glusterfs.git, branch v3.1.3qa3

Solaris xattr support for symlink and special files.

2011-03-04T08:41:15+00:00

Since glusterfs uses xattr for storing gfid, and xattr support for
symlinks and special files does not exist in solaris. The work around
is provided by creating hidden files under export directory on solaris
hosts only. the hidden files ares maintained in .glusterfs_xattr_inode directory,
and all xattr ops on symlink and special files are redirected to respective
inodes.

All dir entries with name starting as .glusterfs (GF_HIDDEN_PATH) will
not be shown in readdir ops.

Signed-off-by: shishir gowda 
Signed-off-by: Anand V. Avati 

BUG: 2213 (Symlink fails with ENODATA)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2213

debug/trace: add logs for 'forget', 'release' and 'releasedir'

2011-03-04T08:41:08+00:00

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand V. Avati 

BUG: 2332 (update 'debug/trace' to log gfid instead of 'ia_ino')
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2332

glusterd: check for the variable before dereferencing it

2011-03-04T08:41:00+00:00

check if 'peerinfo->mgmt' is set (ie, handshake is completed), before
sending any request to that peer.

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand V. Avati 

BUG: 2486 ([glusterfs-3.1.3qa2]: Core generated due to SegFault in glusterd_sm.c)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2486

glusterd: handle a crash in dump_version_cbk (in handshake)

2011-03-04T08:40:53+00:00

reset 'frame->local' to NULL before STACK_DESTROY(), which prevents
freeing of valid 'peerctx' pointer.

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand V. Avati 

BUG: 2485 ([glusterfs-3.1.3qa2]: Core generated due to SegFault in glusterd_rpc_notify)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2485

Make sure we are looking at the right errno in the fix for bug 2452.

2011-03-04T07:30:08+00:00

Signed-off-by: Pavan T C 
Signed-off-by: Anand V. Avati 

BUG: 2452 (Excessive CPU usage /very low throughput while using NFS mounts)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2452

Eliminate syscall tight loop when handling EAGAIN in NFS.

2011-03-04T02:26:41+00:00

Signed-off-by: Pavan T C 
Signed-off-by: Anand V. Avati 

BUG: 2452 (Excessive CPU usage /very low throughput while using NFS mounts)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2452

NFS : Check for duplicate entries while filling up child_up array.

2011-03-03T07:58:16+00:00

Signed-off-by: Gaurav 
Signed-off-by: Anand V. Avati 

BUG: 2483 (Showmount output is not consistent with exported volumes.)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2483

Display correct usage of rebalance command when it is wrongly entered.

2011-03-02T06:07:21+00:00

Signed-off-by: Rahul 
Signed-off-by: Anand V. Avati 

BUG: 1922 (Volume not present wrong message displayed on command line)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1922

gluster rebalance: send the proper 'procnum' to glusterd

2011-03-01T23:10:46+00:00

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand V. Avati 

BUG: 2333 (make glusterd more rpc friendly)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2333

gluster rebalance: give option to split the command

2011-03-01T23:10:40+00:00

the 'gluster volume rebalance  start' is enhanced with two more options:

* 'gluster volume rebalance  fix-layout start' (for fixing layout only)
* 'gluster volume rebalance  migrate-data start' (for migrating data only)

Also the old way of running rebalance in one shot will still work fine

* 'gluster volume rebalance  start'

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand V. Avati 

BUG: 2258 (enhance gluster volume rebalance)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2258