diff options
| author | Pranith Kumar K <pkarampu@redhat.com> | 2016-03-17 19:42:00 +0530 |
|---|---|---|
| committer | Pranith Kumar Karampuri <pkarampu@redhat.com> | 2016-04-15 02:51:06 -0700 |
| commit | 8deedef565df49def75083678f8d1558c7b1f7d3 (patch) | |
| tree | ee593ac5d6a88411ba1110e9cdef80270b8b4a43 /xlators/cluster/afr/src/afr-self-heal-entry.c | |
| parent | d5409aae63a7ac5e5b3ea6cfa16c6250a028291c (diff) | |
cluster/afr: Fix partial heals in 3-way replication
Problem:
When there are 2 sources and one sink and if two self-heal daemons
try to acquire locks at the same time, there is a chance that it
gets a lock on one source and sink leading partial to heal. This will
need one more heal from the remaining source to sink for the complete
self-heal. This is not optimal.
Fix:
Upgrade non-blocking locks to blocking lock on all the subvolumes, if
the number of locks acquired is majority and there were eagains.
BUG: 1318751
Change-Id: Iae10b8d3402756c4164b98cc49876056ff7a61e5
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/13766
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Diffstat (limited to 'xlators/cluster/afr/src/afr-self-heal-entry.c')
| -rw-r--r-- | xlators/cluster/afr/src/afr-self-heal-entry.c | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/xlators/cluster/afr/src/afr-self-heal-entry.c b/xlators/cluster/afr/src/afr-self-heal-entry.c index e4d616ea20a..c8e2c98db0e 100644 --- a/xlators/cluster/afr/src/afr-self-heal-entry.c +++ b/xlators/cluster/afr/src/afr-self-heal-entry.c @@ -699,7 +699,7 @@ postop_unlock: out: if (did_sh) afr_log_selfheal (fd->inode->gfid, this, ret, "entry", source, - healed_sinks); + sources, healed_sinks); else ret = 1; @@ -754,8 +754,9 @@ afr_selfheal_entry (call_frame_t *frame, xlator_t *this, inode_t *inode) locked_on = alloca0 (priv->child_count); long_name_locked = alloca0 (priv->child_count); - ret = afr_selfheal_tryentrylk (frame, this, inode, priv->sh_domain, NULL, - locked_on); + ret = afr_selfheal_tie_breaker_entrylk (frame, this, inode, + priv->sh_domain, NULL, + locked_on); { if (ret < AFR_SH_MIN_PARTICIPANTS) { gf_msg_debug (this->name, 0, "%s: Skipping " |