glusterfs.git/xlators/cluster/afr/src, branch v3.5.4

cluster/afr: Make read child match check in afr optional

2015-03-18T11:21:28+00:00

        Backport of: http://review.gluster.org/#/c/9917

Having this particular check which was introduced by
commit bb2df4e63fa8a5d65f18b4a5efc757e8d475fbff causes a drop in
performance in readdirp. So the behavior is made configurable with this
patch.

Change-Id: I9012a6bb955229a0cbb48f06e4e2edc0782dfead
BUG: 1202675
Signed-off-by: Krutika Dhananjay 
Reviewed-on: http://review.gluster.org/9924
Reviewed-by: Pranith Kumar Karampuri 
Reviewed-by: Atin Mukherjee 
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

afr: exit out of stack winds in for loops if call_count is zero

2015-03-12T04:14:38+00:00

....in order to avoid a race where the fop cbk frees the frame's local
variables and the fop tries to access it at a later point in time.

Change-Id: I91d2696e5e183c61ea1368b3a538f9ed7f3851de
BUG: 1200764
Signed-off-by: Ravishankar N 
Reviewed-on: http://review.gluster.org/9856
Tested-by: Gluster Build System 
Reviewed-by: pranith karampuri 
Reviewed-by: Niels de Vos

afr: Don't write to sparse regions of sink.

2015-02-12T12:09:13+00:00

Corresponding afr-v2 fix: http://review.gluster.org/#/c/9480/

Problem:
When data-self-heal-algorithm is set to 'full', shd just reads from
source and writes to sink. If source file happened to be sparse (VM
workloads), we end up actually writing 0s to the corresponding regions
of the sink causing it to lose its sparseness.

Fix:
If the source file is sparse, and the data read from source and sink are
both zeros for that range, skip writing that range to the sink.

Change-Id: Iade957e4173c87e45a2881df501ba2ad3eb1a172
BUG: 1190633
Signed-off-by: Ravishankar N 
Reviewed-on: http://review.gluster.org/9611
Reviewed-by: Pranith Kumar Karampuri 
Reviewed-by: Krutika Dhananjay 
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

features/marker: Filter internal xattrs in lookup

2015-02-12T12:08:12+00:00

        Backport of http://review.gluster.com/9061

Afr should ignore quota-size-key as part of self-heal
but should heal quota-limit key.

BUG: 1162230
Change-Id: I639cfabbc44468da29914096afc7e2eca1ff1292
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/9091
Tested-by: Gluster Build System 
Reviewed-by: Ravishankar N 
Reviewed-by: Raghavendra G 
Reviewed-by: Krutika Dhananjay 
Reviewed-by: Niels de Vos

cluster/afr: serialize inode locks

2015-02-11T10:15:30+00:00

      Backport of http://review.gluster.com/9372

Problem:
Afr winds inodelk calls without any order, so blocking inodelks
from two different mounts can lead to dead lock when mount1 gets
the lock on brick-1 and blocked on brick-2 where as mount2 gets
lock on brick-2 and blocked on brick-1

Fix:
Serialize the inodelks whether they are blocking inodelks or
non-blocking inodelks.

        Non-blocking locks also need to be serialized.
Otherwise there is a chance that both the mounts which issued same
non-blocking inodelk may endup not acquiring the lock on any-brick.
Ex:
Mount1 and Mount2 request for full length lock on file f1.  Mount1 afr may
acquire the partial lock on brick-1 and may not acquire the lock on brick-2
because Mount2 already got the lock on brick-2, vice versa. Since both the
mounts only got partial locks, afr treats them as failure in gaining the locks
and unwinds with EAGAIN errno.

Change-Id: I939a1d101e313a9f0abf212b94cdce1392611a5e
BUG: 1177928
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/9374
Reviewed-by: Krutika Dhananjay 
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

cluster/afr: Preserve errno in case of failures on all subvols

2015-02-11T10:13:46+00:00

        Partly backported from http://review.gluster.org/8984

Problem:
When quorum is enabled and the fop fails on all the subvolumes,
op_errno is set to EROFS which overrides the actual errno returned
from bricks.

Fix:
Don't override the errno when fop fails on all subvols.

PS: Afr-v2 code differs from afr-v1 so that pre-op part of code doesn't apply

Change-Id: I61e57bbf1a69407230ec172a983de18d1c624fd2
BUG: 1162150
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/9088
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

cluster/afr: When parent and entry read subvols are different, set entry->inode to NULL

2015-02-05T14:52:45+00:00

        Backport of: http://review.gluster.org/#/c/9477

That way a lookup would be forced on the entry, and its attributes will
always be selected from its read subvol.

Additionally, directory write fops as well as LOOKUP have been made to
unwind parent attributes from parent's read child in AFR.

Change-Id: I9fca49fa91cc3a65f53db855fedb90b08f1ca7f4
BUG: 1186121
Signed-off-by: Krutika Dhananjay 
Reviewed-on: http://review.gluster.org/9504
Tested-by: Gluster Build System 
Reviewed-by: Pranith Kumar Karampuri 
Tested-by: Pranith Kumar Karampuri

core: fix Ubuntu code audit (cppcheck) results

2014-11-26T08:30:12+00:00

See http://review.gluster.org/#/c/7583/ BZ 1086460

AFAICT these are false positives:

[geo-replication/src/gsyncd.c:99]: (error) Memory leak: str
[geo-replication/src/gsyncd.c:395]: (error) Memory leak: argv
[xlators/nfs/server/src/nlm4.c:1200]: (error) Possible null pointer dereference: fde

Program exits, resource leak not an issue
[extras/geo-rep/gsync-sync-gfid.c:105]: (error) Resource leak: fp

Test program:
[extras/test/test-ffop.c:27]: (error) Buffer overrun possible for long command line arguments.

Not built:
[xlators/cluster/ha/src/ha.c:2699]: (error) Possible null pointer dereference: priv

The remainder are fixed with this change-set:

[heal/src/glfs-heal.c:357]: (error) Possible null pointer dereference: remote_subvol
[libglusterfs/src/xlator.c:648]: (error) Uninitialized variable: gfid
[libglusterfs/src/xlator.c:649]: (error) Uninitialized variable: gfid
[xlators/cluster/afr/src/afr-inode-write.c:469]: (error) Possible null pointer dereference: frame
[xlators/cluster/afr/src/afr-self-heal-common.c:1704]: (error) Possible null pointer dereference: local
[xlators/cluster/dht/src/dht-rebalance.c:1643]: (error) Possible null pointer dereference: ctx
[xlators/cluster/stripe/src/stripe.c:4963]: (error) Possible null pointer dereference: local
[xlators/features/changelog/src/changelog.c:1464]: (error) Possible null pointer dereference: priv
[xlators/mgmt/glusterd/src/glusterd-geo-rep.c:1656]: (error) Possible null pointer dereference: command
[xlators/mgmt/glusterd/src/glusterd-replace-brick.c:914]: (error) Resource leak: file
[xlators/mgmt/glusterd/src/glusterd-replace-brick.c:998]: (error) Resource leak: file
[xlators/mgmt/glusterd/src/glusterd-sm.c:248]: (error) Possible null pointer dereference: new_ev_ctx
[xlators/mgmt/glusterd/src/glusterd-store.c:1332]: (error) Possible null pointer dereference: handle
[xlators/mgmt/glusterd/src/glusterd-utils.c:4706]: (error) Possible null pointer dereference: this
[xlators/mgmt/glusterd/src/glusterd-utils.c:5613]: (error) Possible null pointer dereference: this
[xlators/mgmt/glusterd/src/glusterd-utils.c:6342]: (error) Possible null pointer dereference: path_tokens
[xlators/mgmt/glusterd/src/glusterd-utils.c:6343]: (error) Possible null pointer dereference: path_tokens
[xlators/mount/fuse/src/fuse-bridge.c:4591]: (error) Uninitialized variable: finh
[xlators/mount/fuse/src/fuse-bridge.c:3004]: (error) Possible null pointer dereference: state
[xlators/nfs/server/src/nfs-common.c:89]: (error) Dangerous usage of 'volname' (strncpy doesn't always null-terminate it).
[xlators/performance/quick-read/src/quick-read.c:585]: (error) Possible null pointer dereference: iobuf

Rerunning cppcheck afterwards:

As before, test program:
[extras/test/test-ffop.c:27]: (error) Buffer overrun possible for long command line arguments.

As before, believed to be false positive:
[geo-replication/src/gsyncd.c:99]: (error) Memory leak: str
[geo-replication/src/gsyncd.c:395]: (error) Memory leak: argv
[xlators/nfs/server/src/nlm4.c:1200]: (error) Possible null pointer dereference: fde

As before, not built:
[xlators/cluster/ha/src/ha.c:2699]: (error) Possible null pointer dereference: priv

False positive after fix:
[heal/src/glfs-heal.c:356]: (error) Possible null pointer dereference: remote_subvol
[xlators/cluster/stripe/src/stripe.c:4963]: (error) Possible null pointer dereference: local

Change-Id: Ib3029d3223f5a13e2ac386a527d64d5ffe3ecb90
BUG: 1092037
Signed-off-by: Kaleb S. KEITHLEY 
Reviewed-on: http://review.gluster.org/7605
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

cluster/afr : Prevent excessive logging of split-brain messages.

2014-11-13T19:55:41+00:00

        Running the volume heal info command would result in
excessive logging of split-brain messages. After this patch,
running heal info command will not log the split brain messages.
This info is now displayed in the output of heal info command
instead.
If a file is in split-brain, a message "Is in split-brain"
will be written against its name.

Change-Id: Ib8979be04f5ac7c59ce3ad1185886bb54b8be808
BUG: 1161102
Signed-off-by: Anuradha 
Reviewed-on: http://review.gluster.org/9069
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

cluster/afr: Fix xattr heal comparison checks

2014-11-13T19:53:07+00:00

        Backport of part of the fixes in http://review.gluster.org/8558

Problem:
While implementing list-xattr based meta-data self-heal for afr-v2 we found
2 issues, with afr-v1's implementation.

1) change in QUOTA_SIZE_KEY xattr value can trigger spurious  metadata
   self-heal.
2) xattr comparison function that is implemented for afr-v1 checks if the
   number of xattrs in both the xattrs is same and then checks that the xattrs
   present in brick-1's response are present and equal. But what we observed me
   was that count also contains the gluster internal/virtual xattrs where as
   the compare function should only compare on-disk external xattrs that can be
   healed. So the correct implementation should check that the external xattrs
   in first brick's response are present in second brick's response and vide
   versa.

Fix:
This patch is partly backported from afr-v2's implementation. Will be providing
the links where necessary.

1) Added QUOTA_SIZE_KEY xattr to the list of xattrs that need to be ignored.
   (http://review.gluster.org/#/c/8558/10/xlators/cluster/afr/src/afr-common.c
   line: 1155)
2) For xattrs to be equal, check all keys in xattr-dict1 are in xattr-dict2 and
   equal and vice versa.
   (http://review.gluster.org/#/c/8558/10/xlators/cluster/afr/src/afr-common.c
   line: 1195)

Change-Id: I63aa74858c6f608b98d1fe425b3fa56f925bb5b3
BUG: 1162230
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/9090
Reviewed-by: Niels de Vos 
Tested-by: Gluster Build System