glusterfs.git/geo-replication/syncdaemon, branch v3.7.0beta1

geo-rep: Log Rsync performance

2015-04-02T13:47:16+00:00

Introducing configurable option to log the rsync performance.

gluster volume geo-replication  :: \
        config log-rsync-performance true

Default value is False.

Example log:
[2015-03-31 16:48:34.572022] I [resource(/bricks/b1):857:rsync] SSH: rsync
performance: Number of files: 2 (reg: 1, dir: 1), Number of regular files
transferred: 1, Total file size: 178 bytes, Total transferred file
size: 178 bytes, Literal data: 178 bytes, Matched data: 0 bytes,
Total bytes sent: 294, Total bytes received: 32, sent 294 bytes
received 32 bytes  652.00 bytes/sec

Change-Id: If11467e29e6ac502fa114bd5742a8434b7084f98
Signed-off-by: Aravinda VK 
BUG: 764827
Reviewed-on: http://review.gluster.org/10070
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

geo-rep: Add support for acls

2015-03-30T13:57:16+00:00

This patch adds support for ACLS. When it sees SETXATTR
in Changelog, it adds the file to data queue. rsync/tar+ssh
will take care of syncing ACLS. User set ACLS will be
synced to Slave.

This requires "system.posix_acl_access" to go through when
client-pid is equal GF_CLIENT_PID_GSYNCD in fuse layer.

New config interface is introduced, sync-acls
Which can be set using geo-rep config(Default is True)

gluster volume geo-replication  :: \
               config sync-acls false

Change-Id: I7eb3523fa72b8fed830efc98138891244e830d65
BUG: 1187021
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/10001
Reviewed-by: Aravinda VK 
Tested-by: Gluster Build System 
Reviewed-by: Venky Shankar 
Tested-by: Venky Shankar

geo-rep: Use gfchangelog context init API

2015-03-27T09:04:09+00:00

With the RPC based changes to {libgf}changelog, changelog_init
is required before changelog_register.

Change-Id: Id125b2bd2e51aaaffa22ecab463dfb739c50d83c
Signed-off-by: Venky Shankar 
BUG: 1170075
Reviewed-on: http://review.gluster.org/9993
Reviewed-by: Saravanakumar Arumugam 
Tested-by: Saravanakumar Arumugam 
Tested-by: Gluster Build System

geo-rep: Ignore .trashcan during xsync crawl.

2015-03-19T04:22:55+00:00

With trash feature, .trashcan directory gets created
at each export directory. Xsync picks .trashcan to sync
and fails with EPERM. Xsync should ignore .trashcan
directory.

Change-Id: I45bd226c96011ace2c40dd2de878d886c7d34ce5
BUG: 1203293
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/9934
Reviewed-by: Vijay Bellur 
Reviewed-by: Aravinda VK 
Reviewed-by: Venky Shankar 
Tested-by: Gluster Build System

geo-rep / gsyncd: use RTLD_GLOBAL while loading libgfchangelog

2015-03-19T01:22:50+00:00

With the RPC based changes to {libgf}changelog, loading shared
objects dynamically would need symbols to be available from
other shared libraries. As an example, creating an RPC listner
loads the RPC transport shared object which requires symbols
to be available from already loaded shared objects.

Using RTLD_GLOBAL makes the symbols available for symbol
resolution of subsequently loaded libraries.

Change-Id: I3d3ef790eded82911f05836c707509157680645c
BUG: 1170075
Signed-off-by: Venky Shankar 
Reviewed-on: http://review.gluster.org/9814
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

geo-rep: Do not use xsync_upper_limit for change detection

2015-03-16T05:55:31+00:00

Use register time(xsync_upper_limit) only for stime update, do not
use for change detection.

Problem 1:
If a file created before geo-rep, xtime xattr does not exist.
Geo-rep updates xtime of the file to current time if not exists.
xtime > upper_limit so geo-rep will not pick those files. Changelog
either will have SETXATTR, and fails to sync the file.

Problem 2:
If a file is created before geo-rep create and updated after
geo-rep start. xtime of the file is greater than upper limit(geo-rep
start time/changelog register time). Geo-rep(XSync) will not pick this
file for syncing. Changelog will have only DATA recorded for that file.
Geo-rep tries DATA without any ENTRY ops and fails with rsync error.

BUG: 1200733
Change-Id: Ie4e8f284db689d2c755ef8e7ecbb658db1c0785f
Signed-off-by: Aravinda VK 
Reviewed-on: http://review.gluster.org/9855
Reviewed-by: Kotresh HR 
Reviewed-by: Saravanakumar Arumugam 
Tested-by: Saravanakumar Arumugam 
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

feature/geo-rep: Active Passive Switching logic flock

2015-03-16T04:53:59+00:00

CURRENT DESIGN AND ITS LIMITATIONS:
-----------------------------------
Geo-replication syncs changes across geography using changelogs captured
by changelog translator. Changelog translator sits on server side just
above posix translator. Hence, in distributed replicated setup, both
replica pairs collect changelogs w.r.t their bricks. Geo-replication
syncs the changes using only one brick among the replica pair at a time,
calling it as "ACTIVE" and other non syncing brick as "PASSIVE".

Let's consider below example of distributed replicated setup where
NODE-1 as b1 and its replicated brick b1r is in NODE-2

        NODE-1                         NODE-2
          b1                            b1r

At the beginning, geo-replication chooses to sync changes from NODE-1:b1
and NODE-2:b1r will be "PASSIVE". The logic depends on virtual getxattr
'trusted.glusterfs.node-uuid' which always returns first up subvolume
i.e., NODE-1. When NODE-1 goes down, the above xattr returns NODE-2 and
that is made 'ACTIVE'. But when NODE-1 comes back again, the above xattr
returns NODE-1 and it is made 'ACTIVE' again. So for a brief interval of
time, if NODE-2 had not finished processing the changelog, both NODE-2
and NODE-1 will be ACTIVE causing rename race as mentioned in the bug.

SOLUTION:
---------
1. Have a shared replicated storage, a glusterfs management volume specific
   to geo-replication.

2. Geo-rep creates a file per replica set on management volume.

3. fcntl lock on the above said file is used for synchronization
   between geo-rep workers belonging to same replica set.

4. If management volume is not configured, geo-replication will back
   to previous logic of using first up sub volume.

Each worker tries to lock the file on shared storage, who ever wins will
be ACTIVE. With this, we are able to solve the problem but there is an
issue when the shared replicated storage goes down (when all replicas
goes down). In that case, the lock state is lost. So AFR needs to rebuild the
lock state after brick comes up.

NOTE:
-----
This patch brings in the, pre-requisite step of setting up management volume
for geo-replication during creation.

1. Create mgmt-vol for geo-replicatoin and start it. Management volume should
   be part of master cluster and recommended to be three way replicated
   volume having each brick in different nodes for availability.
2. Create geo-rep session.
3. Configure mgmt-vol created with geo-replication session as follows.
   gluster vol geo-rep  slavenode:: config meta_volume \
   
4. Start geo-rep session.

Backward Compatiability:
-----------------------
If management volume is not configured, it falls back to previous logic of
using node-uuid virtual xattr. But it is not recommended.

Change-Id: I7319d2289516f534b69edd00c9d0db5a3725661a
BUG: 1196632
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/9759
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

geo-rep: Handle ENOENT during cleanup

2015-03-06T02:26:57+00:00

shutil.rmtree was failing to remove file if file was not
exists. Added error handling function to ignore ENOENT if
a file/dir not present.

BUG: 1198101
Change-Id: I1796db2642f81d9e2b5e52c6be34b4ad6f1c9786
Signed-off-by: Aravinda VK 
Reviewed-on: http://review.gluster.org/9792
Tested-by: Gluster Build System 
Reviewed-by: Prashanth Pai 
Reviewed-by: Venky Shankar 
Reviewed-by: Kotresh HR 
Reviewed-by: Saravanakumar Arumugam

geo-rep: Add support for xattrs

2015-03-02T10:05:45+00:00

This patch adds support for xattrs. When it sees SETXATTR
in Changelog, it adds the file to data queue. rsync/tar+ssh
will take care of syncing xattrs. User set xattrs will be
synced to Slave.

New config interface is introduced, sync-xattrs
Which can be set using geo-rep config(Default is True)

gluster volume geo-replication  :: \
               config sync-xattrs false

Change-Id: I70626d854a0d616469dd54d61e5ef155ed8b67d8
BUG: 1196690
Signed-off-by: Aravinda VK 
Reviewed-on: http://review.gluster.org/9499
Tested-by: Gluster Build System 
Reviewed-by: Kotresh HR 
Reviewed-by: Saravanakumar Arumugam 
Reviewed-by: Vijay Bellur

geo-rep: Archive Changelogs and avoid generating empty XSync changelogs

2015-02-20T02:56:06+00:00

With this patch,
- Hybrid Crawl will not generate empty Changelogs
- Archives Changelogs when processed(Hybrid(XSync), History,
  and Changelog Crawl
- Passive worker cleans up its processing directory

BUG: 1169331
Change-Id: I1383ffaed261cdf50da91b14260b4d43177657d1
Signed-off-by: Aravinda VK 
Reviewed-on: http://review.gluster.org/9453
Reviewed-by: Venky Shankar 
Tested-by: Venky Shankar