summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* protocol: utilize the version 4 xdrv4.0.0rc0Amar Tumballi2018-02-018-152/+581
| | | | | | | updates #384 Change-Id: Id80bf470988dbecc69779de9eb64088559cb1f6a Signed-off-by: Amar Tumballi <amarts@redhat.com>
* gfapi : New APIs have been added to use lease feature in glusterSoumya Koduri2018-02-0111-29/+280
| | | | | | | | | | | Following APIs glfs_h_lease(), glfs_lease() added, so that gfapi applications can set and get lease which enables more efficient client side caching. Updates: #350 Change-Id: Iede85be9af1d4df969b890d0937ed0afa4ca6596 Signed-off-by: Poornima G <pgurusid@redhat.com> Signed-off-by: Soumya Koduri <skoduri@redhat.com> Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
* glusterd: optimize glusterd import volumes code pathAtin Mukherjee2018-01-311-5/+7
| | | | | | | | | | | | | In case there's a version mismatch detected for one of the volumes glusterd was ending up with updating all the volumes which is a overkill. >mainline patch : https://review.gluster.org/#/c/19358/ Change-Id: I6df792db391ce3a1697cfa9260f7dbc3f59aa62d BUG: 1540554 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> (cherry picked from commit bb34b07fd2ec5e6c3eed4fe0cdf33479dbf5127b)
* core: remove experimental xlators and associated testsKaleb S. KEITHLEY2018-01-3157-6032/+2
| | | | | | | | | | | | experimental xlators removed from 4.0 > Cherry picked from 4231c40973c60999f5ef759db450d25e129ef6ba: > Reviewed-on: https://review.gluster.org/17953 > Reviewed-by: jiffin tony Thottan <jthottan@redhat.com> Signed-off-by: ShyamsundarR <srangana@redhat.com> Change-Id: I34419ce22ca09b7626b8f9382c377a614fd9fed8 BUG: 1539842
* protocol: Remove lock recovery logic from client and serverAnoop C S2018-01-3111-924/+36
| | | | | | | Change-Id: I27f5e1e34fe3eac96c7dd88e90753fb5d3d14550 BUG: 1540438 Signed-off-by: Anoop C S <anoopcs@redhat.com> (cherry picked from commit 3e78ea991b213422fc423ff94994e1eb295569c7)
* build: use libtirpc by default, even if ipv6 is not the defaultKaleb S. KEITHLEY2018-01-303-3/+4
| | | | | | | | | | | | | | Another error snuck in with Change-Id I86f847dfd, or more accurately I think, with Change-Id: Ic47065e9c2... All libs, not just libgfrpc, need to be linked with libtirpc, especially on systems that still have xdr functions in (g)libc where you will get a mixture of calls to libtirpc functions and glibc functions, with catas Change-Id: I97dc39c7844f44c36fe210aa813480c219e1e415 BUG: 1538723 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
* geo-rep: Remove unused working directory check in gsyncdAravinda VK2018-01-301-37/+1
| | | | | | | | | | | | To append the default config file path, gsyncd calls gluster command to get the workdir path and constructs config file path. This is not required now since the Config management in Geo-replication is changed with patch 18257(Issue #73) BUG: 1539701 Change-Id: Ia7eb39e36ed59ece4de65ea7ec71a0f615e338bb Signed-off-by: Aravinda VK <avishwan@redhat.com> (cherry picked from commit 596143a286079338c30c8abc4b862f8d6ddddb3a)
* dentry fop serializer: added new server side xlator for dentry fop serializationSakshi Bansal2018-01-3011-4/+1532
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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 Signed-off-by: Sakshi Bansal <sabansal@redhat.com> Signed-off-by: Amar Tumballi <amarts@redhat.com> Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
* doc: Initial draft of 4.0.0 release notesShyamsundarR2018-01-291-0/+328
| | | | | | | | | | | This is an initial draft, that should render right in github. Need inputs from contributors to complete the notes. Change-Id: I0e6a4f88b44ff46685e66f139758a2247954639d BUG: 1539842 Signed-off-by: ShyamsundarR <srangana@redhat.com>
* geo-rep: Detailed JSON output for configAravinda VK2018-01-293-10/+45
| | | | | | | | | | | | | | | | | | JSON output of `config-get` command now returns in the following format { "name": CONFIG_NAME, "value": CONFIG_VALUE, "default_value": DEFAULT_VALUE, # Only if modified == true "configurable": true|false, "modified": true|false } Change-Id: I6193de48cd33655df7ecef5a0d83d7cb147089cf Fixes: #361 Signed-off-by: Aravinda VK <avishwan@redhat.com> (cherry picked from commit 5d3c90d14839fc134580a4ff944d68f3ff3c605a)
* tests: Disable geo-rep testsNigel Babu2018-01-292-0/+2
| | | | | | | | | | | | | | These tests are prone to issues at the moment that need further debugging and fixing. > BUG: 1537602 > Change-Id: Ic59ca620925c6f43948b8a751eaddb571b791969 > Signed-off-by: Nigel Babu <nigelb@redhat.com> (cherry picked from commit 5cbe21b1ddf6eae4782afa33ce841fc4aa195974) Change-Id: I5ccb7751203526082a6f71fa96cff47b56372bf7 BUG: 1539657 Signed-off-by: Nigel Babu <nigelb@redhat.com>
* Update rfc.sh for release-4.0v4.0.0alphaShyamsundarR2018-01-231-1/+1
| | | | Signed-off-by: ShyamsundarR <srangana@redhat.com>
* libglusterfs: Reset errno before callv4.1devNigel Babu2018-01-231-1/+4
| | | | | | | | This was causing Gluster to return a failure when testing on Centos7. BUG: 1536913 Change-Id: Idb90baef05058123a7f69e94a51dd79abd371815 Signed-off-by: Nigel Babu <nigelb@redhat.com>
* features/error-gen: update volume options for GD2Atin Mukherjee2018-01-231-1/+11
| | | | | | | Updates #302 Change-Id: I8d04bb4f46b1e53a8d176a75da2382fdbf0cfc6f Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* geo-rep: Support for using Volinfo from Conf fileAravinda VK2018-01-234-50/+128
| | | | | | | | | | | | | | | | | | | | | | | | | | | Once Geo-replication is started, it runs Gluster commands to get Volume info from Master and Slave. With this patch, Georep can get Volume info from Conf file if `--use-gconf-volinfo` argument is specified to monitor Create a config(Or add to the config if exists) with following fields [vars] master-bricks=NODEID:HOSTNAME:PATH,.. slave-bricks=NODEID:HOSTNAME,.. master-volume-id= slave-volume-id= master-replica-count= master-disperse_count= Note: Exising Geo-replication is not affected since this is activated only when `--use-gconf-volinfo` is passed while spawning `gsyncd monitor` Tiering support is not yet added since Tiering + Glusterd2 is still under discussion. Fixes: #396 Change-Id: I281baccbad03686c00f6488a8511dd6db0edc57a Signed-off-by: Aravinda VK <avishwan@redhat.com>
* performance/read-ahead: volume option fixes for GD2Csaba Henk2018-01-221-0/+7
| | | | | | | Updates #302 Change-Id: I842ab746502bc4ff3b312ee19f397a6177edc6f7 Signed-off-by: Csaba Henk <csaba@redhat.com>
* protocol/server: Fix "auth-path" in options tableKotresh HR2018-01-221-0/+5
| | | | | | | | | | | Options set will crash the brick with glusterd2. glusterd used to set "auth-path" during volfile generation. With glusterd2, it should come from options table. Updates: #302 Change-Id: Ie41a17779c185b87ace7e0bce0d0ba594b415a75 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* performance/write-behind: volume option fixes for GD2Csaba Henk2018-01-221-0/+18
| | | | | | | Updates #302 Change-Id: I2ade3dc4aef4da619c5b64058872594f0f1e004f Signed-off-by: Csaba Henk <csaba@redhat.com>
* libgfapi: Add new api for supporting mandatory-locksAnoop C S2018-01-2210-36/+694
| | | | | | | | | | | | | | | | The current API for byte-range locks [glfs_posix_lock()] doesn't allow applications to specify whether it is advisory or mandatory type locks. This particular change is to introduce an extended byte-range lock API with an additional argument for including the byte-range lock mode to be one among advisory(default) or mandatory. Patch also includes a gfapi test case which make use of this new api to acquire mandatory locks. Ref: https://github.com/gluster/glusterfs-specs/blob/master/done/GlusterFS%203.8/Mandatory%20Locks.md Change-Id: Ia09042c755d891895d96da857321abc4ce03e20c Updates #393 Signed-off-by: Anoop C S <anoopcs@redhat.com>
* cluster/ec : EC options for GD2Sunil Kumar Acharya2018-01-221-1/+34
| | | | | | | Updates #302 Change-Id: I31b4648f7b1a394fceece5cba8120c579c66edd9 Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
* geo-rep: Improve geo-rep pre-validation logsKotresh HR2018-01-221-4/+5
| | | | | | | | | | | | | | | | | | | | | | Geo-rep runs gverify.sh which does pre-validation. As part of it, master and slave volume is mounted to verify the size. If for some reason, the mount fails, the error message does not point out the mount log file location. Also both master and slave mount logs are same. Patch does following improvements. 1. Master and slave mount logs are separated and error message points the log file to be looked for. 2. The log location is changed to /var/log/glusterfs/geo-replication instead of /var/log/glusterfs/geo-replication-slaves 3. The log file name is changed to "gverify-mastermnt.log" and "gverify-slavemnt.log" for master and slave mount respectively Fixes: #395 Change-Id: Ia644ec0afebbdaae92e01adf03c635e5f8866a02 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* md-cache: Implement dynamic configuration of xattr list for cachingPoornima G2018-01-226-277/+188
| | | | | | | | | | | | | | | Currently, the list of xattrs that md-cache can cache is hard coded in the md-cache.c file, this necessiates code change and rebuild everytime a new xattr needs to be added to md-cache xattr cache list. With this patch, the user will be able to configure a comma seperated list of xattrs to be cached by md-cache Updates #297 Change-Id: Ie35ed607d17182d53f6bb6e6c6563ac52bc3132e Signed-off-by: Poornima G <pgurusid@redhat.com>
* protocol: make on-wire-change of protocol using new XDR definition.Amar Tumballi2018-01-1919-879/+18796
| | | | | | | | | | | | | | | | | | | | | | With this patchset, some major things are changed in XDR, mainly: * Naming: Instead of gfs3/gfs4 settle for gfx_ for xdr structures * add iattx as a separate structure, and add conversion methods * the *_rsp structure is now changed, and is also reduced in number (ie, no need for different strucutes if it is similar to other response). * use proper XDR methods for sending dict on wire. Also, with the change of xdr structure, there are changes needed outside of xlator protocol layer to handle these properly. Mainly because the abstraction was broken to support 0-copy RDMA with payload for write and read FOP. This made transport layer know about the xdr payload, hence with the change of xdr payload structure, transport layer needed to know about the change. Updates #384 Change-Id: I1448fbe9deab0a1b06cb8351f2f37488cefe461f Signed-off-by: Amar Tumballi <amarts@redhat.com>
* gfapi : added glfs_setfsleaseid() for setting lease idSoumya Koduri2018-01-195-0/+32
| | | | | | | | | | | | A new function glfs_setfsleaseid() added in gfapi. Currently lock owner is saved in the thread context. Similarly the leaseid attribute can be saved using glfs_setfsleaseid(). Updates: #350 Change-Id: I55966cca01d0f2649c32b87bd255568c3ffd1262 Signed-off-by: Poornima G <pgurusid@redhat.com> Signed-off-by: Soumya Koduri <skoduri@redhat.com> Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
* afr: add quorum checks in post-opRavishankar N2018-01-192-1/+30
| | | | | | | | | | afr relies on pending changelog xattrs to identify source and sinks and the setting of these xattrs happen in post-op. So if post-op fails, we need to unwind the write txn with a failure. Change-Id: I0f019ac03890108324ee7672883d774918b20be1 BUG: 1506140 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
* readdir-ahead: Cleanup the xattr request codePoornima G2018-01-191-40/+2
| | | | | | | Updates #297 Change-Id: Ia0c697583751290a455da3cd1894e0c5685d1bd8 Signed-off-by: Poornima G <pgurusid@redhat.com>
* upcall: Allow md-cache to specify invalidations on xattr with wildcardPoornima G2018-01-192-4/+21
| | | | | | | | | | | | | Currently, md-cache sends a list of xattrs, it is inttrested in recieving invalidations for. But, it cannot specify any wildcard in the xattr names Eg: user.* - invalidate on updating any xattr with user. prefix. This patch, enable upcall to honor wildcard in the xattr key names Updates: #297 Change-Id: I98caf0ed72f11ef10770bf2067d4428880e0a03a Signed-off-by: Poornima G <pgurusid@redhat.com>
* geo-rep: Validate availability of gluster binary on slaveKotresh HR2018-01-192-1/+77
| | | | | | | | | | 1. Adds validation to check if gluster binary is available on slave 2. Add a simple geo-rep setup test case to verify whether setup is fine. It's named in such a way that it runs first. BUG: 1532591 Change-Id: Ie777e55ae13db8fa97d4e32464ad82269ee5fd07 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* features/locks: volume option fixes for GD2Pranith Kumar K2018-01-191-0/+18
| | | | | | Updates #302 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Change-Id: I1667fc32b7a31562b289fcab878d94be202407e4
* posix: In getxattr, honor the wildcard '*'Poornima G2018-01-192-23/+43
| | | | | | | | | | | | | | | | | Currently, the posix_xattr_fill performas a sys_getxattr on all the keys requested, there are requirements where the keys could contain a wildcard, in which case sys_getxattr would return ENODATA, eg: if the xattr requested is user.* all the xattrs with prefix user. should be returned, with their values. This patch, changes posix_xattr_fill, to honor wildcard in the keys requested. Updates #297 Change-Id: I3d52da2957ac386fca3c156e26ff4cdf0b2c79a9 Signed-off-by: Poornima G <pgurusid@redhat.com>
* cluster/afr: Adding option to take full file lockkarthik-us2018-01-195-3/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: In replica 3 volumes there is a possibilities of ending up in split brain scenario, when multiple clients writing data on the same file at non overlapping regions in parallel. Scenario: - Initially all the copies are good and all the clients gets the value of data readables as all good. - Client C0 performs write W1 which fails on brick B0 and succeeds on other two bricks. - C1 performs write W2 which fails on B1 and succeeds on other two bricks. - C2 performs write W3 which fails on B2 and succeeds on other two bricks. - All the 3 writes above happen in parallel and fall on different ranges so afr takes granular locks and all the writes are performed in parallel. Since each client had data-readables as good, it does not see file going into split-brain in the in_flight_split_brain check, hence performs the post-op marking the pending xattrs. Now all the bricks are being blamed by each other, ending up in split-brain. Fix: Have an option to take either full lock or range lock on files while doing data transactions, to prevent the possibility of ending up in split brains. With this change, by default the files will take full lock while doing IO. If you want to make use of the old range lock change the value of "cluster.full-lock" to "no". Change-Id: I7893fa33005328ed63daa2f7c35eeed7c5218962 BUG: 1535438 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* distributed-tests: Framework for running tests on a worker clusterkrad2018-01-186-0/+1153
| | | | | | | | | | | | Summary: This framework helps distribute running of unit tests on a cluster of machines. Test Plan: Run tests on fb cluster Reviewers: sshreyas, jdarcy Change-Id: If309f504d9aa959cc8b01c85bff3b5503a890ff1 updates #374 Signed-off-by: krad <krad@fb.com>
* selinux-xlator : validate dict before calling dict_rename_key()Jiffin Tony Thottan2018-01-181-4/+4
| | | | | | Change-Id: I71da3b64e5e8c82e8842e119b2b05da3e2ace550 BUG: 1535772 Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
* md-cache: Serve nameless lookup from cachePoornima G2018-01-181-10/+8
| | | | | | Updates #232 Change-Id: I97e92312a53a50c2d1660bf8d657201fc05a76eb Signed-off-by: Poornima G <pgurusid@redhat.com>
* fix features/leases: volume option fixes for GD2Atin Mukherjee2018-01-181-0/+4
| | | | | | | Updates #302 Change-Id: I31091765b5bfd118fc5370b3515128f0e84e360f Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* readdir-ahead: Add parallel-readdir option in readdir-aheadPoornima G2018-01-182-0/+16
| | | | | | | | | | parallel-readdir option is defined as belonging to readdir-ahead in glusterd-volume-set.c, but was not defined in options of readdir-ahead, fixing the same. Change-Id: I97cc88b38ab99ade5f066519ca1cb9bfed03a7da BUG: 1506197 Signed-off-by: Poornima G <pgurusid@redhat.com>
* performance/io-threads: Fix checked_return coverity errorVarsha Rao2018-01-181-5/+18
| | | | | | | | | Change the return type of set_stack_size function to integer and also check the return value of pthread_attr_init(). This fixes the Checked_Return coverity issue. Change-Id: I270b8bd168b09f0b071437c117e4e23b01398534 Signed-off-by: Varsha Rao <varao@redhat.com>
* features/bit-rot: Fix default value for scrub-throttleKotresh HR2018-01-181-0/+1
| | | | | | Updates #302 Change-Id: Ifc23d5f8b5bc78b95f7c9d92d6df37a9168102f7 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* make sure geo-rep tests run firstAmar Tumballi2018-01-172-0/+0
| | | | | | | | | as we run regression with a 'sort' function, geo-rep thus becomes last test to run. instead, make sure it is the first test by changing the name of directory, and thus any setup failures would be noticed much earlier. Change-Id: I9e8d81824274900be42c4c49c752a1602497fa31 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* rpc/*: auth-header changesAmar Tumballi2018-01-1713-130/+413
| | | | | | | | | | | | | | | | | Introduce another authentication header which can now send more data. This is useful because this data can be common for all the fops, and we don't need to change all the signatures. As part of this, made rpc-clnt.c little more modular to support multiple authentication structures. stack.h changes are placeholder for the ctime etc, can be moved later based on need. updates #384 Change-Id: I6111c13cfd2ec92e2b4e9295896bf62a8a33b2c7 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* dict: add another type to handle backward compatibilityAmar Tumballi2018-01-175-6/+38
| | | | | | | | | | | | | | | | This new type helps to avoid excessive logs. It should be set only in case of * volume graph building (graph.y) * dict unserialize (happens once a dictionary is received on wire in old protocol) All other dict set and get should have proper check and warning logs if there is a mismatch. updates #220 Change-Id: I1cccb304a877aa80c07aaac95f10f5005e35b9c5 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* features/bit-rot: fix volume options for GD2Atin Mukherjee2018-01-171-1/+11
| | | | | | | Updates #302 Change-Id: I8809f269b93253bce049fdbf28a7f44e85a2b9e7 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* fuse: write out reverse notification to fuse dumpCsaba Henk2018-01-171-30/+59
| | | | | | BUG: 1534602 Change-Id: Ide42cf9cffe462d0cc46272b327c2a05999f09ba Signed-off-by: Csaba Henk <csaba@redhat.com>
* core: fix some of the dict_{get,set} with proper APIsAmar Tumballi2018-01-1718-58/+43
| | | | | | | updates #220 Change-Id: I6e25dbb69b2c7021e00073e8f025d212db7de0be Signed-off-by: Amar Tumballi <amarts@redhat.com>
* locks: added inodelk/entrylk contention upcall notificationsXavier Hernandez2018-01-1622-264/+1203
| | | | | | | | | | | | | | The locks xlator now is able to send a contention notification to the current owner of the lock. This is only a notification that can be used to improve performance of some client side operations that might benefit from extended duration of lock ownership. Nothing is done if the lock owner decides to ignore the message and to not release the lock. For forced release of acquired resources, leases must be used. Change-Id: I7f1ad32a0b4b445505b09908a050080ad848f8e0 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es>
* posix: delete stale gfid handles in nameless lookupRavishankar N2018-01-162-1/+81
| | | | | | | | | ..in order for self-heal of symlinks to work properly (see BZ for details). Change-Id: I9a011d00b07a690446f7fd3589e96f840e8b7501 BUG: 1529488 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
* tests: EC test fails with brick mux enabledSunil Kumar Acharya2018-01-151-1/+1
| | | | | | | | | | | | | Problem: With brick mux enabled get_fd_count count was returning wrong value due to parsing issue. Solution: Updated the code to fix parsing problem. BUG: 1533594 Change-Id: I5d7ff6843b4760f866c4a5aab2f13ff7380f248e Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
* storage/posix: Set f_bfree to 0 if brick fullN Balachandran2018-01-151-1/+12
| | | | | | | | | Return 0 free blocks if the brick is full or has less than the reserved limit. Change-Id: I2c5feda0303d0f4abe5af22fac903011792b2dc8 BUG: 1533736 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* performance/readdir-ahead: fix cache usage update issueZhang Huan2018-01-152-6/+9
| | | | | | | | | Use atomic operation to modify cache-size to protect it from concurrent modification. Change-Id: Ie73cdd4abbaf0232b1db4ac856c01d24603890ad BUG: 1533804 Signed-off-by: Zhang Huan <zhanghuan@open-fs.com>
* cluster/afr: Fixing the flaws in arbiter becoming source patchkarthik-us2018-01-137-180/+271
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Setting the write_subvol value to read_subvol in case of metadata transaction during pre-op (commit 19f9bcff4aada589d4321356c2670ed283f02c03) might lead to the original problem of arbiter becoming source. Scenario: 1) All bricks are up and good 2) 2 writes w1 and w2 are in progress in parallel 3) ctx->read_subvol is good for all the subvolumes 4) w1 succeeds on brick0 and fails on brick1, yet to do post-op on the disk 5) read/lookup comes on the same file and refreshes read_subvols back to all good 6) metadata transaction happens which makes ctx->write_subvol to be assigned with ctx->read_subvol which is all good 7) w2 succeeds on brick1 and fails on brick0 and this will update the brick in reverse order leading to arbiter becoming source Fix: Instead of setting the ctx->write_subvol to ctx->read_subvol in the pre-op statge, if there is a metadata transaction, check in the function __afr_set_in_flight_sb_status() if it is a data/metadata transaction. Use the value of ctx->write_subvol if it is a data transactions and ctx->read_subvol value for other transactions. With this patch we assign the value of ctx->write_subvol in the afr_transaction_perform_fop() with the on disk value, instead of assigning it in the afr_changelog_pre_op() with the in memory value. Change-Id: Id2025a7e965f0578af35b1abaac793b019c43cc4 BUG: 1482064 Signed-off-by: karthik-us <ksubrahm@redhat.com>
generated by cgit (git 2.34.1) at 2024-06-17 16:19:54 +0000