summaryrefslogtreecommitdiffstats
path: root/xlators/features/sdfs
Commit message (Collapse)AuthorAgeFilesLines
* graph/shd: Use top down approach while cleaning xlatorMohammed Rafi KC2019-06-271-0/+1
| | | | | | | | | | | | | | We were cleaning xlator from botton to top, which might lead to problems when upper xlators trying to access the xlator object loaded below. One such scenario is when fd_unref happens as part of the fini call which might lead to calling the releasedir to lower xlator. This will lead to invalid mem access Change-Id: I8a6cb619256fab0b0c01a2d564fc88287c4415a0 Updates: bz#1716695 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
* across: clang-scan: fix NULL dereferencing warningsAmar Tumballi2019-06-041-1/+2
| | | | | | | | | All these checks are done after analyzing clang-scan report produced by the CI job @ https://build.gluster.org/job/clang-scan updates: bz#1622665 Change-Id: I590305af4ceb779be952974b2a36066ffc4865ca Signed-off-by: Amar Tumballi <amarts@redhat.com>
* features/sdfs: Assign unique lk-owner for entrylk fopPranith Kumar K2019-04-221-0/+6
| | | | | | | | | | | | | sdfs is supposed to serialize entry fops by doing entrylk, but all the locks are being done with all-zero lk-owner. In essence sdfs doesn't achieve its goal of mutual exclusion when conflicting operations are executed by same client because two locks on same entry with same all-zero-owner will get locks. Fixed this up by assigning lk-owner before taking entrylk updates bz#1624701 Change-Id: Ifabfc998c9f1724915d38e90ed8287e05797d769 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* features/sdfs: disable by defaultAmar Tumballi2019-01-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the feature enabled, some of the performance testing results, specially those which create millions of small files, got approximately 4x regression compared to version before enabling this. On master without this patch: 765 creates/sec On master with this patch : 3380 creates/sec Also there seems to be regression caused by this in 'ls -l' workload. On master without this patch: 3030 files/sec On master with this patch : 16610 files/sec This is a feature added to handle multiple clients parallely operating (specially those which race for file creates with same name) on a single namespace/directory. Considering that is < 3% of Gluster's usecase right now, it makes sense to disable the feature by default, so we don't penalize the default users who doesn't bother about this usecase. Also note that the client side translators, specially, distribute, replicate and disperse already handle the issue upto 99.5% of the cases without SDFS, so it makes sense to keep the feature disabled by default. Credits: Shyamsunder <srangana@redhat.com> for running the tests and getting the numbers. Change-Id: Iec49ce1d82e621e9db25eb633fcb1d932e74f4fc Updates: bz#1670031 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* xlator: make 'xlator_api' mandatoryAmar Tumballi2018-12-131-3/+14
| | | | | | | | | | | | | | * Remove the options to load old symbol. * keep only 'xlator_api' symbol from being exported using xlator.sym * add xlator_api to all the xlators where its missing NOTE: This covers all the xlators which has at least a test case to validate its loading. If there is a translator, which doesn't have any test, then we should probably remove that from codebase. fixes: #164 Change-Id: Ibcdc8c9844cda6b4463d907a15813745d14c1ebb Signed-off-by: Amar Tumballi <amarts@redhat.com>
* libglusterfs: Move devel headers under glusterfs directoryShyamsundarR2018-12-052-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | libglusterfs devel package headers are referenced in code using include semantics for a program, this while it works can be better especially when dealing with out of tree xlator builds or in general out of tree devel package usage. Towards this, the following changes are done, - moved all devel headers under a glusterfs directory - Included these headers using system header notation <> in all code outside of libglusterfs - Included these headers using own program notation "" within libglusterfs This change although big, is just moving around the headers and making it correct when including these headers from other sources. This helps us correctly include libglusterfs includes without namespace conflicts. Change-Id: Id2a98854e671a7ee5d73be44da5ba1a74252423b Updates: bz#1193929 Signed-off-by: ShyamsundarR <srangana@redhat.com>
* Land part 2 of clang-format changesGluster Ant2018-09-121-1126/+1062
| | | | | Change-Id: Ia84cc24c8924e6d22d02ac15f611c10e26db99b4 Signed-off-by: Nigel Babu <nigelb@redhat.com>
* Land clang-format changesGluster Ant2018-09-122-32/+30
| | | | Change-Id: I6f5d8140a06f3c1b2d196849299f8d483028d33b
* features/sdfs: Minor improvementsVijay Bellur2018-08-211-3/+3
| | | | | | | | | | | | | | | | Addresses: CID 1389688: Prevent op_errno from being -ve in sdfs_rename() CID 1389286: Check for retval from loc_copy() to avoid logically dead code and Initializes op_errno to ENOMEM in sdfs_link() Change-Id: I12e17a98faa5887da94a33ba9ca775e8e0fef359 updates: bz#789278 Signed-off-by: Vijay Bellur <vbellur@redhat.com>
* sdfs: Fix missing NULL option list treminationShyamsundarR2018-07-251-0/+1
| | | | | | | | | | | | | Option list for volume_options in sdfs was not NULL terminated. This resulted in a crash when running in lcov based builds. This is rectified by this patch. fixes: bz#1608566 Change-Id: I5d8730f1ae963ed6adf21d970e4921c5d5d92f62 Signed-off-by: ShyamsundarR <srangana@redhat.com>
* sdfs: enable by defaultAmar Tumballi2018-05-241-0/+25
| | | | | | | | also provide an option for pass-through to enable/disable xlator fixes: #421 Change-Id: Ie30a91ad09620db62ab07b797e23123fd1200d1f Signed-off-by: Amar Tumballi <amarts@redhat.com>
* features/sdfs: implement readdirpRaghavendra G2018-05-241-3/+144
| | | | | | | | | | Since readdirp acts as a batched lookup for all dentries it reads, it has to synchronize with any entry operation within the directory being read. Change-Id: I923a6ebd21856dbaa5fa5db4a26a29b7b29b3159 Signed-off-by: Raghavendra G <rgowdapp@redhat.com> fixes: #421
* build: add --without-server optionNiels de Vos2018-02-191-0/+2
| | | | | | | | | | | | | | | | With Gluster 4.0 we will not provide the server components for EL6 and older. At one point Gluster 4.x will get GlusterD2, which requires Golang tools in the distribution. EL6 does not contain these at the moment. With this change, it is possible to `./configure --without-server` which prevents building glusterd and the xlators for the bricks. Building RPMs can pass `--without server` and the glusterfs-server sub-package will not be created. Change-Id: I97f5ccf9f2c76e60d9af83915fc59fae57ad6d25 BUG: 1074947 Signed-off-by: Niels de Vos <ndevos@redhat.com>
* sdfs: crash fixesAmar Tumballi2018-02-011-8/+14
| | | | | | | | | | | | | * from the patch which got tested in experimental branch, there was a code cleanup involved, which missed setting of a local variable, which led to crash immediately after enabling the feature. * added a sanity test case to validate all the fops of sdfs. Updates: #397 Change-Id: I7e0bebfc195c344620577cb16c1afc5f4e7d2d92 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* dentry fop serializer: added new server side xlator for dentry fop serializationSakshi Bansal2018-01-245-0/+1489
Problems addressed by this xlator : [1]. To prevent race between parallel mkdir,mkdir and lookup etc. Fops like mkdir/create, lookup, rename, unlink, link that happen on a particular dentry must be serialized to ensure atomicity. Another possible case can be a fresh lookup to find existance of a path whose gfid is not set yet. Further, storage/posix employs a ctime based heuristic 'is_fresh_file' (interval time is less than 1 second of current time) to check fresh-ness of file. With serialization of these two fops (lookup & mkdir), we eliminate the race altogether. [2]. Staleness of dentries This causes exponential increase in traversal time for any inode in the subtree of the directory pointed by stale dentry. Cause : Stale dentry is created because of following two operations: a. dentry creation due to inode_link, done during operations like lookup, mkdir, create, mknod, symlink, create and b. dentry unlinking due to various operations like rmdir, rename, unlink. The reason is __inode_link uses __is_dentry_cyclic, which explores all possible path to avoid cyclic link formation during inode linkage. __is_dentry_cyclic explores stale-dentry(ies) and its all ancestors which is increases traversing time exponentially. Implementation : To acheive this all fops on dentry must take entry locks before they proceed, once they have acquired locks, they perform the fop and then release the lock. Some documentation from email conversation: [1] http://www.gluster.org/pipermail/gluster-devel/2015-December/047314.html [2] http://www.gluster.org/pipermail/gluster-devel/2015-August/046428.html With this patch, the feature is optional, enable it by running: `gluster volume set $volname features.sdfs enable` Also the feature is tested for a month without issues in the experiemental branch for all the regression. Change-Id: I6e80ba3cabfa6facd5dda63bd482b9bf18b6b79b Fixes: #397 BUG: 1304962 Signed-off-by: Sakshi Bansal <sabansal@redhat.com> Signed-off-by: Amar Tumballi <amarts@redhat.com> Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
generated by cgit (git 2.34.1) at 2024-04-30 12:46:31 +0000