glusterfs.git/xlators/mgmt/glusterd, branch v3.6.0beta3

glusterd/quota: Heal pgfid xattr on existing data when the quota is enable

2014-09-30T16:42:40+00:00

This is a backport of http://review.gluster.org/#/c/8878/

The pgfid extended attributes are used to construct the ancestry path
(from the file to the volume root) for nameless lookups on files.
As NFS relies on nameless lookups heavily, quota enforcement through NFS
would be inconsistent if quota were to be enabled on a volume with
existing data.

Solution is to heal the pgfid extended attributes as a part of lookup
perfomed by quota-crawl process. In a posix lookup check for pgfid xattr
and if it is missing set the xattr.

BUG: 1147953
Change-Id: I707d91a056e07452bfd1e070af5eddaa752a84ac
Signed-off-by: vmallika 
Reviewed-on: http://review.gluster.org/8890
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

glusterd: Improve debugging experience for glusterd locks

2014-09-26T16:58:21+00:00

Today, when glusterd's internal locking mechanism fails
with invalid type or when another competing lock is being
held, the log message doesn't provide enough information
directly as to which command saw this (first). Following
is a snippet of how a failure would look in the log file.
This would greatly assist in debugging.

[2014-09-03 04:57:58.549418] E
[glusterd-locks.c:520:glusterd_mgmt_v3_lock]
(-->/usr/local/lib/glusterfs/3.7dev/xlator/mgmt/glusterd.so(__glusterd_handle_create_volume+0x801)
[0x7f30b071e651]
(-->/usr/local/lib/glusterfs/3.7dev/xlator/mgmt/glusterd.so(glusterd_op_begin_synctask+0x2c)
[0x7f30b072e19c]
(-->/usr/local/lib/glusterfs/3.7dev/xlator/mgmt/glusterd.so(gd_sync_task_begin+0x55d)
[0x7f30b072de6d]))) 0-management: Invalid entity. Cannot perform locking
operation on vol types

Change-Id: I0595f49d60e620e8b065f3506bdb147ccee383a7
BUG: 1145093
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/8842
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

OSX/FreeBSD: Regression fix

2014-09-24T23:34:05+00:00

Introduced in "1f6e992f1aaa676be5bd47d17e58f1171825cf43"

Change-Id: Id684e2f082def7d01ef3c258ea6598da6205591f
BUG: 1117822
Signed-off-by: Harshavardhana 
Reviewed-on: http://review.gluster.org/8840
Reviewed-by: Justin Clift 
Tested-by: Justin Clift

Do not hardcode setfattr(1) path

2014-09-24T07:03:53+00:00

Turn setfattr(1) absolute path into an OS-dependant macro. Let compiler
option override it to fit custom installation if needed.

Backport of I8f469c5741a85b6e8d8f6299a9540b3d64611d2f

BUG: 1138897
Change-Id: I279752f2ec5db1abc25830cb9a23290cc401d517
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/8828
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

glusterd: Add last successful glusterd lock backtrace

2014-09-24T07:00:56+00:00

Also, moved the backtrace fetching logic to a separate function.
Modified the backtrace fetching logic able to work under memory pressure
conditions.

Change-Id: Ie38bea425a085770f41831314aeda95595177ece
BUG:1145093
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/8794
Tested-by: Gluster Build System 
Reviewed-by: Atin Mukherjee 
Reviewed-by: Niels de Vos 
Reviewed-by: Vijay Bellur

glusterd: Authenticate management handshake requests

2014-09-24T07:00:13+00:00

Backport of
  371bb42 glusterd: Authenticate management handshake requests
from master.

Management handshake requests, which are used to validate op-version
supported by the peers, are now only allowed if,
- the glusterd doesn't have any other peer, or
- the request was sent by another peer.

This prevents the op-version of a peer being changed because of a
connection attempt by an invalid peer.

BUG: 1144978
Change-Id: I5a909dad37e9873efe8b75dad41b7af71ce91c3d
Signed-off-by: Kaushal M 
Reviewed-on: http://review.gluster.org/8819
Reviewed-by: Atin Mukherjee 
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

glusterd: Set the rlimit for Open FDs to higher value.

2014-09-23T10:14:54+00:00

Default 'open FD limit' is 1024. As the number of volumes/bricks
increases, brick-to-glusterd socket FDs also increases in glusterd and
runs out of the limit.

Solution is to set the 'Open FD' limit to higher value in glusterd

Change-Id: Iaa60b2155df2fa5a0759e054bdebffbc09f63ec1
BUG: 1145095
Signed-off-by: Vijaikumar M 
Reviewed-on: http://review.gluster.org/8578
Reviewed-by: Rajesh Joseph 
Reviewed-by: Atin Mukherjee 
Tested-by: Gluster Build System 
Reviewed-by: Krishnan Parthasarathi 
Tested-by: Krishnan Parthasarathi 
Signed-off-by: Sachin Pandit 
Reviewed-on: http://review.gluster.org/8807

glusterd/snapshot : Fail the snapshot create operation if geo-rep is running

2014-09-23T10:13:25+00:00

As one of the recommandations for taking a snapshot is not to have
an active geo-replication session, its better to display an error
saying session is active when snapshot create command is issued.

Change-Id: I94593dbd2659610e033ca316176dda1ac8dc5ce6
BUG: 1145091
Signed-off-by: Sachin Pandit 
Reviewed-on: http://review.gluster.org/8461
Tested-by: Gluster Build System 
Reviewed-by: Rajesh Joseph 
Reviewed-by: Atin Mukherjee 
Reviewed-by: Vijaikumar Mallikarjuna 
Reviewed-by: Krishnan Parthasarathi 
Tested-by: Krishnan Parthasarathi 
Signed-off-by: Sachin Pandit 
Reviewed-on: http://review.gluster.org/8804

glusterd/snapshot: Inherit the mount options of a original brick when creating snapshots.

2014-09-23T10:11:42+00:00

When creating a snapshot a LVM is created at the backend and is mounted
under /var/run/gluster/snaps/... However, this mount does not inherit
the mount options for the original brick acting as the parent for the
snap.

If the snap is restored, this could lead to performance degredations,
functional limitations, or in extreme scenarios even potential data
loss.

Change-Id: I67d70fd83430d83dacc5380c6c928e27fb9c9e1b
BUG: 1145088
Signed-off-by: Vijaikumar M 
Reviewed-on: http://review.gluster.org/8394
Reviewed-by: Krishnan Parthasarathi 
Tested-by: Krishnan Parthasarathi 
Signed-off-by: Sachin Pandit 
Reviewed-on: http://review.gluster.org/8802
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

glusterd: Error msg for snapshot status for no existing snaps should be aligned with error messages of info and list.

2014-09-23T09:01:37+00:00

When a snapshot operation like status, info, list
performed on a non-existing snapshot.
For Status error message is displayed as 'Snap not found'
For List and Info error message is displayed as 'Snapshot does not exist'
Have the consistant error message all the places

Change-Id: I7b241217dba62fda844481731a6858e4ecb12897
BUG: 1145087
Signed-off-by: Vijaikumar M 
Reviewed-on: http://review.gluster.org/8309
Reviewed-by: Krishnan Parthasarathi 
Tested-by: Krishnan Parthasarathi 
Signed-off-by: Sachin Pandit 
Reviewed-on: http://review.gluster.org/8801
Tested-by: Gluster Build System