REVIEW: http://review.gluster.org/16055 (cluster/tier: Check for phase2 migration on EINVAL error) posted (#1) for review on master by Dan Lambright (firstname.lastname@example.org)
REVIEW: http://review.gluster.org/16163 (cluster/tier: Don't update cached subvol) posted (#1) for review on master by N Balachandran (email@example.com)
COMMIT: https://review.gluster.org/16163 committed in master by Kaleb KEITHLEY (firstname.lastname@example.org)
Author: N Balachandran <email@example.com>
Date: Fri Dec 16 18:55:28 2016 +0530
cluster/tier: Don't update cached subvol
tier_readdirp_cbk updates the cached subvol to
the hot tier if it finds a linkto file.
However, if no lookup has been sent to the hot tier,
lower layers will not have updated the inode-ctx causing
later fops to fail.
Signed-off-by: N Balachandran <firstname.lastname@example.org>
Smoke: Gluster Build System <email@example.com>
Reviewed-by: Dan Lambright <firstname.lastname@example.org>
Tested-by: Dan Lambright <email@example.com>
NetBSD-regression: NetBSD Build System <firstname.lastname@example.org>
CentOS-regression: Gluster Build System <email@example.com>
Reviewed-by: Kaleb KEITHLEY <firstname.lastname@example.org>
I'm reopening this as this will need another patch as well. That patch is still WIP.
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.12.0, please open a new bug report.
glusterfs-3.12.0 has been announced on the Gluster mailinglists , packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist  and the update infrastructure for your distribution.