summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* glusterd: Handling eof during gf_store_read_and_tokenizeAvra Sengupta2014-02-041-4/+4
| | | | | | | | | | | Change-Id: I88eeffd64c521db78b9e7539ae57ed046854cb63 BUG: 1049278 Signed-off-by: Avra Sengupta <asengupt@redhat.com> Reviewed-on: http://review.gluster.org/6903 Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-by: Sachin Pandit <spandit@redhat.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Tested-by: Rajesh Joseph <rjoseph@redhat.com>
* glusterd/snapshot : snapshot delete - check if process is killed before ↵Sachin Pandit2014-02-042-3/+23
| | | | | | | | | | | | | unmounting. Change-Id: Idf0cf63429212142795e1aeb4fd4962b51620426 BUG: 1049353 Signed-off-by: Sachin Pandit <spandit@redhat.com> Reviewed-on: http://review.gluster.org/6772 Reviewed-by: Vijaikumar Mallikarjuna <vmallika@redhat.com> Reviewed-by: Avra Sengupta <asengupt@redhat.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Tested-by: Rajesh Joseph <rjoseph@redhat.com>
* mgmt/glusterd: perform store on only newly created snapshotRaghavendra Bhat2014-02-023-13/+103
| | | | | | | | | | | | | | | * Instead performing store on all the snaps of the volume when a new snap is created, store only the new snap's info. Otherwise with more and more snapshots in the volume, time creation for new snapshot becomes very large as glusterd has to store all the snapshots related info everytime (that too by doing fsyncs) Change-Id: I0e005d1d4c044b07a8abde8e6ba55e66a1bbd590 BUG: 1059146 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/6841 Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Tested-by: Rajesh Joseph <rjoseph@redhat.com>
* glusterd/vol-locks: Dict_ref while adding req_ctx->dict to txn_opinfosAvra Sengupta2014-01-217-83/+71
| | | | | | | | | | | Introducing a wrapper function glusterd_txn_opinfo_init(), to initialize the opinfo to be set in the txn_id engine. Removed glusterd_op_fini_ctx() as the txn opinfo should only be cleared by glusterd_clear_txn_opinfo(). Change-Id: I17e85a162d6a3bca79941f8603d0c2b579f0d194 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* Merge "glusterd/vol-locks: Releasing vol-locks on receiving peer disconnect" ↵Rajesh Joseph2014-01-211-0/+14
|\ | | | | | | into development
| * glusterd/vol-locks: Releasing vol-locks on receiving peer disconnectAvra Sengupta2014-01-211-0/+14
| | | | | | | | | | | | | | | | On receiving peer disconnect, release volume locks held by the disconnected peer Change-Id: I0e36ba2fba4761225092ce2fb7b9b3c2d2dca0ad Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* | glusterd/snapshot: Move out code from prevalidate functionVijaikumar M2014-01-211-156/+225
|/ | | | | | | | Move the code part that need to executed at the origin glusterd to a seperate function. Change-Id: Ieb53fbd4fb12e3e4d5087db306be37616657ee04 Signed-off-by: Vijaikumar M <vmallika@redhat.com>
* Merge "protocol/server: get the client object from frame->root even in ↵Rajesh Joseph2014-01-202-3/+1
|\ | | | | | | barrier" into development
| * protocol/server: get the client object from frame->root even in barrierRaghavendra Bhat2014-01-152-3/+1
| | | | | | | | | | | | | | * Remove the reference to client object from server_state_t structure Change-Id: Ic2dbf9dd487f1177bd889fe481e63bc063fd6183 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* | glusterd/snapshot : Making snap time persistent.Sachin Pandit2014-01-161-11/+61
|/ | | | | Change-Id: I2d07717ee01751e481571ce420b0f84711ea9597 Signed-off-by: Sachin Pandit <spandit@redhat.com>
* Merge "protocol/server: copy the response into the payload instead of using ↵Rajesh Joseph2014-01-153-3/+7
|\ | | | | | | the pointer" into development
| * protocol/server: copy the response into the payload instead of using the pointerRaghavendra Bhat2014-01-133-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The response structure filled up in server_submit_reply is local to the function (i.e stack allocated) whose address is stored in the barrier payload to use later while sending the replies. But after the function is exited (server_submit_reply) the pointer is not valid and contains invalid data, which either leads in a segfault due to illegal memory access or reply not being sent as the total length of the reply obtained from that memory might not be valid. So instead of saving the pointer inside the payload, save the complete reply itself. Change-Id: I7d0b7b181584865199357a67165b99bf35def5ab Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* | Merge "cli: fix displaying of different delete message" into developmentRajesh Joseph2014-01-151-1/+1
|\ \
| * | cli: fix displaying of different delete messageRaghavendra Bhat2014-01-071-1/+1
| | | | | | | | | | | | | | | | | | Change-Id: Ife2395a92997168bb147a7db4bba346d3adc916b BUG: 1048126 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* | | mgmt/glusterd : Having a separate list for snapshots.Sachin Pandit2014-01-157-9/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | Creating a separate list for snaps taken, as cluttering snaps in the volume list does not look neat. Change-Id: Ida4a183e95e8694b85ebb5a680d06b7d29a460a0 BUG: 1040947 Signed-off-by: Sachin Pandit <spandit@redhat.com>
* | | glusterd/snapshot : Update the parent volume name of snap during restore.Sachin Pandit2014-01-151-0/+1
| | | | | | | | | | | | | | | | | | Change-Id: I9663342e49073a172667220d4839b2beb65ffc0e BUG: 1040947 Signed-off-by: Sachin Pandit <spandit@redhat.com>
* | | Merge "glusterd/snapshot: A Warning must be displayed before the snapshot is ↵Rajesh Joseph2014-01-152-9/+40
|\ \ \ | | | | | | | | | | | | deleted" into development
| * | | glusterd/snapshot: A Warning must be displayed before the snapshot is deletedVijaikumar M2014-01-152-9/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, on deleting snapshot , there is no warning message displayed to the user. A Warning must be displayed before the snapshot is deleted as follows: Deleting snap will erase all information about the snap. Do you want to continue? (y/n) Change-Id: I60f4f346da36b6564576c7482f5989a17d4bb5e1 BUG: 1048216 Signed-off-by: Vijaikumar M <vmallika@redhat.com>
* | | | Snapshot: Minor change to fix last merge conflictRajesh Joseph2014-01-152-1/+1
|/ / / | | | | | | | | | | | | | | | | | | Some changes were missed in resolving conflict in the restore patch. Change-Id: Ie0cf01237bf975027056c10bb3681e122334f83e Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
* | | Merge "glusterd: Keeping snapshot delete error messages consistent with that ↵Rajesh Joseph2014-01-151-7/+8
|\ \ \ | | | | | | | | | | | | of vol delete." into development
| * | | glusterd: Keeping snapshot delete error messages consistent with that of vol ↵Avra Sengupta2014-01-151-7/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | delete. Change-Id: I5056c0e46015278bd8d0c39307e4671fd4259961 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* | | | Merge "CLI/snapshot : Aligning the snapshot list output." into developmentRajesh Joseph2014-01-151-19/+47
|\ \ \ \
| * | | | CLI/snapshot : Aligning the snapshot list output.Sachin Pandit2014-01-151-19/+47
| | |_|/ | |/| | | | | | | | | | | | | | Change-Id: Id98d6e2d0436621486c311889f128077558e59f8 Signed-off-by: Sachin Pandit <spandit@redhat.com>
* | | | Merge "cli: check the description length before appending it to the buffer" ↵Rajesh Joseph2014-01-151-35/+29
|\ \ \ \ | | | | | | | | | | | | | | | into development
| * | | | cli: check the description length before appending it to the bufferRaghavendra Bhat2014-01-071-35/+29
| | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I34cf0f2ee42074bb40edda875e0b290069d856ae Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* | | | | Snapshot: Gluster snapshot restore featureRajesh Joseph2014-01-158-249/+739
| |_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Implemented gluster snapshot restore feature. The restore is done by replacing the origin volume with the snap volume. TODO: After the restore the snapshot volume should be deleted. As of now the deletion work is pending. Change-Id: Ib137fb6bb84a74030607ffa47f89cd705dc7e1ff Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
* | | | glusterd/snapshot : Making CG Name persistent and handle Snap Description Issue.Sachin Pandit2014-01-134-19/+78
| |/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1)Write the CG Name in the "snap_list.info" file, so that CG Name is not lost when we restart the glusterd. 2)Fixes the issue where Description given for CG, as a part of create command, was getting stored as snap description rather than CG Description. 3)Fixes the problem with glusterd restart when we have multiple words in Snap Descripition Change-Id: I3129c53d1ec54dd170ca1300583f278f58c4e0e2 BUG: 1044476 Signed-off-by: Sachin Pandit <spandit@redhat.com>
* | | glusterd: Pass generated snap-name/cg-name to cli.Avra Sengupta2014-01-092-9/+53
| | | | | | | | | | | | | | | | | | | | | | | | Also block operation on response aggregation errors. Change-Id: I641ddc98d986fd44627d790ab84488a6b849f807 BUG: 1043792 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* | | glusterd/multiple volume locks: Fix for lock requests received in synctasked ↵Avra Sengupta2014-01-092-30/+48
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | volume locks The synctasked volume lock function was locking every node with it's own UUID, instead of received UUID, which resulted in deadlock when multiple snapshot commands were executed. Change-Id: Ida76da1a057eae04178c202d626f9c1e671c1fee BUG: 1043862 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* | | glusterd/snapshot: Updating snapshot options during glusterd-handshake.Avra Sengupta2014-01-092-5/+91
| |/ |/| | | | | | | Change-Id: Icdd20825f51a01f4186841639fb645cb56a2fd12 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* | glusterd/snapshot: Defining snap-max-soft-limit as a percentage of ↵Avra Sengupta2014-01-077-189/+390
|/ | | | | | | | | | | | | snap-max-hard-limit. This patch also prohibits configuration of snap-max-hard-limit and snap-max-soft-limit for snap volumes. Also displaying the snapshot configs by reading data only from local node, as all config data will be in sync across the cluster. Change-Id: I635b925c02ed5b108cd10c7193b154ad82d5afad BUG: 1043792 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* Merge "glusterd/snapshot: "gluster help" to list the usage for snapshot ↵Rajesh Joseph2014-01-061-1/+2
|\ | | | | | | commands" into development
| * glusterd/snapshot: "gluster help" to list the usage for snapshot commandsVijaikumar M2014-01-061-1/+2
| | | | | | | | | | | | | | | | | | Currently "gluster help" only list the volume and peer command family. With this fix it will also list the usage for snapshot commands. Change-Id: I75601496e4f027250c2381eff94e4283bb98299a BUG: 1043472 Signed-off-by: Vijaikumar M <vmallika@redhat.com>
* | glusterd/snapshot: Introducing snap-max-hard-limit and snap-max-soft-limitAvra Sengupta2014-01-0611-334/+550
| | | | | | | | | | | | | | | | | | Note: Manually adding this patch again as this patch got missed in git reset option done on remote development branch Change-Id: I9e81c5ec003c1e1722d0fcb27dd87c365ee43ff4 Signed-off-by: Avra Sengupta <asengupt@redhat.com> Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
* | glusterd/snapshot : Fix for CG ID and Name not getting displayed.Sachin Pandit2014-01-062-8/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | CG ID was not getting initiated during snapshot create, hence there was problem in listing the CG ID and CG Name. Note: Manually adding this patch again as this patch got missed in git reset option done on remote development branch Change-Id: I81951b42292912c98bab5964fc732b630ff66d14 BUG: 1040435 Signed-off-by: Sachin Pandit <spandit@redhat.com> Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
* | glusterd/snapshot : Making snap uuid same across different nodes.Sachin Pandit2014-01-061-3/+1
|/ | | | | | | | | | | | | | Till now we were generation a new uuid for a snap, that uuid was brick id, so the snap uuid was different for different nodes. Now we are copying the snap volume id to snap uuid so that it is unique across nodes. Note: Manually adding this patch again as this patch got missed in git reset option done on remote development branch Change-Id: Icb80caca9672b82de1b2129af47540c651a034ec Signed-off-by: Sachin Pandit <spandit@redhat.com> Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
* glusterd/snapshot: Fixing glusterd_do_quorum_action to start snap bricks for ↵Avra Sengupta2013-12-171-7/+29
| | | | | | | snap volumes Change-Id: I97f59fcf1e78ded35fd15996d587ecd043c7dc17 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* glusterd/snapshot: Fix for cksum mismatches at snap create.Avra Sengupta2013-12-173-11/+84
| | | | | | | Also fixes peer rejects on glusterd restart Change-Id: I1671416c1f3fd2afea450cc3b4c632de187351ca Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* cli/glusterd: implement the snap and cg delete functionalitiesRaghavendra Bhat2013-12-128-140/+1258
| | | | | Change-Id: Icdb66c89acdd043d0d6368c48ce2e01b1a40966f Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* mgmt/glusterd: snapshot related cleanupRaghavendra Bhat2013-12-122-24/+47
| | | | | Change-Id: I277a70f732666d047ba5dff7a7e6925e0679741b Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* glusterd/snapshot: Fix Displaying Port, Online Status and Pid for snap vols ↵Avra Sengupta2013-12-105-2/+36
| | | | | | | | | | | in volume status Added a parent_volname member in glusterd_volinfo_ structure to help point the snap vol to the parent volname. Using this to fetch the pidfile location during volume status. Change-Id: I30a16646561394d0f7d16f66abff14c425f31f06 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* glusterd/snap: Fix for socket connect failures on brick startAvra Sengupta2013-12-102-33/+34
| | | | | Change-Id: I9f53bd4d83794c69c54e4a03f59425a1ca6a4ac3 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* mgmt/glusterd: set the snap count properly while restoringRaghavendra Bhat2013-12-031-2/+3
| | | | | Change-Id: I37eb7ab12767fdd11aa2e58441d26e6d6d9dd245 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* glusterd/snapshot: Fixes for cg name space collisionsAvra Sengupta2013-12-031-18/+11
| | | | | Change-Id: I6137ecb0dbbd37af81433c415568dfede1d740a6 Signed-off-by: Avra Sengupta <asengupt@redhat.com>
* Merge "mgmt/glusterd: use runner apis to mount the snap device" into developmentRajesh Joseph2013-12-031-2/+17
|\
| * mgmt/glusterd: use runner apis to mount the snap deviceRaghavendra Bhat2013-12-031-2/+17
| | | | | | | | | | | | | | | | Instead of using mount api to mount the snap device, use runner apis to execute the mount command which mounts the snap device. Change-Id: I8cf1fd4ade44d35dcd48c93fab10f5bea7f96f73 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* | glusterd/mgmtv3: Fetching snap mount path per brick.Avra Sengupta2013-12-033-144/+576
|/ | | | | | | | | | | | | | | | | Fetching snap mount path per brick during pre-validate and aggregating it before commit to generate all brick volfiles and all client volfiles on all nodes. Added changes to propagate the same cg-id by generating it in pre-validate and sending the same to all the nodes during commit Also fixed volume-id mismatches caused in snap-volumes leading to failures of volume status Change-Id: I0b8ed81ac93077cd50313e6eeed00f71932cf2ac Signed-off-by: Avra Sengupta <asengupt@redhat.com> Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
* mgmt/glusterd: create pid file for the snap bricks in proper pathRaghavendra Bhat2013-12-031-1/+1
| | | | | Change-Id: I98fcd25e80b1b39e0292eae059e2d624ca9094fb Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* mgmt/glusterd: handle issues present in snapshot create and snap managementRaghavendra Bhat2013-12-028-294/+322
| | | | | Change-Id: I94b5f6e00be7d1ff0c454e291c779dae7b423748 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* Merge "snapshot: setting --setactivationskip as 'n' during lvcreate" into ↵Rajesh Joseph2013-12-021-2/+2
|\ | | | | | | development