summaryrefslogtreecommitdiffstats
path: root/xlators/cluster/dht/src/dht-rebalance.c
Commit message (Collapse)AuthorAgeFilesLines
* cluster/dht: Add migration checks to dht_(f)xattropN Balachandran2018-01-301-39/+52
| | | | | | | | | | | | | | | | The dht_(f)xattrop implementation did not implement migration phase1/phase2 checks which could cause issues with rebalance on sharded volumes. This does not solve the issue where fops may reach the target out of order. > Change-Id: I2416fc35115e60659e35b4b717fd51f20746586c > BUG: 1471031 > Signed-off-by: N Balachandran <nbalacha@redhat.com> Change-Id: I2416fc35115e60659e35b4b717fd51f20746586c BUG: 1498081 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cluster/rebalance: Fix hardlink migration failuresSusant Palai2017-08-111-13/+81
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A brief about how hardlink migration works: - Different hardlinks (to the same file) may hash to different bricks, but their cached subvol will be same. Rebalance picks up the first hardlink, calculates it's hash(call it TARGET) and set the hashed subvolume as an xattr on the data file. - Now all the hardlinks those come after this will fetch that xattr and will create linkto files on TARGET (all linkto files for the hardlinks will be hardlink to each other on TARGET). - When number of hardlinks on source is equal to the number of hardlinks on TARGET, the data migration will happen. RACE:1 Since rebalance is multi-threaded, the first lookup (which decides where the TARGET subvol should be), can be called by two hardlink migration parallely and they may end up creating linkto files on two different TARGET subvols. Hence, hardlinks won't be migrated. Fix: Rely on the xattr response of lookup inside gf_defrag_handle_hardlink since it is executed under synclock. RACE:2 The linkto files on TARGET can be created by other clients also if they are doing lookup on the hardlinks. Consider a scenario where you have 100 hardlinks. When rebalance is migrating 99th hardlink, as a result of continuous lookups from other client, linkcount on TARGET is equal to source linkcount. Rebalance will migrate data on the 99th hardlink itself. On 100th hardlink migration, hardlink will have TARGET as cached subvolume. If it's hash is also the same, then a migration will be triggered from TARGET to TARGET leading to data loss. Fix: Make sure before the final data migration, source is not same as destination. RACE:3 Since a hardlink can be migrating to a non-hashed subvolume, a lookup from other client or even the rebalance it self, might delete the linkto file on TARGET leading to hardlinks never getting migrated. This will be addressed in a different patch in future. > Change-Id: If0f6852f0e662384ee3875a2ac9d19ac4a6cea98 > BUG: 1469964 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/17755 > Smoke: Gluster Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: N Balachandran <nbalacha@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: If0f6852f0e662384ee3875a2ac9d19ac4a6cea98 BUG: 1473141 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17838 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: fix on demand migration files from clientSusant Palai2017-08-111-16/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On demand migration of files i.e. migration done by clients triggered by a setfattr was broken. Dependency on defrag led to crash when migration was triggered from client. Note: This functionality is not available for tiered volumes. Migration from tier served client will fail with ENOTSUP. usage (But refer to the steps mentioned below to avoid any issues) : setfattr -n "trusted.distribute.migrate-data" -v "1" <filename> The purpose of fixing the on-demand client migration was to give a workaround where the user has lots of empty directories compared to files and want to do a remove-brick process. Here are the steps to trigger file migration for remove-brick process from client. (This is highly recommended to follow below steps as is) Let's say it is a replica volume and user want to remove a replica pair named brick1 and brick2. (Make sure healing is completed before you run these steps) Step-1: Start remove-brick process - gluster v remove-brick <volname> brick1 brick2 start Step-2: Kill the rebalance daemon - ps aux | grep glusterfs | grep rebalance\/ | awk '{print $2}' | xargs kill Step-3: Do a fresh mount as mentioned here - glusterfs -s ${localhostname} --volfile-id rebalance/$volume-name /tmp/mount/point Step-4: Go to one of the bricks (among brick1 and brick2) - cd <brick1 path> Step-5: Run the following command. - find . -not \( -path ./.glusterfs -prune \) -type f -not -perm 01000 -exec bash -c 'setfattr -n "distribute.fix.layout" -v "1" ${mountpoint}/$(dirname '{}')' \; -exec setfattr -n "trusted.distribute.migrate-data" -v "1" ${mountpoint}/'{}' \; This command will ignore the linkto files and empty directories. Do a fix-layout of the parent directory. And trigger a migration operation on the files. Step-6: Once this process is completed do "remove-brick force" - gluster v remove-brick <volname> brick1 brick2 force Note: Use the above script only when there are large number of empty directories. Since the script does a crawl on the brick side directly and avoids directories those are empty, the time spent on fixing layout on those directories are eliminated(even if the script does not do fix-layout on empty directories, post remove-brick a fresh layout will be built for the directory, hence not affecting application continuity). Detailing the expectation for hardlink migartion with this patch: Hardlink is migrated only for remove-brick process. It is highly essential to have a new mount(step-3) for the hardlink migration to happen. Why?: setfattr operation is an inode based operation. Since, we are doing setfattr from fuse mount here, inode_path will try to build path from the linked dentries to the inode. For a file without hardlinks the path construction will be correct. But for hardlinks, the inode will have multiple dentries linked. Without fresh mount, inode_path will always get the most recently linked dentry. e.g. if there are three hardlinks named dir1/link1, dir2/link2, dir3/link3, on a client where these hardlinks are looked up, inode_path will always return the path dir3/link3 if dir3/link3 was looked up most recently. Hence, we won't be able to create linkto files for all other hardlinks on destination (read gf_defrag_handle_hardlink for more details on hardlink migration). With a fresh mount, the lookup and setfattr become serialized. e.g. link2 won't be looked up until link1 is looked up and migrated. Hence, inode_path will always have the correct path, in this case link1 dentry is picked up(as this is the most recently looked up inode) and the path is built right. Note: If you run the above script on an existing mount(all entries looked up), hard links may not be migrated, but there should not be any other issue. Please raise a bug, if you find any issue. Tests: Manual > Change-Id: I9854cdd4955d9e24494f348fb29ba856ea7ac50a > BUG: 1450975 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/17115 > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Smoke: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: I9854cdd4955d9e24494f348fb29ba856ea7ac50a BUG: 1473140 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17837 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: Make rebalance throttle option tuned by numberSusant Palai2017-08-111-4/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Current rebalance throttle options: lazy/normal/aggressive may not always be sufficient for the purpose of throttling. In our recent test, we observed for certain setups, normal and aggressive modes behaved similarly consuming full disk bandwidth. So in cases like this admin should be able to tune it down(or vice versa) depending on the need. Along with old throttle configurations, thread counts are tuned based on number. e.g. gluster v set vol-name cluster-rebal.throttle 5. Admin can tune up/down between 0 and the number of cores available. Note: For heterogenous servers, validation will fail on the old server if "number" is given for throttle configuration. The message looks something like this: "volume set: failed: Staging failed on vm2. Error: cluster.rebal-throttle should be {lazy|normal|aggressive}" Test: Manual test by logging active thread number after reconfiguring throttle option. testcase: tests/basic/distribute/throttle-rebal.t > Change-Id: I46e3cde546900307831028b344ecf601fd9b02c3 > BUG: 1438370 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/16980 > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > Smoke: Gluster Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Atin Mukherjee <amukherj@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: I46e3cde546900307831028b344ecf601fd9b02c3 BUG: 1473136 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17835 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: rebalance perf enhancementSusant Palai2017-08-111-106/+244
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Throttle settings "normal" and "aggressive" for rebalance did not have performance difference. normal mode spawns $(no. of cores - 4)/2 threads and aggressive spawns $(no. of cores - 4) threads. Though aggressive mode has twice the number of threads compared to that of normal mode, there was no performance gain when switched to aggressive mode from normal mode. RCA: During the course of debugging the above problem, we tried assigning migration job to migration threads spawned by rebalance, rather than synctasks(as there is more overhead associated to manage the task queue and threads). This gave us a significant improvement over rebalance under synctasks. This patch does not really gurantee that there will be a clear performance difference between normal and aggressive mode, but this patch certainly maximized the disk utilization for 1GBfiles run. Results: Test enviroment: Gluster Config: Number of Bricks: 2 (one brick per disk(RAID-6 12 disk)) Bricks: Brick1: server1:/brick/test1/1 Brick2: server2:/brick/test1/1 Options Reconfigured: performance.readdir-ahead: on server.event-threads: 4 client.event-threads: 4 1000 files with 1GB each were created/renamed such that all files will have server1 as cached and server2 as hashed, so that all files will be migrated. Test machines had 24 cores each. Results with/without synctask based migration: ----------------------------------------------- mode normal(10threads) aggressive(20threads) timetaken 0:55:30 (h:m:s) 0:56:3 (h:m:s) withsynctask timetaken with migrator 0:38:3 (h:m:s) 0:23:41 (h:m:s) threads From above table it can be seen that, there is a clear 2x perf gain between rebalance with synctask vs rebalance with migrator threads. Additionally this patch modifies the code so that caller will have the exact error number returned by dht_migrate_file(earlier the errno meaning was overloaded). This will help avoiding scenarios where migration failure due to ENOENT, can result in rebalance abort/failure. > Change-Id: I8904e2fb147419d4a51c1267be11a08ffd52168e > BUG: 1420166 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/16427 > Smoke: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: N Balachandran <nbalacha@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: I8904e2fb147419d4a51c1267be11a08ffd52168e BUG: 1473134 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17834 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: correct space check for rebalanceSusant Palai2017-08-111-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | With rebalance doing fallocate on destination, we don't need to add file size to the "destination available space" to decide whether to migrate the file or not. Notes: Fallocate would have already occupied the file size space on destination > Change-Id: If7f6a6654e6257726680cf20d618482a6e9095a6 > BUG: 1441508 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/17104 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Amar Tumballi <amarts@redhat.com> > Reviewed-by: N Balachandran <nbalacha@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: If7f6a6654e6257726680cf20d618482a6e9095a6 BUG: 1473133 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17833 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: Skip file migration if the subvol that meets min-free-diskSusant Palai2017-08-111-18/+78
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... criteria happens to be the same subvol containing data-file Rebalance need to figure out a new subvol in case the hashed subvol does not have enough space. In the process of figuring out the new subvol, we need to ignore the source subvol, otherwise it will lead to data loss. Test: Manual Ran the following sizeof /tmp/1: 1.5GB sizeof /brick/1: 16GB sizeof /tmp/2: 1.5GB <start> glusterd; gluster v create test1 vm1:/brick/1 vm1:/tmp/1; gluster v start test1; mount -t glusterfs vm1:test1 /mnt; for i in {1..2000} do dd if=/dev/zero of=/mnt/file$i bs=1KB count=1 &> /dev/null; done gluster v add-brick test1 vm1:/tmp/2 gluster v set test1 min-free-disk 12GB gluster v remove-brick test1 vm1:/tmp/1 star <end> file count and data were intact. > Change-Id: Ib8fc8467a3d48a7c12958824c4f0b88e160b86c1 > BUG: 1441508 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/17064 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: Ib8fc8467a3d48a7c12958824c4f0b88e160b86c1 BUG: 1473133 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17832 Smoke: Gluster Build System <jenkins@build.gluster.org> Tested-by: Shyamsundar Ranganathan <srangana@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: Make rebalance honor min-free-diskSusant Palai2017-08-111-12/+202
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | test: Manual created files of size 1K on 2 brick(of size 1GB) setup . added a brick of size 16GB. set min-free-disk to 12GB(so that first two bricks won't receive any files). removed one of the 1st brick of size 1GB. Logs from test: [2017-04-12 08:52:08.196484] W [MSGID: 0] [dht-rebalance.c:895:__dht_check_free_space] 0-test1-dht: Write will cross min-free-disk for file - /tile32 on subvol - test1-client-1. Looking for new subvol. [2017-04-12 08:52:08.196904] I [MSGID: 0] [dht-rebalance.c:925:__dht_check_free_space] 0-test1-dht: new target found - test1-client-2 for file - /tile32 - Post migration we have two files. The new destination (/brick/1) has the data file [root@vm1 ~]# ll /brick/1/tile32 -rw-r--r--. 2 root root 0 Apr 12 14:22 /brick/1/tile32 - On the old target the linkto file is there with linkto xattr pointing to /brick/1 [root@vm1 ~]# ll /tmp/2/tile32 ---------T. 2 root root 1000 Apr 12 14:22 /tmp/2/tile32 [root@vm1 ~]# getfattr -m . -de text /tmp/2/tile32 getfattr: Removing leading '/' from absolute path names security.selinux="unconfined_u:object_r:user_tmp_t:s0" trusted.gfid="����:Aс�#�/'b2" trusted.glusterfs.dht.linkto="test1-client-2" Marking ./tests/features/worm_sh.t as bad test. Reason being, this patch failed on master branch as well and it has nothing to do with rebalance/remove-brick. > BUG: 1441508 > Change-Id: I90bae251cda3d957a49cdceda90cd08311a392fb > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/17034 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > Reviewed-by: Amar Tumballi <amarts@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Signed-off-by: Susant Palai <spalai@redhat.com> Change-Id: I90bae251cda3d957a49cdceda90cd08311a392fb BUG: 1473132 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17831 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* dht/rebalance: Crawler performance improvementSusant Palai2017-08-111-127/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The job of the crawler in rebalance is to fetch files from each local subvolume and push them to migration queue if it is eligible for migration. And we do a lookup on the entries received to figure out the eligibilty. Since, the lookup done is on a local subvolume we receive linkto files and regular files as well. This requires us to do two lookups. first: do a lookup on the file to figure out whether it is a linkto file second: do a lookup on the file to figure out if it should be migrated Note: The migrator thread also does one lookup for the file before migration. Optimization: Remove the lookup done by the crawler. Offload these task to the migrator threads. For linkto file verification get the stat and xattr information from readdirp. So in total we have one lookup instead of three for each entry. Performance numbers: Create two node, two brick setup. Created 100000 files. And started rebalance. Since, there is no add-brick, no files will be migrated and we will get the crawler performance. Without patch: [root@gprfs039 ~]# grs Node Rebalanced-files size scanned failures skipped status run time in h:m:s --------- ----------- ----------- ----------- ----------- ----------- ------------ -------------- localhost 0 0Bytes 50070 0 0 completed 0:0:48 server2 0 0Bytes 49930 0 0 completed 0:0:44 volume rebalance: test1: success Total: 48 seconds WiththecurrentPatch: [root@gprfs039 mnt]# gluster v rebalance test1 status Node Rebalanced-files size scanned failures skipped status run time in h:m:s --------- ----------- ----------- ----------- ----------- ----------- ------------ -------------- localhost 0 0Bytes 50070 0 0 completed 0:0:12 server2 0 0Bytes 49930 0 0 completed 0:0:12 volume rebalance: test1: success Total: 12 seconds That's 4X speed gain. :) > Updates glusterfs#155 > Change-Id: Idc8e5b366e76c54aa40d698876ae62fe1630b6cc > BUG: 1439571 > Signed-off-by: Susant Palai <spalai@redhat.com> > Reviewed-on: https://review.gluster.org/15781 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Updates glusterfs#155 Change-Id: Idc8e5b366e76c54aa40d698876ae62fe1630b6cc BUG: 1473129 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: https://review.gluster.org/17830 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: Additional checks for rebalance estimatesN Balachandran2017-07-041-5/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The rebalance estimates calculation was not handling calculations correctly when no files had been processed, i.e., when rate_lookedup was 0. Now, the estimated time is set to 0 in such scenarios as there is no way for rebalance to figure out how long the process will take to complete without knowing the rate at which the files are being processed. > BUG: 1457985 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > Reviewed-on: https://review.gluster.org/17564 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Amar Tumballi <amarts@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Change-Id: I7b6378e297e1ba139852bcb2239adf2477336b5b BUG: 1460914 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17599 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-by: Raghavendra Talur <rtalur@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* cluster/dht: Include dirs in rebalance estimatesN Balachandran2017-06-201-26/+77
| | | | | | | | | | | | | | | | | | | | | | | | | Empty directories were not being considered while calculating rebalance estimates leading to negative time-left values being displayed as part of the rebalance status. > BUG: 1457985 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > Reviewed-on: https://review.gluster.org/17448 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Amar Tumballi <amarts@redhat.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Change-Id: I48d41d702e72db30af10e6b87b628baa605afa98 BUG: 1460914 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17530 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra Talur <rtalur@redhat.com>
* cluster/dht: Rebalance on all nodes should migrate filesN Balachandran2017-05-181-12/+85
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Rebalance compares the node-uuid of a file against its own to and migrates a file only if they match. However, the current behaviour in both AFR and EC is to return the node-uuid of the first brick in a replica set for all files. This means a single node ends up migrating all the files if the first brick of every replica set is on the same node. Fix: AFR and EC will return all node-uuids for the replica set. The rebalance process will divide the files to be migrated among all the nodes by hashing the gfid of the file and using that value to select a node to perform the migration. This patch makes the required DHT and tiering changes. Some tests in rebal-all-nodes-migrate.t will need to be uncommented once the AFR and EC changes are merged. > BUG: 1366817 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > Reviewed-on: https://review.gluster.org/17239 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Amar Tumballi <amarts@redhat.com> > Reviewed-by: Jeff Darcy <jeff@pl.atyp.us> > Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com> (cherry picked from commit b23bd3dbc2c153171d0bb1205e6804afe022a55f) Change-Id: I5ce41600f5ba0e244ddfd986e2ba8fa23329ff0c BUG: 1451561 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17311 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: Fix ret checkN Balachandran2017-05-101-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Fixed an incorrect return code check in the rebalance code. > BUG: 1448640 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > Reviewed-on: https://review.gluster.org/17197 > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Smoke: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> (cherry picked from commit 67598f538efb24a9e5ac561b294a05e707e15761) Change-Id: I60804ff121cec7a2f0419e2ee70dd22ea7533c0c BUG: 1448864 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17204 Reviewed-by: MOHIT AGRAWAL <moagrawa@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* cluster/dht: Don't update layout in rebalance_task_completionN Balachandran2017-02-071-24/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Updating the layout in the dht inode_ctx in rebalance_task_completion after the file is migrated is erroneous in case of files with hardlinks. This step can be skipped as the layout will be set in the syncop_lookup call post the migration in dht_migrate_file. > Change-Id: I24ac798a919585d91a117d6a207e6a31b88486c6 > BUG: 1415761 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > Reviewed-on: https://review.gluster.org/16457 > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > Smoke: Gluster Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Reviewed-by: Susant Palai <spalai@redhat.com> (cherry picked from commit ddf05f3d1e39cc920251c809e9ba42fe42b2c5f2) Signed-off-by: N Balachandran <nbalacha@redhat.com> Change-Id: I46176f8605cfb782aa17c2bc9ffd7a5f0d07dd88 BUG: 1419855 Reviewed-on: https://review.gluster.org/16554 Tested-by: N Balachandran <nbalacha@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* dht/rebalance Estimate time to complete rebalanceN Balachandran2017-01-241-1/+101
| | | | | | | | | | | | | | | | | | | | | | | | | The estimates will be logged to the rebalance log on running gluster v rebalance <vol> status > Change-Id: I9d51b139cd4c8dfde1ff2c2050720ae606c13fc6 > BUG: 1396004 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > Reviewed-on: http://review.gluster.org/15893 > Smoke: Gluster Build System <jenkins@build.gluster.org> > NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> > CentOS-regression: Gluster Build System <jenkins@build.gluster.org> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> (cherry picked from commit 2edd75ec8de17da89004859375844f60890a4df0) Signed-off-by: N Balachandran <nbalacha@redhat.com> Change-Id: I58b0550210149443966798d9a26e73cb598eeb6a BUG: 1415915 Reviewed-on: https://review.gluster.org/16458 Tested-by: N Balachandran <nbalacha@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* tier : Tier as a servicehari gowtham2017-01-161-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tierd is implemented by separating from rebalance process. The commands affected: 1) Attach tier will trigger this process instead of old one 2) tier start and tier start force will also trigger this process. 3) volume status [tier] will show tier daemon as a process instead of task and normal tier status and tier detach status works. 4) tier stop implemented. 5) detach tier implemented separately along with new detach tier status 6) volume tier volname status will work using the changes. 7) volume set works This patch has separated the tier translator from the legacy DHT rebalance code. It now sends the RPCs from the CLI to glusterd separate to the DHT rebalance code. The daemon is now a service, similar to the snapshot daemon, and can be viewed using the volume status command. The code for the validation and commit phase are the same as the earlier tier validation code in DHT rebalance. The “brickop” phase has been changed so that the status command can use this framework. The service management framework is now used. DHT rebalance does not use this framework. This service framework takes care of : *) spawning the daemon, killing it and other such processes. *) volume set options , which are written on the volfile. *) restart and reconfigure functions. Restart is to restart the daemon at two points 1)after gluster goes down and comes up. 2) to stop detach tier. *) reconfigure is used to make immediate volfile changes. By doing this, we don’t restart the daemon. it has the code to rewrite the volfile for topological changes too (which comes into place during add and remove brick). With this patch the log, pid, and volfile are separated and put into respective directories. Change-Id: I3681d0d66894714b55aa02ca2a30ac000362a399 BUG: 1313838 Signed-off-by: hari gowtham <hgowtham@redhat.com> Reviewed-on: http://review.gluster.org/13365 Smoke: Gluster Build System <jenkins@build.gluster.org> Tested-by: hari gowtham <hari.gowtham005@gmail.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
* dht/rebalance: remove errno check for failure detectionSusant Palai2017-01-061-16/+13
| | | | | | | | | | | BUG: 1410355 Change-Id: I867419ca36a81ef7209e6911a46c1c2c898b8eab Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/16328 NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* dht/rebalance: reverify lookup failuresSusant Palai2016-12-221-56/+121
| | | | | | | | | | | | | | | | | | race: readdirp has read one entry, and doing a lookup on that entry, but user might have renamed/removed that entry just after readdirp but before lookup. Since remove-brick is a costly opertaion,will ingore any ENOENT/ESTALE failures and move on. Change-Id: I62c7fa93c0b9b7e764065ad1574b97acf51b5996 BUG: 1408115 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/15846 Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
* cluster/dht: A hard link is lost during rebalance + lookupMohit Agrawal2016-11-291-40/+62
| | | | | | | | | | | | | | | | | | | | | | Problem: A hard link is lost during rebalance + lookup.Rebalance skip files if file has hardlink.In dht_migrate_file __is_file_migratable () function checks if a file has hardlink, if yes file is not migrated but if link is created after call this function then link will lost. Solution: Call __check_file_has_hardlink to check hardlink existence after (S+T) bits in migration process ,if file has hardlink then skip the file for migrate rebalance process. BUG: 1396048 Change-Id: Ia53c07ef42f1128c2eedf959a757e8df517b9d12 Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> Reviewed-on: http://review.gluster.org/15866 Reviewed-by: Susant Palai <spalai@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* events: Add FMT_WARN for gf_eventPranith Kumar K2016-11-181-1/+1
| | | | | | | | | | | | | | | | | Raghavendra G found that posix is trying to print %s but passing an int when HEALTH_CHECK fails in posix. These are the kind of bugs that should be caught at compilation itself. Also fixed the problematic gf_event() callers. BUG: 1386097 Change-Id: Id7bd6d9a9690237cec3ca1aefa2aac085e8a1270 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/15671 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* cluster/dht: Incorrect volname in rebalance eventsN Balachandran2016-10-251-6/+50
| | | | | | | | | | | | | | | | The rebalance event code was using strtok to parse the volume name which is incorrect. Reworked the code to get the correct volume name using strstr. Change-Id: Ib5f3305a34e6bf1ecfef677d87c5aff96bdeb0e6 BUG: 1388010 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/15712 NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* dht/events: Added rebalance eventsN Balachandran2016-09-161-6/+50
| | | | | | | | | | | | | | | | | The rebalance process will now send an event when it is complete. Also fixed a problem where the run-time was not always set causing spurious rebalance failure events to be sent. Change-Id: Ib445171c78c9560940022bca20c887d31a9bb1ca BUG: 1371874 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/15501 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* dht: Proper log message if data migration is skippedankit2016-09-111-8/+8
| | | | | | | | | | | | Change-Id: Id0af15a2aec96bdbe675b4c959b56f0fc8e72504 BUG: 1341948 Signed-off-by: ankit <anraj@redhat.com> Reviewed-on: http://review.gluster.org/15345 Tested-by: ankitraj Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: N Balachandran <nbalacha@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* cluster/tier: add tiering eventsMilind Changire2016-09-081-0/+5
| | | | | | | | | | | | | | | | | | | | | | Add events for: * tier attach and detach * tier pause and resume * tier rising and dropping hi and lo watermarks Update eventskeygen.py with tiering events. Update cli help with: * attach: add optional force argument * detach: make force available as non-optional argument on its own Change-Id: I43990d3a8742151a4a7889bafa19cb572fe661bd BUG: 1368336 Signed-off-by: Milind Changire <mchangir@redhat.com> Reviewed-on: http://review.gluster.org/15232 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* dht/rebalance: allocate migrator thread pool dynamicallySusant Palai2016-07-281-3/+14
| | | | | | | | | | | | | | | | | | Problems: The maximum number of migratior threads created was static set to "40". And the number of these threads get created in rebalance depends on the number of cores user has. If the number of cores exceeds 40, a crash or memory corruption can be seen. Fix: Make the migratior thread pool dynamic. Change-Id: Ifbdac8a1a396363dd75e2f6bcb454070cfdbf839 BUG: 1359711 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/15000 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* cluster/tier: fix statfs for dht/tiered volumesDan Lambright2016-07-271-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | Return the correct size of the tiered volume in statfs. It should be the size of the cold tier, not the sum of the hot and cold tier, because the hot tier is a cache and not an extension of the volume's capacity. The number of free blocks, etc is the cold tier's capacity subtracted by the sum of utilization on the hot and cold tiers. Note if both tiers are part of the same file system this must be accounted for as well. The patch also fixes a pre-existing bug in the DHT/tier translators. If statfs was taken on a file, the code only calculated free space on the cached subvolume, not all subvolumes in the replica group. With the fix, this is corrected, except in the case where quota is used with the deem-statfs option set to "on". Change-Id: I2b8bcb4511edf83f12130960aad0a609fcf8f513 BUG: 1339689 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/14536 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* dht : add metalock/unlockSusant Palai2016-06-031-5/+98
| | | | | | | | | | | | Change-Id: I842a7ea1b286f1b893b200fe647597e7fd0f2105 BUG: 1331720 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/14252 Smoke: Gluster Build System <jenkins@build.gluster.com> Tested-by: Prasanna Kumar Kalever <pkalever@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* dht/rebalance: mark hardlink failures as skipped in rebalanceSusant Palai2016-05-251-0/+8
| | | | | | | | | | | | | | | Since rebalance(not remove-brick) process does not migrate hardlinks mark them as skipped rather than failed as it creates confusion for the users. Change-Id: I5d469d10146274f00bb91482d0373c5235a9b8b2 BUG: 1339071 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/14493 NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> Smoke: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: N Balachandran <nbalacha@redhat.com>
* tier/detach: Clear tier-fix-layout-complete xattr after migration threads joinJoseph Fernandes2016-05-141-33/+42
| | | | | | | | | | | | | | | | | | | | | Previously we had wrongly placed the clearing tier-fix-layout-complete xattr before the joining of migration threads. This would lead to situations where failure of clearing the xattr would cause the premature death of migration threads. Now we clear the xattr only after the data movement threads join, ensuring that all migration is done. Change-Id: I829b671efa165ae13dbff7b00707434970b37a09 BUG: 1334839 Signed-off-by: Joseph Fernandes <josferna@redhat.com> Reviewed-on: http://review.gluster.org/14285 Smoke: Gluster Build System <jenkins@build.gluster.com> Tested-by: Joseph Fernandes CentOS-regression: Gluster Build System <jenkins@build.gluster.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* tier/detach : During detach check if background fixlayout is doneJoseph Fernandes2016-05-061-1/+16
| | | | | | | | | | | | | | | During detach check if background fixlayout is done, if not done ignore the case and continue detach. Change-Id: I5d5cfc0e73d0eb217fdeab54c432dc4af8bc598d BUG: 1332136 Signed-off-by: Joseph Fernandes <josferna@redhat.com> Reviewed-on: http://review.gluster.org/14147 Smoke: Gluster Build System <jenkins@build.gluster.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* dht/rebalance: add lock migration fop to dht_migrate_fileSusant Palai2016-05-011-63/+105
| | | | | | | | | | | Change-Id: Id0e7400c8ae950c90d42a3ddf8b558a14959a1f8 BUG: 1326085 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/14074 Smoke: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
* dht/rebalance: Handle GF_DEFRAG_STOPSusant Palai2016-04-251-0/+17
| | | | | | | | | | | | | | | | Problem: On a rebal stop, the migrator threads don't intimate the crawler thread to wake up in case it is waiting on signal from migrator thread. Change-Id: I3cc4be41a4db25f48fee059ebb79a97ee99dcd00 BUG: 1327507 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/14004 Smoke: Gluster Build System <jenkins@build.gluster.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* tier/dht : Attach tier fix layout to run in backgroundJoseph Fernandes2016-03-281-16/+228
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1. Spawn a thread for background fix-layout for tier process. 2. Once the fix-layout is completed a marker xttr is set on the root of volume to mark the completion of the background fixlayout, so that even if the tier process is spawned again, fixlayout will not be issued, if it was completed last time. 3. Please note that promotion of legacy files will happen eventually as the ctr lookup heal in the fixlayout slowly heals the ctr db for legacy files OR the ctr lookup heal happend due to a name lookup. 4. When a detach tier is successful in evacuation data from hot tier, we remove the marker xattr is removed. So that next attach tier runs the background tier fixlayout. what is remaining ? 1. Instead of clearing the marker xattr of tiering fix layout at the end of detach start clear it during detach commit. But the issue is detach commit is a glusterd operation and the volume is not mounted in glusterd. The reason we want to do it in detach commit is that if the admin wants to attach the same tier again, then a background fixlayout will be triggered, which would not be needed. 2. Clearing the CTR DB of the cold bricks when there is a detach commit, as it will be having entries which will be stale when the volume is used, with ctr off (ctr is switched off only when we have detach commit.) Change-Id: Ibe343572e95865325cd0eef4d0b976b626a3c0c5 BUG: 1313228 Signed-off-by: Joseph Fernandes <josferna@redhat.com> Reviewed-on: http://review.gluster.org/13491 Smoke: Gluster Build System <jenkins@build.gluster.com> Tested-by: Joseph Fernandes NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* cluster/dht : Rebalance process crashes due to double fd_unrefN Balachandran2016-01-101-6/+11
| | | | | | | | | | | | | The dst_fd was being unrefed twice in case the call to __dht_rebalance_create_dst_file failed. Change-Id: I56c5aff7fa3827887e67936b8aa1ecbd1a08a9e9 BUG: 1296611 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/13193 Reviewed-by: Susant Palai <spalai@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com>
* cluster/tier: allow db queries to be interruptableDan Lambright2016-01-071-14/+55
| | | | | | | | | | | | | | | | | | | | A query to the database may take a long time if the database has many entries. The tier daemon also sends IPC calls to the bricks which can run slowly, espcially in RHEL6. While it is possible to track down each such instance, the snapshot feature should not be affected by database operations. It requires no migration be underway. Therefore it is okay to pause tiering at any time except when DHT is moving a file. This fix implements this strategy by monitoring when control passes to DHT to migrate a file using the GF_XATTR_FILE_MIGRATE_KEY trigger. If it is not, the pause operation is successful. Change-Id: I21f168b1bd424077ad5f38cf82f794060a1fabf6 BUG: 1287842 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/13104 Reviewed-by: Joseph Fernandes Tested-by: Gluster Build System <jenkins@build.gluster.com>
* Tier: "tier start force" command implementationhari gowtham2015-12-221-2/+1
| | | | | | | | | | | | | | | | The start command doesnt restart the tier deamon if the deamon is running at one node. hence to bring up the tierd on the nodes where the deamon is down, the force command is implemented. It skips the check for tierd running. Change-Id: I0037d3e5ecfe56637d0da201a97903c435d26436 BUG: 1292112 Signed-off-by: hari gowtham <hgowtham@redhat.com> Reviewed-on: http://review.gluster.org/12983 Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* cluster/tier: do not block in synctask created from pause tierDan Lambright2015-12-211-13/+70
| | | | | | | | | | | | | | | | | | | | | | | | | | | We had run sleep() in the pause tier callback. Blocking within a synctask is dangerous. The sleep() call does not inform the synctask scheduler that a thread is no longer running. It therefore believes it is running. If a second synctask already exists, it may not be able to run. This occurs if the thread limit in the pool has been reached. Note the pool size only grows when a synctask is created, not when it is moved from wait state to run state, as is the case when an FOP completes. When the tier is paused during migration, synctasks already exist waiting for responses to FOPs to the server with high probability. The fix is to yield() in the RPC callback, which will place the synctask into the wait queue and free up a thread for the FOP callback. A timer wakes the callback after sufficient time has elapsed for the pause to occur. Change-Id: I6a947ee04c6e5649946cb6d8207ba17263a67fc6 BUG: 1267950 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/12987 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com>
* tier: Demotion failed if the file was renamed when it was in coldMohammed Rafi KC2015-12-211-11/+26
| | | | | | | | | | | | | | | | | | | | During migration if the file is present we just open the file in hashed subvol. Now if the linkfile present on hashed is just linkfile to another subvol, we actually open in hashed subvol. But subsequent operation will go to linkto subvol ie, to non-hashed subvol. This operation will get failed since we haven't opened d on non-hashed. Change-Id: I9753ad3a48f0384c25509612ba76e7e10645add3 BUG: 1292067 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: http://review.gluster.org/12980 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Susant Palai <spalai@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* tier/dht : Properly free file descriptors during data migrationJoseph Fernandes2015-12-211-6/+5
| | | | | | | | | | | | | | | While tier migration, free src and dst fd's when create of destination or open of source fails. Change-Id: I62978a669c6c9fbab5fed9df2716b9b2ba00ddf1 BUG: 1291566 Signed-off-by: Joseph Fernandes <josferna@redhat.com> Reviewed-on: http://review.gluster.org/12969 Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* cluster/tier: fix tier-max-files bookeeping and helpDan Lambright2015-12-181-1/+0
| | | | | | | | | | | | | | | | | | The option tier-max-files represents the maximum number of files trasferred by a node in a gives cycle. Fix help message to reflect the "per node" aspect. The code transferred one more file per cycle than the given value. Also change the default values of max file and max bytes to very large values, effectively we will not throttle migration unless the administrator requests it via CLI. Change-Id: Ic2949ed3d8c35afe7c9ae4db72195603cfb2e28f BUG: 1292671 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/12984 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* dht/rebalance: Use seperate return variable for destination cleanupJoseph Fernandes2015-12-121-3/+3
| | | | | | | | | | | | | | | Use seperate local return variable for destination cleanup, without messing with the function return variable. Change-Id: Iaea9ed2927234fdb888aef7a31ec362090e98196 BUG: 1290975 Signed-off-by: Joseph Fernandes <josferna@redhat.com> Reviewed-on: http://review.gluster.org/12956 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* cluster/tier: fix loading tier.so into glusterdN Balachandran2015-12-031-9/+2
| | | | | | | | | | | | | | | | | glusterd occasionally loads shared libraries of translators. This failed for tiering due to a reference to dht_methods which is defined as a global variable which is not necessary. The global variable has been removed and this is now a member of dht_conf and is now initialised in the *_init calls. Change-Id: Ifa0a21e3962b5cd8d9b927ef1d087d3b25312953 BUG: 1287842 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/12863 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* cluster/tier: Do not delete linkto file on demotionN Balachandran2015-11-181-2/+7
| | | | | | | | | | | | | | | | | The current DHT migration code will always delete the src linkto file after migration as dht always moves files to the hashed subvol. This is not the case in tiering. The lack of linkto files causes rename to fail leaving 2 files with the same name but different gfids on the volume. Modified to leave the linkto file behind if the source volume is the hashed subvolume. Change-Id: I2b99f7d34b4b719aee6232dc40c6a8f8ba88225d BUG: 1279376 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/12551 Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
* cluster/tier: Disallow detach commit when detach in progressDan Lambright2015-11-161-3/+0
| | | | | | | | | | | | | 1. Check if detach is running, disallow detach commit if so. 2. Cleanup shutdown of tier daemon on detach: do not rerun fix-layout, do not send incorrect status back to glusterd. Change-Id: I97202f748773c1176396a4ffd32a4c7fa9b9c1bc BUG: 1279637 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/12560 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
* cluster/tier: fix lookup-unhashed on tiered volumesDan Lambright2015-11-051-2/+1
| | | | | | | | | | | | During attach tier the commit hash must be copied to the hot tier. Change-Id: I91b92fd8e98696993433856e1436409b657c439d BUG: 1277716 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/12498 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/dht/rebalance: rebalance failure handlingSusant Palai2015-10-291-23/+144
| | | | | | | | | | | | | | | | | | | | | | | | At current state rebalance aborts basically on any failure like fix-layout of a directory, readdirp, opendir etc. Unless it is not a remove-brick process we can ignore these failures. Major impact: Any failure in the gf_defrag_process_dir means there are files left unmigrated in the directory. Fix-layout(setxattr) failure will impact it's child subtree i.e. the child subtree will not be rebalanced. Settle-hash (commit-hash)failure will trigger lookup_everywhere for immediate children until the next commit-hash. Note: Remove-brick opertaion is still sensitive to any kind of failure. Change-Id: I08ab71909bc832f03cc1517172525376f7aed14a BUG: 1257076 Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/12013 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* core: use syscall wrappers instead of direct syscalls - miscellaneousKaleb S. KEITHLEY2015-10-281-1/+2
| | | | | | | | | | | | | | | various xlators and other components are invoking system calls directly instead of using the libglusterfs/syscall.[ch] wrappers. If not using the system call wrappers there should be a comment in the source explaining why the wrapper isn't used. Change-Id: I1f47820534c890a00b452fa61f7438eb2b3f667c BUG: 1267967 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: http://review.gluster.org/12276 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/tier do not log error message on lookup heal for files on hot tierDan Lambright2015-10-281-4/+11
| | | | | | | | | | | | | | | | | | On fix-layout heal files are scanned. Files found are exist on the hot or cold subvolume. Those not found in the cold tier would exist on the hot. They should not be flagged as an error. Replace INFO with TRACE for common tier migration logs. Frequent migration was growing the log files too quickly. On migratation failures, do not acrue files towards cycle limit's budget. Change-Id: Ie832ee07c43bce5477ae81c939d1fe8416a11615 BUG: 1275383 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/12430 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Joseph Fernandes
* cluster/tier: add pause tier for snapshotsDan Lambright2015-10-211-3/+122
| | | | | | | | | | | | | | | | | | | | Snaps of tiered volumes cannot handle files undergoing migration. We implement a helper mechanism to "pause" migration. Any files undergoing migration are aborted. Clean up is done to remove sticky bits and data at the destination. Migration is restarted after snap completes. For testing an internal switch is added. It is not exposed externally. gluster volume set vol1 tier-pause [true|false] Change-Id: Ia85bbf89ac142e9b7e73fcbef98bb9da86097799 BUG: 1267950 Signed-off-by: Dan Lambright <dlambrig@redhat.com> Reviewed-on: http://review.gluster.org/12304 Reviewed-by: N Balachandran <nbalacha@redhat.com> Tested-by: NetBSD Build System <jenkins@build.gluster.org> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/dht : Do not migrate files with POSIX locks heldN Balachandran2015-10-151-10/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | dht_migrate_file does not migrate file locks to the dst file. Any locks held on the source file are lost once the migration is complete. This issue is magnified in the case of a tier volume as file migrations occur more frequently and repeatedly as compared to a DHT rebalance. The fix makes 2 changes: 1. Before starting the actual migration process, check if there are any locks held on the file. If yes, do not migrate the file. 2. The rebalance process tries to lock on the entire file just before moving into the Phase 2 of the file migration. If the lock acquisition fails, the file migration does not proceed. If the lock is granted, the file migration proceeds. This still leaves a small window where conflicting locks can be granted to different clients. If client1 requests a lock on the src file just after it is converted to a linkto file and client2 requests a lock on the dst data file, they will both be granted, but all FOPs will be redirected to the dst data file. This issue will be taken up in a subsequent patch. Change-Id: I8c895fc3cced50dd2894259d40a827c7b43d58ac BUG: 1271148 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/12347 Tested-by: NetBSD Build System <jenkins@build.gluster.org> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: Dan Lambright <dlambrig@redhat.com>
generated by cgit (git 2.34.1) at 2025-08-27 03:25:20 +0000