glusterfs.git/libglusterfs/src/glusterfs.h, branch v3.3.0qa29

core: bring a cmdline option to set memory-accounting

2012-03-15T10:33:18+00:00

currently this is implemented as a command line option, and not
as an easier translator option. this is because as of now, before
even the volume files are parsed, we would need memory accounting
enabled. there is scope for improving this behavior, but for now,
this approach solves the problem.

Also, this feature's major consumers are the testers who are
looking for leaks, hence option is hidden from usage output.

Change-Id: I09a5b13743ae43ff42c251989f921319e94cabe3
Signed-off-by: Amar Tumballi 
BUG: 799199
Reviewed-on: http://review.gluster.com/2856
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

core: dict allocations through mem-pool

2012-03-15T05:15:25+00:00

Change-Id: I7401639060957d437808779745a1e46c3f9f4585
Signed-off-by: Amar Tumballi 
BUG: 798503
Reviewed-on: http://review.gluster.com/2851
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

Introduce new extended attribute: node-uuid

2012-02-23T07:19:44+00:00

Request for trusted.glusterfs.node-uuid returns pathinfo
like string but containing the UUID of glusterd instead
of the backend path for the requested file. This info
is benificial for tasks like parallel rebalance that will
make use of the UUID for data locality.

Change-Id: I766a09cc4a5f63aebd11c73107924a1b29242dcf
BUG: 772610
Signed-off-by: Venky Shankar 
Reviewed-on: http://review.gluster.com/2614
Tested-by: Gluster Build System 
Reviewed-by: Shishir Gowda 
Reviewed-by: Vijay Bellur

fuse-bridge: Handle graph-switch.

2012-02-21T09:05:14+00:00

The purpose of this patch is to let protocol/client know when its transports can
be disconnected, without application running on gluster mount noticing any
effects of graph switch.

In order to do this, we migrate all fds and blocked locks to new graph.
Once this migration is complete and there are no in-transit frames as viewed
by fuse-bridge, we send a PARENT_DOWN event to its children. protocol/client
on receiving this event, can disconnect up its transports.

Change-Id: Idcea4bc43e23fb077ac16538b61335ebad84ba16
BUG: 767862
Reviewed-on: http://review.gluster.com/2734
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

cluster/afr: Add commands to see self-heald ops

2012-02-21T05:23:37+00:00

Change-Id: Id92d3276e65a6c0fe61ab328b58b3954ae116c74
BUG: 763820
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/2775
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

storage/posix: Add xattr for gfid2path

2012-02-20T18:05:38+00:00

Change-Id: I1fe987d255bf50e8433043749b482b67554a0ac3
BUG: 763820
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/2774
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

features/index: Index translator implementation

2012-02-20T16:52:33+00:00

Change-Id: If8a11ecbdd010f64fb4409add5751080f4b59086
BUG: 763820
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.com/2722
Tested-by: Gluster Build System 
Reviewed-by: Jeff Darcy 
Reviewed-by: Vijay Bellur

cluster/dht: Rebalance will be a new glusterfs process

2012-02-19T09:31:19+00:00

rebalance will not use any maintainance clients. It is replaced by syncops,
with the volfile. Brickop (communication between glusterd<->glusterfs process)
is used for status and stop commands.

Dept-first traversal of dir is maintained, but data is migrated as and when
encounterd.
fix-layout (dir)
do
        Complete migrate-data of dir
        fix-layout (subdir)
done

Rebalance state is saved in the vol file, for restart-ability.
A disconnect event and pidfile state determine the defrag-status

Signed-off-by: shishirng 
Change-Id: Iec6c80c84bbb2142d840242c28db3d5f5be94d01
BUG: 763844
Reviewed-on: http://review.gluster.com/2540
Tested-by: Gluster Build System 
Reviewed-by: Amar Tumballi

posix: handle some internal behavior in posix_mknod()

2012-02-17T06:48:57+00:00

assume a case of link() systemcall, which is handled in distribute by
creating a 'linkfile' in hashed subvolume, if the 'oldloc' is present
in different subvolume. we have same 'gfid' for the linkfile as that
of file for consistency. Now, a file with multiple hardlinks, we may
end up with 'hardlinked' linkfiles. dht create linkfile using 'mknod()'
fop, and as now posix_mknod() is not equipped to handle this situation.

this patch fixes the situation by looking at the 'internal' key set in
the dictionary to differentiate the call which originates from inside
with regular system calls.

Change-Id: Ibff7c31f8e0c8bdae035c705c93a295f080ff985
BUG: 763844
Signed-off-by: Amar Tumballi 
Reviewed-on: http://review.gluster.com/2755
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

locks: Added a getxattr interface to clear locks on a given inode.

2012-02-14T10:48:59+00:00

getxattr returns a summary of no. of inodelks/entrylks cleared.
cmd_structure: trusted.glusterfs.clrlk.t.k[.{range|basename}]
where,
type = "inode"| "entry"| "posix"
kind = "granted"| "blocked" | "all"
range = off,a-b, where a, b = 'start', 'len' from offset 'off'

Change-Id: I8a771530531030a9d4268643bc6823786ccb51f2
BUG: 789858
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.com/2551
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur