glusterfs.git/xlators/features, branch v3.7.9

features/bit-rot-stub: get frame->local before unwinding

2016-03-09T19:29:26+00:00

In bit-rot-stub, if unlink fails, then it was unwinding
directly. Then it was trying to cleanup local. But local
would be NULL, since it was unwinding directly without getting
the value of frame->local. The NULL cleanup of local was
causing the brick process to crash.

Change-Id: I8544ba73b2e8dc0c50b1a53ff8027d85588d087b
BUG: 1315552
Signed-off-by: Raghavendra Bhat 
Signed-off-by: Venky Shankar 
Reviewed-on: http://review.gluster.org/13630
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System

features/shard: Return ENOTSUP for unsupported fallocate flags

2016-03-09T12:05:08+00:00

        Backport of: http://review.gluster.org/13523

Basis: http://lists.gnu.org/archive/html/qemu-devel/2016-02/msg05101.html

Change-Id: I681eb4d0c43c635cf96a2deab0996dec7a255fe5
BUG: 1299712
Signed-off-by: Krutika Dhananjay 
Reviewed-on: http://review.gluster.org/13652
Smoke: Gluster Build System 
Reviewed-by: Ravishankar N 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Vijay Bellur

afr: misc performance improvements

2016-03-09T02:16:32+00:00

Backport of http://review.gluster.org/#/c/13595/
1. In afr_getxattr_cbk, consider the errno value before blindly
launching an inode refresh and a subsequent retry on other children.

2. We want to accuse small files only when we know for sure that there
is no
IO happening on that inode. Otherwise, the ia_sizes obtained in the
post-inode-refresh replies may mismatch due to a race between
inode-refresh and ongoing writes, causing spurious heal launches.

Change-Id: I9858485d1061db67353ccf99c59530731649c847
BUG: 1309462
Signed-off-by: Ravishankar N 
Reviewed-on: http://review.gluster.org/13644
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Pranith Kumar Karampuri 
Reviewed-by: Krutika Dhananjay 
CentOS-regression: Gluster Build System

features/trash: Handle unlink unwind properly

2016-03-08T15:29:51+00:00

When enabled, trash translator does a rename
internally for every unlink request and unwinds
the original unlink call. But this was unwinded
back with prerparent and postparent as NULL which
resulted in changing the parent directory
permissions to 000.

This issue is consistently seen as a failure
when a non-root user executes vim commands which
internally tries to perform stat operations (as
part of swap/backup file creation) on a file
whose parent directory's permission was modified
to 000 due to recent unlink for another file
inside the same directory.

>Change-Id: I161a036b37fb815866d50d2d6260ff0ad22d7223
>BUG: 1302307
>Signed-off-by: Anoop C S 
>Reviewed-on: http://review.gluster.org/13346
>Smoke: Gluster Build System 
>Tested-by: jiffin tony Thottan 
>Reviewed-by: jiffin tony Thottan 
>CentOS-regression: Gluster Build System 
>NetBSD-regression: NetBSD Build System 
(cherry picked from commit b609a55be4119c44b19252bd951780a78deb21c9)

Signed-off-by: Anoop C S 
Change-Id: I88950b7d2e42bda65272bc359e8dc60a2ce04d89
BUG: 1305749
Reviewed-on: http://review.gluster.org/13401
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: jiffin tony Thottan 
Reviewed-by: Vijay Bellur 
Smoke: Gluster Build System

features/bitrot: do not remove the quarantine handle in forget

2016-03-07T13:04:18+00:00

If an object is marked as bad, then an entry is corresponding to the
bad object is created in the .glusterfs/quarantine directory to help
scrub status. The entry name is the gfid of the corrupted object.
The quarantine handle is removed in below 2 cases.

1) When protocol/server revceives the -ve lookup on an entry whose inode
   is there in the inode table (it can happen when the corrupted object
   is deleted directly from the backend for recovery purpose) it sends a
   forget on the inode and bit-rot-stub removes the quarantine handle in
   upon getting the forget.
   refer to the below commit
   f853ed9c61bf65cb39f859470a8ffe8973818868:
   http://review.gluster.org/12743)

2) When bit-rot-stub itself realizes that lookup on a corrupted object
   has failed with ENOENT.

But with step1, there is a problem when the bit-rot-stub receives forget
due to lru limit exceeding in the inode table. In such cases, though the
corrupted object is not deleted (either from the mount point or from the
backend), the handle in the quarantine directory is removed and that object
is not shown in the bad objects list in the scrub status command.

So it is better to follow only 2nd step (i.e. bit-rot-stub removing the handle
from the quarantine directory in -ve lookups). Also the handle has to be removed
when a corrupted object is unlinked from the mount point itself.

Change-Id: Ibc3bbaf4bc8a5f8986085e87b729ab912cbf8cf9
BUG: 1313131
Original author: Raghavendra Bhat 
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/13472
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Venky Shankar 
(cherry picked from commit 2102010edab355ac9882eea41a46edaca8b9d02c)
Reviewed-on: http://review.gluster.org/13552
Tested-by: Venky Shankar

geo-rep: Mask xtime and stime xattrs

2016-03-07T10:24:18+00:00

Allow access to xtime and stime xattrs only
to gsyncd client and mask them for the rest.

This is to prevent afr from performing self
healing on marker xtime and geo-rep stime
xattr which is not expected as each of which
gets updated them from backend brick and
should not be healed.

BUG: 1313623
Change-Id: I9b4b3ce30bbc09d300e6d5c6782e2446f2411c6f
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/13242
NetBSD-regression: NetBSD Build System 
Reviewed-by: Vijaikumar Mallikarjuna 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Venky Shankar 
Reviewed-on: http://review.gluster.org/13572
Reviewed-by: Aravinda VK

features/trash: Retain file permissions during truncate

2016-03-06T00:18:15+00:00

Consider the situation where directory path for a truncated
file does not exists under trash directory. In this scenario
after creating the required path we failed to create the
orginal file with proper permissions. Eventhough we try to
fetch permissions from local->origpath, it was never filled
with required value in truncate and ftruncate call paths.
This change will copy original location to local->origpath
inside both fop handling functions.

>Change-Id: If5930b6d368d08e58f04db999f3f9edb9250bcb9
>BUG: 1309342
>Signed-off-by: Anoop C S 
>Reviewed-on: http://review.gluster.org/13461
>Smoke: Gluster Build System 
>CentOS-regression: Gluster Build System 
>NetBSD-regression: NetBSD Build System 
>Reviewed-by: jiffin tony Thottan 
>Reviewed-by: Jeff Darcy 
(cherry picked from commit b1cb581424305592fac5394a578b307117b22fe7)

Change-Id: I5d964b4d802551bb04a7011f88edb59a1231238e
BUG: 1313233
Signed-off-by: Anoop C S 
Reviewed-on: http://review.gluster.org/13555
NetBSD-regression: NetBSD Build System 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: jiffin tony Thottan 
Reviewed-by: Vijay Bellur

features/shard: Fix NULL-dereference when fsync fails

2016-03-05T01:28:18+00:00

        Backport of: http://review.gluster.org/#/c/13562/

Change-Id: I6a642203949227452e0f64bcffae23ade1a34fd0
BUG: 1313315
Signed-off-by: Krutika Dhananjay 
Reviewed-on: http://review.gluster.org/13563
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Pranith Kumar Karampuri

cluster/ec: Mark self-heal fops as internal

2016-03-05T01:26:21+00:00

 >Change-Id: I8ae7af266d3e00460f0cfdc9389a926e5f2fee36
 >BUG: 1282761
 >Signed-off-by: Pranith Kumar K 
 >Reviewed-on: http://review.gluster.org/12598
 >Tested-by: Gluster Build System 
 >Tested-by: NetBSD Build System 
 >Reviewed-by: Xavier Hernandez 

BUG: 1283757
Change-Id: Ic20d4ee031265305db1a6ed2cf591ce94b7d0749
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/12668
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System

locks: Handle negative values for flock->l_len

2016-02-28T20:14:14+00:00

As per 'man 3 fcntl',
"If l_len is positive, the area affected shall start at
l_start and end at l_start+l_len−1. If l_len is negative,
the area affected shall start at l_start+l_len and end at
l_start−1. Locks may start and extend beyond the current
end of a file, but shall not extend before the beginning
of the file."

Currently we return EINVAL if l_len is found to be negative.
Fixed the same as mentioned in the man page.

This is backport of the below patch
    - http://review.gluster.org/11613

Change-Id: I493ce202c543185fc4ae7266d1aaf9d7e2a66991
BUG: 1312200
Signed-off-by: Soumya Koduri 
Reviewed-on: http://review.gluster.org/11613
Reviewed-by: Kaleb KEITHLEY 
Reviewed-by: Niels de Vos 
Reviewed-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/13526
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System