glusterfs.git/xlators, branch v3.3.0qa36

glusterd: Moved 'hooks' interface code to glusterd-hooks.*

2012-04-19T07:36:26+00:00

Change-Id: Ia7c0e37154414bddb05516e11b3cc4e6ae83ff38
BUG: 806996
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.com/3142
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

cluster/afr: Set errno correctly in find_fresh_parents failures

2012-04-19T07:26:21+00:00

Change-Id: I923103a03c1efe38285cd70842ec91b98a3f6bfd
BUG: 765551
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/3185
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

cluster/afr: Handle transient parent-entry xactions in lookup

2012-04-19T07:25:17+00:00

This patch addresses the case when the lookup on an entry is performed
while it is being renamed. The lookup can possibly return 2 different
gfids when lookup on one subvol reached before rename and on other
after rename. In such cases the conflicting entry self-heal is
triggered to resolve the issue, but if there are lot entry transactions
going on the parent directory of the entry then the non-blocking
locks could fail resulting in EIO. To avoid this, lookup queries
locks xlator if there are any parent-entrylk on entry's basename.
If afr finds that there are such locks and gfids are differing then
it chooses the file with latest ctime as the iatt of the entry.
This solution is not foolproof, but it decreases the probability of
hitting the EIO. The correct solution is to take blocking locks on
the parent-entry to find out the correct source. Taking blocking
locks in lookup is not good. One stale entry lock can hang the whole
filesystem. So we chose to go with this for now.

Change-Id: Ibebb6c3074f56f80a96893b6bf5b77941e30d400
BUG: 765551
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/3179
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

cluster/dht: Remove un-used pathinfo code for link files

2012-04-19T06:42:30+00:00

Currently, we do not return LINK: attributes for a file
in pathinfo requests.

Change-Id: If8667bd6af06de3d11c37dd0a09726dfe1d2e330
BUG: 795289
Signed-off-by: shishir gowda 
Reviewed-on: http://review.gluster.com/3175
Tested-by: Gluster Build System 
Reviewed-by: Venky Shankar

nfs/nlm: disable NLM if rpc.statd is not found.

2012-04-19T06:41:35+00:00

Change-Id: I571e1f14cefdfabecd8996b6d0ffc928464d2b51
BUG: 813258
Signed-off-by: Krishna Srinivas 
Reviewed-on: http://review.gluster.com/3174
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

cluster/afr: Build parent loc for expunge

2012-04-19T05:26:46+00:00

Change-Id: Ifd1a4117924a7f9a90eb8e02dd2d655f2156fabd
BUG: 765551
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/3178
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

features/locks: Find parent-entrylk presence in lookup

2012-04-19T05:20:47+00:00

Change-Id: I8fc85676827f26b07bf578ea1152d3e251adb4af
BUG: 765551
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/3177
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

glusterd/rebalance: Start process with xlator option client-pid -3

2012-04-18T16:30:41+00:00

This would prevent updation of xtime by the marker.
Set frame->root->pid while creating sync-task. Removing the command
option, as it is in-effective

Change-Id: Ifaa587df67a72ec0571c7d0043de8048e2be12a8
BUG: 812287
Signed-off-by: shishir gowda 
Reviewed-on: http://review.gluster.com/3180
Tested-by: Gluster Build System 
Reviewed-by: Amar Tumballi 
Reviewed-by: Vijay Bellur

glusterd (remove-brick): fix data-loss issue

2012-04-18T13:25:39+00:00

* when more than one brick (more than one distribute subvolume)
  was removed using single command, rebalance process used to
  get started only on one node, causing data-loss when 'commit'
  was done, because the brick's data was not migrated at all.
  Fixed the logic of validating rebalance required in remove-
  brick case.

Change-Id: I3a90ff9f9593e5215110a3a0974bb97e7271f910
Signed-off-by: Amar Tumballi 
BUG: 812770
Reviewed-on: http://review.gluster.com/3173
Tested-by: Gluster Build System 
Reviewed-by: Jeff Darcy 
Reviewed-by: Shishir Gowda

performance/quick-read: disable reading from cache if unlink has

2012-04-18T07:00:16+00:00

happened after fd was opened.

If dentry being unlinked is the last one, there is a possibility that
later, a new entry with same name being created. After creation of new
file, we shouldn't be returning the contents of this new file, when
application reads using fd opened before unlink.

Change-Id: Ic98aa9cd76d24e8a37ce02a0aae6c66290043192
BUG: 811976
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.com/3151
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur