glusterfs.git/xlators/nfs/server/src, branch v3.1.3qa4

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

CLI : NFS disable option through volume set.

2011-03-01T22:02:28+00:00

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

BUG: 2094 (Need option to turn off NFS)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2094

NFS : Fail Nfs init if volume-id not found.

2011-02-11T06:18:11+00:00

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

BUG: 2124 (NFS crashes when volume-id option is not given with dynamic-volume option)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2124

mount3: Do not return error when mount list is empty

2011-02-04T19:21:34+00:00

An exports list can be empty when no subvolumes have come up.
No point returning error and confusing the user.

Signed-off-by: Shehjar Tikoo 
Signed-off-by: Anand V. Avati 

BUG: 2345 (Taking distribute brick down returns 10006 error on mount command)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2345

nfs: Set volume started after root lookup succeeds

2011-02-04T19:21:25+00:00

Signed-off-by: Shehjar Tikoo 
Signed-off-by: Anand V. Avati 

BUG: 2345 (Taking distribute brick down returns 10006 error on mount command)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2345

Logging : Use of uuid_utoa and uuid_utoa_r.

2011-02-04T08:39:26+00:00

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

BUG: 2308 (Threadsafe uuid to string conversion function)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2308

features/access-control: skip access-tests if the call is from fuse

2011-01-27T07:42:21+00:00

Fuse cant send aux gids. So access-control checks treat
non-primary-group membership of user as "other". So skip
access-control checks if the call is from fuse. We added a hack to
treat all calls with pid set to 1 as calls from nfs. So for calls with pid
not 1 we skip the access-control checks on all fops.

Signed-off-by: Pranith Kumar K 
Signed-off-by: Anand V. Avati 

BUG: 2296 (svn / subversion fails on gluster volume (replicated and non-replicated))
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2296

nfs3: Handle root resolution failures

2011-01-13T17:46:51+00:00

..to prevent a second reply to the client which results in accessing
a call state structure that gets freed on the first reply.

Signed-off-by: Shehjar Tikoo 
Signed-off-by: Anand V. Avati 

BUG: 2286 ([glusterfs 3.1.2qa3]: Crash due to corruption in gf_dirent_free)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2286

mgmt/Glusterd: make nfs.mem-factor settable through CLI

2011-01-12T07:58:37+00:00

Signed-off-by: Kaushik BV 
Signed-off-by: Anand V. Avati 

BUG: 2277 (Regression in Gluster NFS re-read performance)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2277

nfs: Introduce tunable for memory consumption

2011-01-12T07:58:33+00:00

NFS has used a common define called the mem-factor to determine
how much memory is allocated for the following items:

o inode table
o local structures used by nfs for storing fop state
o local structures used by nfs3 for storing nfs op state

These are all allocated through the mem-pool. The factor is multiplied by
a multiple that is specific to each data structure. For eg.

define GF_NFS_CONCURRENT_OPS_MULT     15
define GF_NFS_INODE_LRU_MULT          6000

The first value is used for allocating a mem-pool for storing
state for each fop or nfs op that is currently being handled. Knowing that linux
allows at most 128 in-flight requests, this multiple combined with the default
mem-factor of 15 gives us 225 slots in the mem-pool for the local structures.

Similarly, 6000*15 gives us a space of 90k inodes in the lru.

That means, increasing the common mem-factor will allow increasing the perf
under some conditions. This patch introduces the mem-factor as a configurable
option.

Signed-off-by: Shehjar Tikoo 
Signed-off-by: Anand V. Avati 

BUG: 2277 (Regression in Gluster NFS re-read performance)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2277