glusterfs.git/xlators/mgmt/glusterd/src/glusterd-utils.h, branch v3.4.0qa5

Revert "glusterd, cli: Task id's for async tasks"

2012-12-04T23:59:52+00:00

This reverts commit ed15521d4e5af2b52b78fd33711e7562f5273bc6

Strangely, the test scripts are "silently" passing for failures too. Reverting patch for now.

Change-Id: I802ec1634c7863dc373cc7dc4a47bd4baa72764e
Reviewed-on: http://review.gluster.org/4267
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

glusterd, cli: Task id's for async tasks

2012-12-04T22:44:36+00:00

This patch introduces task-id's for async tasks like rebalance, remove-brick and
replace-brick. An id is generated for each task when it is started and displayed
to the user in cli output. The status of running tasks is also included in the
output of "volume status" along with its id, so that a user can easily track the
progress of an async task.

Also,
 * added tests for this feature into the regression test suite.
 * added a python script for creating files, 'create-files.py', courtesy
   Vijaykumar Koppad (vkoppad@redhat.com) into the test suite.

Change-Id: Ib0c0d12e0d6c8f72ace48d303d7ff3102157e876
BUG: 857330
Signed-off-by: Kaushal M 
Reviewed-on: http://review.gluster.org/3942
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

mgmt/glusterd: Consider nodesvc to be running after online

2012-11-30T01:34:54+00:00

Definition of online in the message below is that the
RPC_CLNT_CONNECT event arrives for the nfs/self-heal-daemon process.

For automated tests, sometimes the script needs to wait until
self-heal-daemon comes online, so that the relevant
commands can be executed. Gluster volume status before this change
printed whether the self-heal-daemon is running or not based on the
lock availability on the pidfile. But there is a small window where
the lock on pid file is present but the process is still not
online. So the commands that were depending on this kept failing in
the test script.

Change-Id: I0e44e18b08d7b653d34fa170c1f187d91c888cd9
BUG: 858212
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/4236
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

geo-rep / gsyncd,glusterd: do not hardcode socket path

2012-11-29T00:54:07+00:00

... in gsyncd python code. Indeed, use the configuration
mechanism to set it suitably from glusterd.

Change-Id: I9fe2088b14d28588d1e64fe892740cc5755b8365
BUG: 868877
Signed-off-by: Csaba Henk 
Reviewed-on: http://review.gluster.org/4143
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

mgmt/glusterd: Implementation of server-side quorum

2012-11-24T07:14:50+00:00

Feature-page:
http://www.gluster.org/community/documentation/index.php/Features/Server-quorum

Change-Id: I747b222519e71022462343d2c1bcd3626e1f9c86
BUG: 839595
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/3811
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

glusterd: 'volume set' changes for op-version support

2012-10-31T09:35:37+00:00

An op-version check is performed for the given keys during stage. The commit
phase moves the cluster op-version to the required version if needed.

Change-Id: Id5c387094dbec723df736b2ecdc49ff93c179e0e
BUG: 814534
Signed-off-by: Kaushal M 
Reviewed-on: http://review.gluster.org/3780
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

glusterd: volume-start, add-brick and remove-brick to use synctask framework

2012-10-11T19:04:27+00:00

- Added volume-id validation to glusterd-syncop code.
- All daemons are restarted using synctasks in init().
- glusterd_brick_start has wait/nowait variants to support
  volume commands using synctask framework and those that aren't.

Change-Id: Ieec26fe1ea7e5faac88cc7798d93e4cc2b399d34
BUG: 862834
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/3969
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

glusterd: Moved peer rsp handling functions to glusterd-utils

2012-10-11T18:58:33+00:00

- Moved inner functions used in conjunction with synctask, 'out'.

Change-Id: I7fbfd9881ea58645c4295a9fa7163ddd15a45d2f
BUG: 862834
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/4066
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

glusterd: glusterd_brick_stop should be race free wrt pmap

2012-10-11T02:10:50+00:00

This is important for the effort to make glusterd use synctask
framework.

Change-Id: I0affb10a342df99df8daccfd6eef8fa6dd63928c
BUG: 862834
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/4057
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

glusterd: Fix to log command status at the appropriate time

2012-09-20T15:26:22+00:00

PROBLEM:
In the existing implementation, the success/failure of
execution of a command is decided (and logged) in glusterd
handler functions. Strictly speaking, the logging mechanism
must take into account what course the command takes within
the state machine before concluding whether it succeeded or
failed.

FIX:
This patch attempts to fix the above issue for vol commands.
The format of the log message is as follows:
    for failure:
     : FAILED : 
    for success:
     : SUCCESS

APPROACH (in a nutshell):

 * The command string is packed into dict at cli and sent to
   glusterd.
 * glusterd logs the command status just before doing a
   "submit_reply", which is called (either directly or
   indirectly via a call to glusterd_op_cli_send_response)
   at 2 places for every vol command:
   i. in handler functions, and
   ii. in glusterd_op_txn_complete

In short, the failure of a command in the handler implies the
command has indeed failed. However, its success in the handler
does NOT necessarily mean the command succeeded/will succeed.

Change-Id: I5a8a2ddc318ef2dc2a9699f704a6bcd2f0ab0277
BUG: 823081
Signed-off-by: Krutika Dhananjay 
Reviewed-on: http://review.gluster.org/3948
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati