summaryrefslogtreecommitdiffstats
path: root/xlators/cluster/afr/src/afr-lk-common.c
Commit message (Collapse)AuthorAgeFilesLines
* Revert "cluster/afr: eager locking of FD writes"v3.2.4qa3Vijay Bellur2011-09-221-134/+50
| | | | | | | | | This reverts commit 81456ec2dfb312ae60c5c4e6f960a3cbf8aaaa4c. Change-Id: Id03335117f5137f5d09781850bf4fba6eca0f73d Reviewed-on: http://review.gluster.com/492 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: eager locking of FD writesAnand Avati2011-09-081-50/+134
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch is a change in the way write transactions hold a lock which optimizes the case of sequential writes from a single writer. Lock phase of a transaction has two sub-phases. First is an attempt to acquire locks in parallel by broadcasting non-blocking lock requests. If lock aquistion fails on any server, then the held locks are unlocked and revert to a blocking locked mode sequentially on one server after another. The change in this patch is to make the initial broadcasting lock request attempt to acquire lock on the entire file. If this fails, we revert back to the sequential "regional" blocking lock as before. In the case where such an "eager" lock is granted in the non-blocking phase, it gives rise to an opportunity for optimization. i.e, if the next write transaction on the same FD arrives before the unlock phase of the first transaction, it "takes over" the full file lock. Similarly if yet another transaction arrives before the unlock phase of the "optimized" transaction, that in turn "takes over" the lock as well. The actual unlock now happens at the end of the last "optimzed" transaction. Any operation which arrives before the unlock phase of the previous transaction is a potential candidate to become an "optimized" transaction. In cases where the previous transaction had aquired lock as a "regional" blocking lock, and the next transaction comes in before its unlock phase, then it would not be an "optimized" transaction. Implied assumption ------------------ Since two or more transactions can now operate within the same large lock, there is a possibility that overlapping transactions can arrive at oppoosite orders on the servers. However in the larger picture this is not possible as write-behind already ensures that no two overlapping writes on an inode are in transit at the same time. Overlapping writes across clients are not a problem as they compete at locks anyways. Theoretical benefits and potential harms ---------------------------------------- In case of a single writer: The benefits are large for sequential writes. In the best case the entire file write can happen with just one lock and unlock per server, provided writes are coming in fast enough and getting pipelined by write-behind soon enough (which is usually the case). If the writes are not coming in fast enough, then the optimization "kicks in" for only those subsets of writes which are close enough to get "piggybacked". For random writes the benefits are the same as well. In any case the overall performance is better than or equal to the performance without this optimization for a single writer. In case of multiple writers: When multiple writers are not writing concurrently, there is no negative performance impact. When multiple writers are writing concurrently to the same region, there is no negative impact either, as they were previously getting arbitrated at the locks translator too. In the case of multiple writers writing to different regions concurrently, there will be an increased number of "failovers" from failed parallel non-blocking to sequential blocking regional locks. This above "worst case" has a simple workaround that as soon as we detect > 1 open-fd-count in lookup xattr, we can disable this optimization on those fds. Beneficial side-effects ----------------------- There is another similar optimization in AFR for changelogs which goes by the name of "changelog-piggybacking". That works in a similar way where pending flags get 'taken over' or 'piggybacked' by the next transaction if its 'pre-op' phase kicks in before the 'post-op' phase of the previous transaction. It has been observed that this changelog-piggybacking optimization gives a saving of about ~55% savings of xattr calls hitting the wire, measured across various types of network interfaces. The side effect of this eager-lock optimization is that it gives an almost 100% saving of xattr calls by making the optimistic-changelog work much more efficiently as it gives a wider overlap of the xattr phases of two consecutive transactions. Change-Id: I41c02eb3b64c14c68ef66a344610ec3f024cd59d BUG: 3409 Reviewed-on: http://review.gluster.com/243 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@gluster.com>
* Change Copyright current yearPranith Kumar K2011-08-101-1/+1
| | | | | | | | Change-Id: Id1f1a91cf15d933d5621a0073ddaebe02df0f159 BUG: 3348 Reviewed-on: http://review.gluster.com/198 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@gluster.com>
* LICENSE: s/GNU Affero General Public/GNU General Public/Pranith Kumar K2011-08-061-3/+3
| | | | | | | | Change-Id: Ibf5f45431d7a55b70d7304649af652d6f25bb688 BUG: 3348 Reviewed-on: http://review.gluster.com/183 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@gluster.com>
* cluster/afr: propagate proper errno returned by lock fopsv3.2.1qa5v3.2.1Anand Avati2011-06-101-4/+0
| | | | | | | | | | If locks could not be held on any of the servers, then propagate the errno returned by the lock FOPs instead of hardcoding EAGAIN/EINVAL. Signed-off-by: Anand Avati <avati@gluster.com> BUG: 2993 ([glusterfs-3.2.0qa2]: hang while doing the selfheal) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2993
* cluster/afr: Log errors in afr self-heal with GF_LOG_ERRORPranith Kumar K2011-06-081-1/+1
| | | | | | | | Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Signed-off-by: Anand Avati <avati@gluster.com> BUG: 2986 (Failed operations should should be logged `E' or `W') URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2986
* cluster/afr: Send Non-blocking lock in non-blocking entrylkPranith Kumar K2011-05-311-1/+1
| | | | | | | | Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Signed-off-by: Anand Avati <avati@gluster.com> BUG: 2949 (self-heal hangs) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2949
* removed reference to GF_LOG_NORMALAmar Tumballi2011-04-071-4/+4
| | | | | | | | | | instead used GF_LOG_INFO, which is more standard log level. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Vijay Bellur <vijay@gluster.com> BUG: 2669 (RuntimeError: cannot recognize log level "normal") URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2669
* cluster/afr: log enhancement - part 3Amar Tumballi2011-04-011-51/+37
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 2346 (Log message enhancements in GlusterFS - phase 1) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346
* cluster/afr: white-space cleanup - part 2Amar Tumballi2011-03-311-186/+186
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 2346 (Log message enhancements in GlusterFS - phase 1) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346
* cluster/replicate: fix warnings due to format string mismatches during ↵Raghavendra G2010-12-141-1/+1
| | | | | | | | | | invocation of gf_log. Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2211 ((re)introduce warnings for format string/parameter mismatch) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2211
* check the return value after setting the fd context in afrRaghavendra Bhat2010-12-121-2/+2
| | | | | | | | Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 865 (Add locks recovery support in GlusterFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=865
* replicate: fix hang/missing frame during lockingAnand Avati2010-10-271-1/+15
| | | | | | | | | | | | nonblocking style locking would result in a missing frame when all subvolumes are down or when no subvolume on which fd was opened is up. Check for this condition and unlock gracefully Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 918 (AFR write fails when subvolumes' state is swapped) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=918
* Copyright changesVijay Bellur2010-10-111-1/+1
| | | | | | | | Signed-off-by: Vijay Bellur <vijay@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 971 (dynamic volume management) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=971
* features/locks: cluster/afr: Misc fixes for lock recovery.Pavan Sondur2010-10-051-2/+5
| | | | | | | | Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 865 (Add locks recovery support in GlusterFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=865
* Change GNU GPL to GNU AGPLPranith K2010-10-041-3/+3
| | | | | | | | Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 1388 () URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1388
* Set the lock owner properly for lock self heal.Pavan Sondur2010-10-021-0/+2
| | | | | | | | Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 865 (Add locks recovery support in GlusterFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=865
* Changes to replace flock with gf_flock across GlusterFS.Pavan Sondur2010-10-011-13/+13
| | | | | | | | Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 865 (Add locks recovery support in GlusterFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=865
* cluster/afr: Recover locks on child_up from source to sink.Pavan Sondur2010-10-011-0/+383
| | | | | | | | Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 865 (Add locks recovery support in GlusterFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=865
* protocol/client: cluster/afr: Support lock recovery and self heal.Pavan Sondur2010-09-301-0/+101
| | | | | | | | Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 865 (Add locks recovery support in GlusterFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=865
* replicate: replace first-write-to-flush optimizationAnand V. Avati2010-09-291-4/+0
| | | | | | | | | | | | use a changelog piggybacking optimization instead of first-write-to-flush optimization and do other cleanups (removal of post-post-op hook etc.) Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@amp.gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 1235 (Bug for all pump/migrate commits) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1235
* cluster/afr: Fix hang in create when one subvol is down.Pavan Sondur2010-09-061-17/+26
| | | | | | | | Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 1544 (Create fails when 1 server is down) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1544
* cluster/replicate: fix warnings during build.Raghavendra G2010-09-031-2/+2
| | | | | | | | Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 960 () URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=960
* cluster/afr: Break STACK_WIND loop when the call count is reached.Pavan Sondur2010-08-311-0/+18
| | | | | | | | | | | Fix also has a check for self heal relevant to pump. Tested with dbench with AFR client and pump on server. Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 1443 (Crash in afr_nonblocking_entrylk_cbk) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1443
* cluster/afr: Use 2 phase locking for transactions and self heal.Pavan Sondur2010-08-221-0/+1657
Signed-off-by: Pavan Vilas Sondur <pavan@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 960 () URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=960
generated by cgit (git 2.34.1) at 2024-05-30 08:49:50 +0000