal-daemon to off and cluster.quorum-type to none
5) Bring down b1,b2 in a replica pair.
6) Write some data to file "f1"
7) Bring back the down bricks. b1 and b2
8) Now, bring down b3 in the replica pair.
9) Write some data to the file "f1"
10) Bring back b3
Now, file will be in split-brain. Append some data to the file, it throws EIO error.
Hence, moving this BZ to Verified.
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.