glusterfs.git, branch round-robin2

io-threads: distribute work fairly among clients

2016-07-25T12:58:49+00:00

This is the full "queue of queues" approach where each client gets its
own queue (per priority) and we round-robin among them.

Change-Id: I73955d1b9bb93f2ff781b48dfe2509009c519ec6
Signed-off-by: Jeff Darcy

io-threads: remove least-rate-limit option and code

2016-07-25T12:58:31+00:00

This will be unnecessary, and mostly in the way, as real fairness
guarantees are implemented.

Change-Id: Ic61ec1c9e9add58385f1a4eafcfe2cc554ceefc8
Signed-off-by: Jeff Darcy

gfapi/upcall: Fix a ref leak

2016-07-24T09:55:28+00:00

inode_find (used to create the handle) takes a reference
of the inode. This needs to be un'refernced to avoid leak.

Change-Id: I22f03577a8f1d9608cfc62d57202cfc4c2ba12b3
BUG: 1358608
Signed-off-by: Soumya Koduri 
Reviewed-on: http://review.gluster.org/14984
Reviewed-by: jiffin tony Thottan 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy 
NetBSD-regression: NetBSD Build System 
Tested-by: Gluster Build System

tests: Remove hard coding in get_aux

2016-07-22T23:23:28+00:00

Change-Id: Ie007d8006a2f2be0187f0c73d46ec6dda2a68a6b
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/14988
Smoke: Gluster Build System 
Tested-by: Jeff Darcy 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Jeff Darcy

tests: Fix spurious failures with split-brain-favorite-child-policy.t

2016-07-22T21:32:41+00:00

Problem:
It is not guranteed that the self-heal daemon would apply the new option
as soon as volume set is executed because all the command gurantees is that
the process is notified of the change in volfile. Shd still needs to fetch
volfile and reconfigure. If the next volume heal command comes even before
the reconfigure happens, then the heal won't happen.

Fix:
Restart shd to make sure it has the option loaded with new value.

BUG: 1358976
Change-Id: I3ed30ebbec17bd06caa632e79e9412564f431b19
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/14978
Smoke: Gluster Build System 
Reviewed-by: Krutika Dhananjay 
Tested-by: Jeff Darcy 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy

tests: Fix pending-heal-count checks

2016-07-22T17:01:45+00:00

EXPECT_WITHIN takes regular expression to match the count,
so even when there are say 10 entries to heal, it would
think that the heal is complete. Fixed checking
pending heal count with correct regex.

Thanks to Xavi for finding this problem.

Change-Id: Ic593d22468b2b586bfca864962ffa0eda96b1d1f
BUG: 1332054
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/14985
Smoke: Gluster Build System 
Reviewed-by: Xavier Hernandez 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System

tests: Fix timing issue in ec.t

2016-07-22T15:13:56+00:00

Problem:
Because of timing issue sometimes the mount is unmounted
even before the version is updated, this is leading to
not triggering heals.

Fix:
One way to fix this would be to increate 'sleep 2' to 'sleep 10'
but that would slow things down. I changed the way ec learns
it needs xattr healing so that it triggers heals even when the
xattrs are not marked correctly.

Change-Id: I1c82041166443ae7079dd99b89ea2ed170233ba3
BUG: 1359001
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/14980
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy

glfs/upcall: entries should be removed under mutex lock

2016-07-22T15:13:29+00:00

During poll, upcall entries should be removed from the
upcall_list only under upcall_list_mutex lock. Otherwise
it could result in the list corruption if there are entries
being added during poll resulting in memory leak.

Also addressed a probable leak during any failures with upcall
entry addition.

Change-Id: I468183f961eb6faed9a0a1bcb783705f711641fc
BUG: 1358608
Signed-off-by: Soumya Koduri 
Reviewed-on: http://review.gluster.org/14972
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Shyamsundar Ranganathan 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy

changelog/rpc: Fix rpc_clnt_t mem leaks

2016-07-22T15:12:52+00:00

PROBLEM:
   1. Freeing up rpc_clnt object might lead to crashes. Well,
      it was not a necessity to free rpc-clnt object till now
      because all the existing use cases needs to reconnect
      back on disconnects. Hence timer code was not taking
      ref on rpc-clnt object.

      Glusterd had some use-cases that led to crash due to
      ping-timer and they fixed only those code paths that
      involve ping-timer.

      Now, since changelog has an use-case where rpc-clnt
      need to be freed up, we need to fix timer code to take
      refs

   2. In changelog, because of issue 1, only mydata was being
      freed which is incorrect. And there are races where
      rpc-clnt object would access the freed mydata which
      would lead to crashes.

      Since changelog xlator resides on brick side and is long
      living process, if multiple libgfchangelog consumers
      register to changelog and disconnect/reconnect mulitple
      times, it would result in leak of 'rpc-clnt' object
      for every connect/disconnect.

SOLUTION:
   1. Handle ref/unref of 'rpc_clnt' structure in timer
      functions properly.
   2. In changelog, unref 'rpc_clnt' in RPC_CLNT_DISCONNECT
      after disabling timers and free mydata on RPC_CLNT_DESTROY.

RPC SETUP IN CHANGELOG:
   1. changelog xlator initiates rpc server say 'changelog_rpc_server'
   2. libgfchangelog initiates one rpc server say 'libgfchangelog_rpc_server'
   3. libgfchangelog initiates rpc client and connects to 'changelog_rpc_server'
   4. In return changelog_rpc_server initiates a rpc client and connects back
      to 'libgfchangelog_rpc_server'

REF/UNREF HANDLING IN TIMER FUNCTIONS:
Let's say rpc clnt refcount = 1
   1. Take the ref before reigstering callback to timer queue
           >>>>  rpc_clnt_ref (say ref count becomes = 2)
   2. Register a callback to timer say 'callback1'
   3. If register fails:
           >>>> rpc_clnt_unref (ref count = 1)
   4. On timer expiration, 'callback1' gets called. So unref rpc clnt at the end
      in 'callback1'. This is corresponding to ref taken in step 1
           >>>> rpc_clnt_unref (ref count = 1)
   5. The cycle from step-1 to step-4 continues....until timer cancel event happens
   6. timer cancel of say 'callback1'
           If timer cancel fails:
                 Do nothing, Step-4 would have unrefd
           If timer cancel succeeds:
                 >>>> rpc_clnt_unref (ref count = 1)

Change-Id: I91389bc511b8b1a17824941970ee8d2c29a74a09
BUG: 1316178
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/13658
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Raghavendra G

core: coverity, NULL potinter check

2016-07-22T15:12:31+00:00

iobuf_get_page_aligned should check the return pointer
after calling iobuf_get2.

CID: 1357868

Change-Id: I3aa5b09bce2225aa70cf3c60a6b6f4e4eca1dee6
BUG: 1358936
Signed-off-by: Zhou Zhengping 
Reviewed-on: http://review.gluster.org/14976
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy