glusterfs.git/libglusterfs, branch v3.2.3qa1

LICENSE: s/GNU Affero General Public/GNU General Public/

2011-08-06T13:40:14+00:00

Change-Id: Ibf5f45431d7a55b70d7304649af652d6f25bb688
BUG: 3348
Reviewed-on: http://review.gluster.com/183
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

libglusterfs: Bring in os_daemon() routine to replace FreeBSD code

2011-07-28T09:52:26+00:00

Change-Id: I72dda0bb9e3b6cfa8b8ecffe85cc606f5a682313
BUG: 3206
Reviewed-on: http://review.gluster.com/117
Tested-by: Gluster Build System 
Reviewed-by: Amar Tumballi

features/marker: fix race-conditions, memory corruptions and fd-leaks.

2011-07-20T13:12:42+00:00

  - remove xattrs from newpath after rename is complete.
  - hold inodelk on both parents (if they are different) before doing rename
    and gather contribution values of oldpath and newpath to their parents
    while still holding the locks. Use these contribution values to reduce
    parent sizes.
  - performance optimization: abort updation process if delta is zero.
  - libglusterfs/call-stub: Allow unwinding of frames for rename during
    call_resume_unwind.
  - fixes in dirty inode healing codepath:
  	- fix fd-leak of fd opened on the directory.
  	- don't add (instead just assign) next offset at which readdir has to
          be sent to local->d_off.
        - assign to local->d_off before winding lookup call to get child
          contribution.
  - use mutexes while accessing contribution values stored in inode context.
  - use contribution values from backend instead of in-memory while reducing
    parent sizes during rename
  - wipe parent_loc in marker_local_unref.
  - check for refcount being zero holding lock in quota_local_unref.
  - hold parent inodelk during creation of xattrs on directory.
  - use contribution value to reduce parent's size, if the value to be
    subtracted is not passed as argument to reduce_parent_size.
  - skip contribution creation on root.

Change-Id: I97a9ac7efc5cf82abd3837fa6f9766c35676a908
BUG: 2697
Signed-off-by: Junaid 
Reviewed-on: http://review.gluster.com/31
Tested-by: Gluster Build System

libglusterfs/common-utils: Added gf_uint64_2human_readable function.

2011-07-14T13:39:42+00:00

This function converts the given number to its corresponding representation in
KB, MB, etc.

Signed-off-by: Junaid 
Signed-off-by: Anand Avati 

BUG: 2829 (Display the quota limit as it is configured)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2829

glusterfs: add --acl command line option to load ACLs on the client side

2011-07-01T22:58:44+00:00

also set the turn off filtering of ACL keys in FUSE xattr FOPs.

Signed-off-by: Anand Avati 

BUG: 2815 (Server-enforced ACLs)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2815

byte-order: htole/letoh and htobe/betoh for forced endian conversions

2011-07-01T22:58:04+00:00

Signed-off-by: Anand Avati 

BUG: 2815 (Server-enforced ACLs)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2815

stack: make process state dump more debug friendly

2011-06-19T09:16:48+00:00

store points of stack wind and unwind (function names) for each
frame and dump them in process state dump

Signed-off-by: Anand Avati 

BUG: 1059 (enhancements for getting statistics from performance translators)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1059

loc_t: add 'gfid' and 'pargfid' fields

2011-06-14T07:13:51+00:00

these fields are used mainly in case of selfheal path, where
'inode->gfid'||'parent->gfid' is not yet set.

These fields in 'loc' will have lower precedence than 'inode->gfid'
in client protocol.

also contains 'Pranith 's patch to set proper
loc->gfid during afr selfheal

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

BUG: 2346 (Log message enhancements in GlusterFS - phase 1)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346

fix multiple transport related portmap issues in client handshake

2011-06-08T18:19:17+00:00

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

BUG: 2294 (Currently there is no way through cli to make a volume listen on both the transports (socket/rdma))
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2294

syncop: Increase stack size for deep call stack.

2011-06-08T18:18:21+00:00

pump performs a depth-first traversal of directories using the
syncop stack and this necessitates a larger stack when handling
deeper directory trees.

Signed-off-by: Krishnan Parthasarathi 
Signed-off-by: Anand Avati 

BUG: 2489 (GlusterFS crashing with replace-brick)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2489