glusterfs.git/xlators/cluster, branch v4.0.0-2

cluster/afr: Fix dict-leak in pre-op

2018-03-03T04:33:26+00:00

At the time of pre-op, pre_op_xdata is populted with the xattrs we get from the
disk and at the time of post-op it gets over-written without unreffing the
previous value stored leading to a leak.
This is a regression we missed in
https://review.gluster.org/#/q/ba149bac92d169ae2256dbc75202dc9e5d06538e

 >BUG: 1550078
 >Change-Id: I0456f9ad6f77ce6248b747964a037193af3a3da7
 >Signed-off-by: Pranith Kumar K 
 >(cherry picked from commit e7b79c59590c203c65f7ac8548b30d068c232d33)

BUG: 1550808
Change-Id: I0456f9ad6f77ce6248b747964a037193af3a3da7

cluster/dht: Handle single dht child in dht_lookup

2018-02-23T15:05:17+00:00

This patch limits itself to only handling the case
where no file (data or linkto) exists on the subvol.

Additional cases to be handled:
1. A linkto file was found on the only child subvol. This currently
calls dht_lookup_everywhere which eventually deletes it. It can be
deleted directly as it will not be pointing to a valid subvol.
2. Directory lookups - locking might be unnecessary in some cases.

> Change-Id: I940ba34531f2aaee1d36fd9ca45ecfd46be662a4
> BUG: 1546620
> Signed-off-by: N Balachandran 

Change-Id: I940ba34531f2aaee1d36fd9ca45ecfd46be662a4
BUG: 1548271
Signed-off-by: N Balachandran

cluster/dht: Ignore ENODATA from getxattr for posix acls

2018-02-23T15:04:23+00:00

dht_migrate_file no longer prints an error if getxattr for
posix acls fails with ENODATA/ENOATTR.

> Change-Id: Id9ecf6852cb5294c1c154b28d609889ea3420e1c
> BUG: 1546954
> Signed-off-by: N Balachandran 

Change-Id: Id9ecf6852cb5294c1c154b28d609889ea3420e1c
BUG: 1548264
Signed-off-by: N Balachandran

cluster/dht: Fixed a typo

2018-02-23T15:03:05+00:00

Replaced "then" with "than"

> Change-Id: I73090e8c1a639befd7c5458e8d63bd173248bc7d
> BUG: 1547128
> Signed-off-by: N Balachandran 

Change-Id: I73090e8c1a639befd7c5458e8d63bd173248bc7d
BUG: 1547842
Signed-off-by: N Balachandran

posix/afr: handle backward compatibility for rchecksum fop

2018-02-20T03:12:02+00:00

Added a volume option 'fips-mode-rchecksum' tied to op version 4.
If not set, rchecksum fop will use MD5 instead of SHA256.

updates: #230
Change-Id: Id8ea1303777e6450852c0bc25503cda341a6aec2
Signed-off-by: Ravishankar N 
(cherry picked from commit 6daa6535692b2c68b493636a9bbfdcbc475b3d80)

cluster/dht: Unlink linkto files as root

2018-02-09T13:44:10+00:00

Non-privileged users cannot delete linkto
files. However the failure to unlink a stale linkto
causes DHT to fail the lookup with EIO and hence
prevent access to the file.

> Change-Id: Id295362d41e52263790694602f36f1219f0646a2
> BUG: 1542318
> Signed-off-by: N Balachandran 

Change-Id: Id295362d41e52263790694602f36f1219f0646a2
BUG: 1543487
Signed-off-by: N Balachandran

cluster/dht: avoid overwriting client writes during migration

2018-02-06T14:33:30+00:00

For more details on this issue see
https://github.com/gluster/glusterfs/issues/308

Solution:
This is a restrictive solution where a file will not be migrated
if a client writes to it during the migration. This does not
check if the writes from the rebalance and the client actually
do overlap.

If dht_writev_cbk finds that the file is being migrated (PHASE1)
it will set an xattr on the destination file indicating the file
was updated by a non-rebalance client.
Rebalance checks if any other client has written to the dst file
and aborts the file migration if it finds the xattr.

updates gluster/glusterfs#308

Change-Id: I73aec28bc9dbb8da57c7425ec88c6b6af0fbc9dd
Signed-off-by: Susant Palai 
Signed-off-by: Raghavendra G 
Signed-off-by: N Balachandran 
(cherry picked from commit 545a7ce6762a1b3a7b989b43a9d18b5b1b299df0)

afr: capture the correct errno in post-op quorum check

2018-02-06T14:25:31+00:00

If the post-op phase of txn did not meet quorm checks, use that errno to
unwind the FOP rather than blindly setting ENOTCONN.

Change-Id: I0cb0c8771ec75a45f9a25ad4cd8601103deddf0c
BUG: 1542382
Signed-off-by: Ravishankar N 
(cherry picked from commit 440a048f24b006c80af3d7bcd0a1f13fe3459d87)

afr: don't treat all cases all bricks being blamed as split-brain

2018-02-06T14:25:14+00:00

Problem:
We currently don't have a roll-back/undoing of post-ops if quorum is not
met. Though the FOP is still unwound with failure, the xattrs remain on
the disk.  Due to these partial post-ops and partial heals (healing only when
2 bricks are up), we can end up in split-brain purely from the afr
xattrs point of view i.e each brick is blamed by atleast one of the
others. These scenarios are hit when there is frequent
connect/disconnect of the client/shd to the bricks while I/O or heal
are in progress.

Fix:
Instead of undoing the post-op, pick a source based on the xattr values.
If 2 bricks blame one, the blamed one must be treated as sink.
If there is no majority, all are sources. Once we pick a source,
self-heal will then do the heal instead of erroring out due to
split-brain.

Change-Id: I3d0224b883eb0945785ade0e9697a1c828aec0ae
BUG: 1542380
Signed-off-by: Ravishankar N 
(cherry picked from commit 0e6e8216823c2d9dafb81aae0f6ee3497c23d140)

cluster/afr: remove unnecessary child_up initialization

2018-02-06T14:24:48+00:00

The child_up array was initialized with all elements being -1 to
allow afr_notify() to differentiate down bricks from bricks that
haven't reported yet. With current implementation this is not needed
anymore and it was causing unexpected results when other parts of
the code considered that if child_up[i] != 0, it meant that it was up.

Backport of:
> BUG: 1541038

Change-Id: I2a9d712ee64c512f24bd5cd3a48dcb37e3139472
BUG: 1541928
Signed-off-by: Xavier Hernandez