glusterfs.git, branch v3.8.10

doc: release-notes for GlusterFS-3.8.10

2017-03-12T11:04:28+00:00

BUG: 1421593
Change-Id: I026c8d3b4a401a098ccc568dad8cd51de51dc007
Signed-off-by: Niels de Vos 
Reviewed-on: https://review.gluster.org/16890
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Smoke: Gluster Build System

extras: Add more options to group virt

2017-03-12T10:11:02+00:00

        Backport of: https://review.gluster.org/16577

Apart from some of the option configurations already listed
in the group-virt.example file, we also recommend that the
users set certain other options added by this patch for
VM use-case. This also helps Gluster-oVirt users in configuring
virt options for new volumes at the click of a button as opposed
to setting them manually through volume-set command.

Change-Id: I879d779a4829d7163de4df9ff80d15296130af8a
BUG: 1424934
Signed-off-by: Krutika Dhananjay 
Reviewed-on: https://review.gluster.org/16683
NetBSD-regression: NetBSD Build System 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Pranith Kumar Karampuri 
Reviewed-by: Niels de Vos

cluster/afr: Perform new entry mark before creating new entry

2017-03-11T18:09:59+00:00

There is a chance for the source brick to go down just after
the new entry is created and before source brick is marked with
necessary pending markers. If after this any I/O happens then
new entry will become source and reverse heal will happen.
To prevent this mark the pending xattrs before creating the new
entry.

 >BUG: 1417466
 >Change-Id: I233b87e694d32e5d734df5a83b4d2ca711c17503
 >Signed-off-by: Pranith Kumar K 
 >Reviewed-on: https://review.gluster.org/16474
 >Smoke: Gluster Build System 
 >NetBSD-regression: NetBSD Build System 
 >CentOS-regression: Gluster Build System 
 >Reviewed-by: Ravishankar N 
 >Reviewed-by: Krutika Dhananjay 

BUG: 1429312
Change-Id: Ia1bdaf9511acaeff72a336c8185a56a64ea0e2ba
Signed-off-by: Pranith Kumar K 
Reviewed-on: https://review.gluster.org/16850
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Ravishankar N 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos

storage/posix: Execute syscalls in xattrop under different locks

2017-03-11T17:22:35+00:00

        Backport of: https://review.gluster.org/16462 and
                     https://review.gluster.org/16792

... and not inode->lock. This is to prevent the epoll thread from
*potentially* being blocked on this lock in the worst case for
extended period elsewhere in the brick stack, while the syscalls
in xattrop are being performed under the same lock by a different
thread. This could potentially lead to ping-timeout, if the only
available epoll thread is busy waiting on the inode->lock, thereby
preventing it from picking up the ping request from the client(s).

Also removed some unused functions.

>Change-Id: I2054a06701ecab11aed1c04e80ee57bbe2e52564
>BUG: 1421938
>Signed-off-by: Krutika Dhananjay 
>Reviewed-on: https://review.gluster.org/16462
>Smoke: Gluster Build System 
>NetBSD-regression: NetBSD Build System 
>Reviewed-by: Raghavendra G 
>CentOS-regression: Gluster Build System 
(cherry picked from commit b5c26a462caf97bfc5380c81092f5c331ccaf1ae)

Change-Id: I2054a06701ecab11aed1c04e80ee57bbe2e52564
BUG: 1427390
Signed-off-by: Krutika Dhananjay 
Reviewed-on: https://review.gluster.org/16777
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Raghavendra G 
Reviewed-by: Niels de Vos

afr: restore atime/mtime for non-regular files

2017-03-10T23:14:10+00:00

AFR restores atime/mtime only as a part of data heal. For non-regular
files (dirs, symlinks, char/block/socket files etc) which do not undergo
data-heal, atime/mtime is not restored.

This patch restores atime/mtime as a part of metadata heal for such
files.

> Reviewed-on: https://review.gluster.org/16844
> Reviewed-by: Pranith Kumar Karampuri 
> Tested-by: Pranith Kumar Karampuri 
> Smoke: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> CentOS-regression: Gluster Build System 
(cherry picked from commit 804a65f07ea8e2093f781807651d0d07513b2627)

Change-Id: Id8da885fc93fdf65c2f4bae2af3605b146ac1f16
BUG: 1429405
Signed-off-by: Ravishankar N 
Reviewed-on: https://review.gluster.org/16852
Smoke: Gluster Build System 
Reviewed-by: Pranith Kumar Karampuri 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Niels de Vos

cluster/dht: Fix crash in "nuke-dir" feature

2017-03-10T23:00:58+00:00

        Backport of: https://review.gluster.org/16829

My patch at https://review.gluster.org/16419 is resulting
in core dumps everytime I run tests/features/nuke.t.

Turns out dht, upon successfully "nuking" a directory,
which was initiated through a setxattr, unwinds the operation
with rmdir fop signature, resulting in readdir-ahead casting
a struct iatt (preparent) to dict_t, leading to a crash.

Change-Id: Ib970b3198185a6c641092b00e115a672cb3f9111
BUG: 1428743
Signed-off-by: Krutika Dhananjay 
Reviewed-on: https://review.gluster.org/16840
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Raghavendra G 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos

features/shard: Fix EIO error on add-brick

2017-03-10T22:47:05+00:00

        Backport of: https://review.gluster.org/14419

DHT seems to link inode during lookup even before initializing
inode ctx with layout information, which comes after
directory healing.

Consider two parallel writes. As part of the first write,
shard sends lookup on .shard which in its return path would
cause DHT to link .shard inode. Now at this point, when a
second write is wound, inode_find() of .shard succeeds and
as a result of this, shard goes to create the participant
shards by issuing MKNODs under .shard. Since the layout is
yet to be initialized, mknod fails in dht call path with EIO,
leading to VM pauses.

The fix involves shard maintaining a flag to denote whether
a fresh lookup on .shard completed one network trip. If it
didn't, all inode_find()s in fop path will be followed by a
lookup before proceeding with the next stage of the fop.

Big thanks to Raghavendra G and Pranith Kumar K for the RCA
and subsequent inputs and feedback on the patch.

Change-Id: I66a7adf177e338a7691f441f199dde7c2b90c292
BUG: 1387878
Signed-off-by: Krutika Dhananjay 
Reviewed-on: https://review.gluster.org/16750
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Pranith Kumar Karampuri 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos

features/shard: Put onus of choosing the inode to resolve on individual fops

2017-03-10T22:44:47+00:00

        Backport of: https://review.gluster.org/16709

... as opposed to adding checks in "common" functions to choose the inode
to resolve based local->fop, which is rather ugly and prone to errors.

Change-Id: Ib26d3dd5a7ae43cd27839752bdae2cce56d73e8a
BUG: 1387878
Signed-off-by: Krutika Dhananjay 
Reviewed-on: https://review.gluster.org/16749
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Pranith Kumar Karampuri 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos

cluster/dht Fix error assignment in dht_*xattr2 functions

2017-03-10T22:42:42+00:00

Corrected the op_errno assignments and NULL checks in
the dht_sexattr2 and dht_removexattr2 functions. Earlier,
they unwound with the default EINVAL op_errno if the
file had been deleted.

> Change-Id: Iaf837a473d769cea40132487a966c7f452990071
> BUG: 1421653
> Signed-off-by: N Balachandran 
> Reviewed-on: https://review.gluster.org/16610
> Smoke: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> CentOS-regression: Gluster Build System 
> Reviewed-by: MOHIT AGRAWAL 
> Reviewed-by: Shyamsundar Ranganathan 
(cherry picked from commit 028626a86ea409f908783b9007c02877f20be43e)
Signed-off-by: N Balachandran 

Change-Id: Id2e91df47bcd734dda18700fb075608c1627a608
BUG: 1424915
Reviewed-on: https://review.gluster.org/16678
Tested-by: N Balachandran 
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Shyamsundar Ranganathan

extras/rebalance.py: Fix statvfs for FreeBSD in python

2017-03-10T22:41:01+00:00

FreeBSD doesn't return the block size in f_bsize as linux does. It
returns the optimal I/O size, so we need to consider this to avoid
invalid results. On FreeBSD we take f_frsize as the block size.

> Change-Id: I72083d8ae183548439de874c77f1d60d9c2d14a7
> BUG: 1356076
> Signed-off-by: Xavier Hernandez 
> Reviewed-on: https://review.gluster.org/16498
> CentOS-regression: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> Smoke: Gluster Build System 
> Reviewed-by: Jeff Darcy 

Change-Id: I2ec327ca2c185716294a33c9ee8ca25fab70da8c
BUG: 1425307
Signed-off-by: Xavier Hernandez 
Reviewed-on: https://review.gluster.org/16634
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Niels de Vos