glusterfs-afrv1.git, branch v3.4.0beta2

performance/io-cache: check the inode context to be NULL before accessing

2013-05-23T20:04:02+00:00

Change-Id: I475af7f8ffd5e5d8adbd2a74af20e56ad7751f69
BUG: 958108
Signed-off-by: Raghavendra Bhat 
Reviewed-on: http://review.gluster.org/4916
Reviewed-by: Amar Tumballi 
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur 
Reviewed-on: http://review.gluster.org/5077
Reviewed-by: Anand Avati

nfs: change default nfs port to 2049

2013-05-23T19:13:53+00:00

This change makes it possible to mount glusterfs volumes
without specifying vers=3 option.

Change-Id: If5a974e2bdfd2adbeac3d82af774310cdf30f988
BUG: 832939
Signed-off-by: Rajesh Amaravathi 
Reviewed-on: http://review.gluster.org/4840
Reviewed-by: Niels de Vos 
Tested-by: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY 
Reviewed-on: http://review.gluster.org/5078
Reviewed-by: Anand Avati

rpc: change dict key for fqdn

2013-05-23T16:54:31+00:00

changed the key from "client.fqdn", which could be wrongly
construed as belonging to protocol/client, to "fqdn".

This is a backport of 8403f9a2d976c33e01fbd9e4a4b04e8f1e936806.

Change-Id: Ib5f4a875a00b99cd903a29da19bafeb70baaab4e
BUG: 950056
Signed-off-by: Rajesh Amaravathi 
Reviewed-on: http://review.gluster.org/4536
Reviewed-by: Niels de Vos 
Reviewed-by: Kaleb KEITHLEY 
Tested-by: Gluster Build System 
Signed-off-by: Rajesh Amaravathi 
Reviewed-on: http://review.gluster.org/4965
Reviewed-by: Jeff Darcy

syncop: synctask shouldn't yawn, it could miss a 'wake'

2013-05-23T14:24:42+00:00

        Backport of http://review.gluster.org/5058

Change-Id: I7731fd33ca0c925cc52f8d105275b44fc625a1e2
BUG: 948686
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/5071
Tested-by: Gluster Build System 
Reviewed-by: Jeff Darcy

syncop: Remove task from syncbarrier's waitq before 'wake'

2013-05-21T13:03:40+00:00

        Backport of http://review.gluster.org/5047

Removing task from syncbarrier's waitq after wake could result in a
subsequent syncbarrier_wake, wake'ing up the already running task.  This
fix makes the removal from waitq and wake 'atomic'

The root cause and the fix are similar in spirit to what was observed
in synclock's waitq implementation.

Change-Id: I7dd9e6ad5945742bcda20eb5a06a9376bb18528e
BUG: 948686
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/5054
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

syncop: Update synctask state appropriately

2013-05-21T13:03:14+00:00

        Backport of http://review.gluster.org/4985

* Earlier, SYNCOP macro, the only consumer of synctask_yield, would set
the task->state to SYNCTASK_SUSPEND. Today, we have glusterd having its
own wrapper macros which don't set task's state. There is also the
syncbarrier and synclock framework, which also participate in a
synctask's scheduling (and need to keep a task's state up to date). It
only makes more sense to leave a synctask's state to the synctask
library, since its an internal affair.

* Need to 'yawn' before 'yield' to avoid re-running tasks to set
  task->woken appropriately.

Change-Id: Ic7a59e6ebcc46f03e53223ca237668d45a3cba40
BUG: 948686
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/5053
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

tests: Modified test to use remove-brick instead of 'start' variant

2013-05-21T11:29:57+00:00

        Backport of change f75be77 from master

remove-brick start doesn't remove the brick from the volume immediately.
It would wait until migration of data to other bricks are complete. Even
when there is no data to be migrated, one can expect a finite delay from
the time of remove-brick start command's exit and removal of brick(s).
This may cause subsequent checks on brick count to fail in a
non-deterministic manner.

Also, renamed the test file name to reflect bug-id corresponding to
community release.

BUG: 878004
Change-Id: Ic6e1360ae5a5280d0d7efe8c3e9a0aa57dddb508
Signed-off-by: Kaushal M 
Reviewed-on: http://review.gluster.org/5052
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

mgmt/glusterd: Enable write-behind in nfs

2013-05-21T08:16:14+00:00

We observed that the number of write requests thus inodelks
are increasing very rapidly to thousands without write-behind
in the graph.

Change-Id: I901a6a820eb7b21b413d33e1a0a3420c7f4746a8
BUG: 928341
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/4736
Reviewed-by: Anand Avati 
Tested-by: Gluster Build System

md-cache: Make options structure NULL terminated

2013-05-21T06:29:41+00:00

Change-Id: I8aa4f90ba7e1eecf3f978be04f8550049275464f
BUG: 765785
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/5028
Reviewed-by: Amar Tumballi 
Tested-by: Gluster Build System

rpcsvc: fix dangerous setting of pointer on free'd structure.

2013-05-21T06:08:12+00:00

Change-Id: I6b666115a8ce155563078d7fc476d41486bab54f
BUG: 884452
Signed-off-by: Vijay Bellur 
Reviewed-on: http://review.gluster.org/4995
Reviewed-by: Krishnan Parthasarathi 
Reviewed-by: Jeff Darcy 
Tested-by: Gluster Build System