glusterfs.git/xlators/features, branch v3.3.0

geo-rep / gsyncd: add "--super" to rsync invocation

2012-05-27T16:57:46+00:00

This forces rsync to perform supposedly privileged operations on
unprivileged slaves (like chown(2)).

For consistent behavior (with gsyncd's "chown" RPC call that's
being used for symlinks and directories), we also pass
"--numeric-ids" to rsync.

Also took the chance to retire gsyncd's "--rsync-extra" option
which was there for debugging purposes (related to a resolved
issue).

Change-Id: I4ee4d0d3a8c4e0f6746d34d7722c8a567a67491c
BUG: 822121
Signed-off-by: Csaba Henk 
Reviewed-on: http://review.gluster.com/3426
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati 
Reviewed-on: http://review.gluster.com/3453
Reviewed-by: Vijay Bellur

locks: Fixed opt arg parsing in clear-locks

2012-05-24T18:08:08+00:00

Change-Id: I470fd21d5d53e3c6f0bd2a4f84c6327532e18559
BUG: 823151
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.com/3429
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

features/marker: use the gfid from the stat structure instead of inode

2012-05-23T09:10:24+00:00

in fresh lookup, the inode would not have linked to the inode table
until the fop reaches back to protocol/server, thus it would not contain
the gfid within it (gfid would still be null). So use the stat structure
to get the gfid in lookup callback instead of inode's gfid.

Change-Id: Id70277f0228f3db64b05d613108cfb4f070197e6
BUG: 791087
Signed-off-by: Raghavendra Bhat 
Reviewed-on: http://review.gluster.com/3400
Tested-by: Gluster Build System 
Reviewed-by: Amar Tumballi 
Reviewed-by: Vijay Bellur

geo-rep / gsyncd: typo fix

2012-05-23T09:00:08+00:00

fix topy.

Change-Id: I84df3e850dd24d9e86713dfa401c603a84a81ca6
BUG: 763302
Signed-off-by: Csaba Henk 
Reviewed-on: http://review.gluster.com/3375
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur 
Reviewed-on: http://review.gluster.com/3407

geo-rep / gsyncd: recognize ECONNABORTED as termination of aux glusterfs

2012-05-23T08:59:47+00:00

Don't dump stack, rather log the "glusterfs session went down" message.

If the aux glusterfs is already dead when we try to do some file
operation, we get a failure with ENOTCONN, which is already handled
as above. However, it's also possible that glusterfs dies while we
are in a syscall into it -- in that case we get ECONNABORTED, and
so far then we end up with an ugly stack strace. From now on we
take ECONNABORTAD as well into consideration.

Nb. wrt. testing: it's not easy to synthetically force the aux glusterfs
to end this way; for that we have to provoke gsyncd into intensive
synchronization. I succeeded in that with the following ruby oneliner:

ruby -rcgi -e '
  Dir.chdir($*[0])
  a=[]
  Thread.new { loop { while a.size >= 100; File.delete a.shift; end; sleep 1 }}
  loop { a<
Reviewed-on: http://review.gluster.com/3078
Tested-by: Gluster Build System 
Reviewed-by: Venky Shankar 
Reviewed-by: Vijay Bellur 
Reviewed-on: http://review.gluster.com/3408

geo-rep / gsyncd: fixes regarding the command invocation framework

2012-05-22T18:25:23+00:00

Some of the bugs to fix were found by the following stress-test:
make "glusterfs --client-pid=-1" exit immediately on slave
side.

Also fix eintr_wrap which should not "adopt" exceptions generated
by the wrapped call, by re-raising them as GsyncdError.

Change-Id: Ia0d39e0635975ebbbf98d86e1e26f3122e1ed6ff
BUG: 764678
Signed-off-by: Csaba Henk 
Reviewed-on: http://review.gluster.com/3258
Tested-by: Gluster Build System 
Reviewed-by: Venky Shankar 
Reviewed-on: http://review.gluster.com/3409
Reviewed-by: Vijay Bellur

geo-rep / gsyncd.c: fix coverity fix

2012-05-22T18:22:48+00:00

gsyncd wrapper was segfaulting as coverity fix freed
up pointer at wrong place (after it was reused)

Instead of the apporach of the original coverity fix
that added elaborate control flow to hunt down potential
leaks, here we move the code over to static allocations
in place of (the not really necessary) dynamic ones.

Change-Id: Ida3855ff4a4f4371b350d27f858f129ceed51785
BUG: 789278
Signed-off-by: Csaba Henk 
Reviewed-on: http://review.gluster.com/3345
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur 
Reviewed-on: http://review.gluster.com/3405

features/locks: Don't delete blocked locks in pl_flush

2012-05-22T11:30:19+00:00

BUG: 810502
Change-Id: I68c1c11754e1f4413b457c5f4bb7da3bc1aafc23
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/3398
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

features/locks: insert_and_merge should not operate on blocked locks

2012-05-22T11:30:01+00:00

BUG: 810502
Change-Id: I1c560dae638ddb66f47ea64b661979b31708731b
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/3397
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

features/quota: get the top xlator from the active graph outside the list

2012-05-20T07:28:01+00:00

In quota reconfigure we were getting the top of the active graph within the
list which contains the limits set on the volume. Suppose the list is empty,
then the code inside the list traversal loop is not executed, and thus top
will be NULL which might lead to segfault when accessed later.

Change-Id: I1648d69256490878339294537e17300399ddbf2c
BUG: 822827
Signed-off-by: Raghavendra Bhat 
Reviewed-on: http://review.gluster.com/3379
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur