summaryrefslogtreecommitdiffstats
path: root/xlators/cluster
Commit message (Collapse)AuthorAgeFilesLines
* dht: dht_lookup_dir_cbk should set op_errno as local->op_errnov3.4.3beta1shishir gowda2014-03-111-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | Two glusterfs clients return inconsistent errnos when the bricks of the volume were down. Consider two gluster mounts. Mount 1 was done when the bricks were online. Mount 2 was done after the bricks were killed, (using the 'glusterfs' command instead of the mount script). For any request, mount 1 will return ENOTCONN, where as mount 2 will return ENOENT. This happens because for the 2nd mount, a fuse would send a lookup on '/' for any request, as it hadn't been done yet. The client xlator returns ENOTCONN, but the dht_lookup_dir_cbk changed this to ENOENT unconditionally when aggregating. So, fuse returned ENOENT, even though the errno should have been ENOTCONN. backporting http://review.gluster.org/6072 BUG: 1019095 Change-Id: Iaa40dffefddfcaf1ab7736f5423d7f9d2ece1363 Original-author: Kaushal M <kaushal@redhat.com> Signed-off-by: shishir gowda <gowda.shishir@gmail.com> Reviewed-on: http://review.gluster.org/6471 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Harshavardhana <harsha@harshavardhana.net> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Treat FOOL condition as split-brain for entry self-healv3.4.3alpha1Pranith Kumar K2014-02-121-0/+48
| | | | | | | | | | BUG: 1057846 Change-Id: I19051c19a54c8aab37eb7cb32dde9f7e9e77c073 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/6854 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: Make sure loc has gfidPranith Kumar K2014-01-131-0/+2
| | | | | | | | | | | | | | | | | Problem: In some code paths neither loc->gfid nor loc->inode->gfid is populated which leads to EINVAL for linkfile setattr in dht_linkfile_attr_heal. Fix: Populate loc->gfid before dht_linkfile_attr_heal. BUG: 971805 Change-Id: I8e4b7510ee5c38aa9ccf5283c7165c7df25ec62b Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/6691 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: Ignore ENOENT errors for unlink of linkfilesAnand Avati2013-12-241-2/+4
| | | | | | | | | | | | | | | Backport of http://review.gluster.org/4971 If unlink of linkfile returns ENOENT, do not fail unlink. Proceed with unlinking of cached file. Change-Id: If7cec92b40c39d68dd9c3606c6c2c3a6bd67d27b BUG: 966848 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6586 Reviewed-by: Harshavardhana <harsha@harshavardhana.net> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* Revert "core: fix errno for non-existent GFID"Vijay Bellur2013-12-246-13/+9
| | | | | | | | | This reverts commit 837422858c2e4ab447879a4141361fd382645406 Change-Id: I0909f26ce088454bb14b3694b489c672286a4ae6 Reviewed-on: http://review.gluster.org/6575 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: interim fix for reverting 837422858cVijay Bellur2013-12-241-1/+1
| | | | | | | | Change-Id: I74818a03f7c5d7891561515af2fa35ea3775255c BUG: 1032894 Signed-off-by: Vijay Bellur <vbellur@redhat.com> Reviewed-on: http://review.gluster.org/6582 Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/dht: Ignore decommissioned subvol in overlap optimizationshishir gowda2013-12-161-0/+7
| | | | | | | | | | | | | Change-Id: Ib727948c6e21b19fd509f258ff0aea1c5d1a84d1 BUG: 966845 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/5056 Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-on: http://review.gluster.org/6517 Reviewed-by: Shishir Gowda <gowda.shishir@gmail.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: Fix anomaly checkv3.4.2qa3shishir gowda2013-12-141-3/+11
| | | | | | | | | | | | | | | We were wrongly detecting holes/overlaps for already accounted errors. Additionally, sort should also handle zero'ed out layout Change-Id: Ic3d13e1d735b914f9acc01fe919bc90656baea48 BUG: 1003851 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/5762 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6469 Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: Del GF_READDIR_SKIP_DIRS key from dict for first_upshishir gowda2013-12-142-4/+12
| | | | | | | | | | | | | | | | | Currently, we sent GF_READDIR_SKIP_DIRS for all subvolumes if first_subvol != first_up_subvolume. Also first_up_subvolume can change with-in the life of a call and cbk. Saving the first_up_subvol in dht_local for checks. Back porting fix http://review.gluster.org/5577 BUG: 996474 Change-Id: I67b5bbe781e12812557b569b7d0a0beba4224159 Signed-off-by: shishir gowda <gowda.shishir@gmail.com> Reviewed-on: http://review.gluster.org/6468 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: Do migration inprog/complete check only if ENOENTAnand Avati2013-12-132-1/+7
| | | | | | | | | | | | | | Additionally, update op_errno to the lasted failure. If failures found in complete_check, error returned would be EUCLEAN instead of the right failure (in this case ENOENT) Change-Id: Ib813867f4b817af651627b9ea07b0b09fa2b26ce BUG: 966852 Original-author: shishir gowda <sgowda@redhat.com> Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6495 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: set layout in inode ctx even if linkfile failsshishir gowda2013-12-101-1/+1
| | | | | | | | | | | | | | | | | Creating linkfile could have failed, but we dont care about linkfile for setting layout in the inode ctx (could be EEXIST etc.) So ignore @inode in cbk and pick it up from local->loc.inode Backporting http://review.gluster.org/6319 BUG: 1032859 Change-Id: Ic95e303a4c060900d041820d4faa68d1c4685b6a Original-author: Anand Avati <avati@redhat.com> Signed-off-by: shishir gowda <gowda.shishir@gmail.com> Reviewed-on: http://review.gluster.org/6470 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* distribute: Rebalance should provide even disk space distributionshishir gowda2013-12-101-15/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Earlier disk space check had an issue which didn't provide the needed functionality to avoid migration when the destination had lesser available space, scenario we need to avoid is stated below : During rebalance `migrate-data` - Destination subvol experiences a `reduction` in 'blocks' of free space, at the same time source subvol gains certain 'blocks' of free space. A valid check is necessary here to avoid errorneous move to destination where the space could be scantily available. This patch provides a proper fix in place by subtracting necessary file blocks from destination and adding those blocks to source. backporting fix http://review.gluster.org/5961 BUG: 982919 Original-author: Harshavardhana <harsha@harshavardhana.net> Signed-off-by: shishir gowda <gowda.shishir@gmail.com> Change-Id: If5808eaa89e66d7bcaeee7268fe3fe5b1b56f51d Signed-off-by: shishir gowda <gowda.shishir@gmail.com> Reviewed-on: http://review.gluster.org/6461 Reviewed-by: Harshavardhana <harsha@harshavardhana.net> Reviewed-by: Vijay Bellur <vbellur@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/afr: handle NULL check before strlen/strcmp in fgetxattrAnand Avati2013-12-041-1/+1
| | | | | | | | | | | xattr name can legally be NULL. Handle that case without crashing. Change-Id: Ie214cb05ccd52565dc247a9234ad83ae799d3866 BUG: 1036879 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6420 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: handle NULL check before strlen/strcmp in fgetxattrAnand Avati2013-12-031-0/+1
| | | | | | | | | | | @key can legally be NULL. Handle that case without crashing. Change-Id: Iaae293caa7eeb24afc9cd2580799173e2ce00911 BUG: 1036879 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6402 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: set layout in inode ctx even if linkfile failsAnand Avati2013-11-261-1/+1
| | | | | | | | | | | | | Creating linkfile could have failed, but we dont care about linkfile for setting layout in the inode ctx (could be EEXIST etc.) So ignore @inode in cbk and pick it up from local->loc.inode Change-Id: I2952799d7ae0d3441b84b2ca2981afd75d7576e2 BUG: 1032859 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6358 Tested-by: Gluster Build System <jenkins@build.gluster.com>
* core: fix errno for non-existent GFIDAnand Avati2013-11-266-9/+13
| | | | | | | | | | | | | | | | When clients refer to a GFID which does not exist, the errno to be returned in ESTALE (and not ENOENT). Even though ENOENT might look "proper" most of the time, as the application eventually expects ENOENT even if a parent directory does not exist, not returning ESTALE results in resolvers (FUSE and GFAPI) to not retry resolution in uncached mode. This can result in spurious ENOENTs during concurrent path modification operations. Change-Id: I7a06ea6d6a191739f2e9c6e333a1969615e05936 BUG: 1032894 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/6322 Tested-by: Gluster Build System <jenkins@build.gluster.com>
* Disable eager-locks on NetBSD for 3.4 branchEmmanuel Dreyfus2013-10-241-0/+10
| | | | | | | | | | | | | | As described in https://bugzilla.redhat.com/show_bug.cgi?id=1005526 eager-locks are broken on release-3.4, at least for NetBSD. This change disable them by default, leaving the admin the possibility to explicitely enable the feature if needed. BUG: 1005526 Change-Id: I6f1b393865b103ec56ad5eb5143f59bb8672f19c Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/6020 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: Treat migration failures due to space constraints as skippedv3.4.1qa3shishir gowda2013-09-192-5/+30
| | | | | | | | | | | | | | | | Currently rebalance/remove-brick op's display migration failed count even for files which failed due to space issues (not enough space for file, or migration leading to cluster imbalance) These will now be counted as skipped, and rebalance/remove-brick status will display the additional counter BUG: 989846 Change-Id: I4efa7ce69dd43680ff47181afed0c561954c5080 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/5977 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/dht: Ignore subvols with error in min-free-disk/inodesAmar Tumballi2013-09-105-17/+86
| | | | | | | | | | | | | | | | | | | Currently when selecting a alternative subvolume when hashed subvol has exceeded min-free-disk/inodes, we do not check if layouts have errors (including decommissioning). This leads to data being written to those subvolumes, and in case of decommissioning, will lead to data loss. BUG: 982919 > Original-Author: shishir gowda <sgowda@redhat.com> > Reviewed-on: http://review.gluster.org/5299 Change-Id: If301a86cf3ca5fad6529bd2e61382f9901663ba0 Signed-off-by: Amar Tumballi <amarts@redhat.com> Reviewed-on: http://review.gluster.org/5888 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: unlock before aborting transactionAnand Avati2013-09-101-0/+2
| | | | | | | | | | | Else this results in a missing frame causing a hang Change-Id: Ib5f3dc6a3999449faa2853cee2944af2fb065a20 BUG: 1002399 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/5878 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/dht: assign layout onto missing directories tooAnand Avati2013-09-101-4/+28
| | | | | | | | | | | | | | | | | | | | | | | | The current self-healing algorithm is ignoring missing directories for assigning new layout. When lookup() is racing against mkdir() or when self-healing a half-done mkdir(), the layout assignment split must happen based on the final number of directories, and not the currently existing number of directories (because we finish mkdir() of missing directories before hash layout assignment). Without this fix, concurrent mkdir() and lookup() will step on each others feet, create a messed up layout on disk, and end up with different in-memory layouts. Once two clients have different in-memory layouts, creation of subdirectory will not arbitrate on the same hashed subvolume and will result in GFID mismatch of the sub-directory. Change-Id: Ia47acad67c265060405984c822b4d37512b9dbb3 BUG: 907072 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/5871 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Perform delayed changelog wakeups for anon fdv3.4.0beta3Pranith Kumar K2013-06-101-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Nfs xlator never does open on a file for performing writes, afr does not perform changelog wakeup for this fd so operations which do metadata operations as soon as the data operations are completed perceive a delay od 'post-op-delay-secs'. Fix: Perform changelog wakeup on anon-fd if the fd with same pid is not present in inode-list. Note: This approach is a short-term fix. A proper fix needs a new domain for taking metadata locks so that data/metadata locks don't compete with each other. BUG: 966018 Change-Id: Ia9188a253e7943801b665e1b9205e2f551952d87 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/5067 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* Fix crash in dht_migration_complete_check_task because of NULL fdEmmanuel Dreyfus2013-06-082-1/+3
| | | | | | | | | | | This is a backport of Ia5a5d40bcea7bfb320ef7096af1e035b8847d4ff BUG: 960055 Change-Id: Ibf3547a775d7ca2f3a097c880cdf38ffafb322da Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/5139 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* dht,posix: support for case discoveryAnand Avati2013-06-081-0/+73
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is support for discovering a filename in a given directory which has a case insensitive match of a given name. It is implemented as a virtual extended attribute on the directory where the required filename is specified in the key. E.g: sh# getfattr -e "text" -n user.glusterfs.get_real_filename:FiLe-B /mnt/samba/patchy getfattr: Removing leading '/' from absolute path names # file: mnt/samba/patchy user.glusterfs.get_real_filename:FiLe-B="file-b" In reality, there can be multiple "answers" as the backend filesystem is case sensitive and there can be multiple files which can strcasecamp() successfully. In this case we pick the first matched file from the first responding server. If a matching file does not exist, we return ENOENT (and NOT ENODATA). This way the caller can differentiate between "unsupported" glusterfs API and file not existing. This API is used by Samba VFS to perform efficient discovery of the real filename without doing a full scan at the Samba level. Change-Id: I53054c4067cba69e585fd0bbce004495bc6e39e8 BUG: 953694 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/5163 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* gfapi: link inodes in relevant entry FOPsAnand Avati2013-06-081-3/+3
| | | | | | | | | | | | Do not let inode linking to happen only in lookup(). While that works, it is inefficient. Change-Id: I51bbfb6255ec4324ab17ff00566375f49d120c06 BUG: 953694 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/5162 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Avoid order mismatch in blocking entrylksPranith Kumar K2013-06-061-6/+9
| | | | | | | | | | | | | | | | | | | | | | Problem: When taking blocking entrylks, afr orders the entrylks based on uuid_compare of gfids of parent dirs, if they are equal then it orders them based on the basenames. While this approach works fine, the implementation assumes loc->gfids to be populated at the time of the comparison, but loc may have gfid in loc->inode->gfid instead of loc->gfid which was leading to order mismatches and dead-locks. Fix: Implemented loc_gfid which gives gfid by checking both loc->gfid, loc->inode->gfid. Used this for ordering the blocking entrylks. Change-Id: I2743fcaff3d670fbeb6b8e0a496f106a3585dde1 BUG: 965987 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/5063 Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: Backport of vme table changes from masterKaushal M2013-06-051-0/+2
| | | | | | | | | | | | | | | | | | | | | This patch backports the following changes from the master branch 99fe09f glusterd: Moved the volume entry table to a separate file. e306d08 glusterd: Changing the volume entry table's representation. eac54f6 glusterd: Added option description, and validation function fields. bcb4235 glusterd: Added validation function for performance cache max and min size. 8897d08 glusterd: Added validation function for quota-timeout. 4579609 glusterd: Added validation function for stripe-block-size. 6788bad glusterd: Fix some options in vme table 549231d glusterd: Added the validation function for subvols-per-directory 9636e63 glusterd: Added description for nfs.transport-type option in volume set help. Change-Id: I4a64ad94f17df4b45a3a32262a83e2c35fb5f7da BUG: 907311 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.org/4956 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Don't queue transactions during open-fd fixPranith Kumar K2013-05-086-385/+167
| | | | | | | | | | | | | | | | | | | | | Before Anonymous fds are available, afr had to queue up transactions if the file is not opened on one of its subvolumes. This happens until the attempt to open the file either succeeds or fails. These attempts happen until the file is successfully opened on the subvolume. Now client xlator uses anonymous fds to perform the fops if the fd used for the fop is not 'opened'. Fops will be successful even when the file is not opened so there is no need to queue up the transactions anymore in afr. Open is attempted on the subvolume where it is not opened independent of the fop. Change-Id: I6d59293023e2de41c606395028c8980b83faca3f BUG: 953887 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4868 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Turn on eager-lock for fd DATA transactionsEmmanuel Dreyfus2013-05-072-21/+7
| | | | | | | | | | | | | | | | | | | | | | | Problem: With the present implementation, eager-lock is issued for any fd fop. eager-lock is being transferred to metadata transactions. But the lk-owner is set to local->fd address only for DATA transactions, but for METADATA transactions it is frame->root. Because of this unlock on the eager-lock fails and rebalance hangs. Fix: Enable eager-lock for fd DATA transactions This is a backport of change If30df7486a0b2f5e4150d3259d1261f81473ce8a http://review.gluster.org/#/c/4588/ BUG: 916226 Change-Id: Id41ac17f467c37e7fd8863e0c19932d7b16344f8 Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/4899 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Preserve mtime in self-healPranith Kumar K2013-04-121-14/+25
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Data self-heal may choose sink iatt to set mtimes. This happens because after syncing of data is done self-heal does one more xattrops/fstat to determine sources sinks to set the inode-ctx. Since this is done after data syncing and erase of xattrs, old source and old sink are now sources, but the mtimes of them differ. Old code just takes the first source from the list and update mtimes, which could be sink before the self-heal started. Fix: Set mtime from 'sources before syncing'. Change-Id: Id769e1b99aa4f041eaee775f64cbf2c57b799723 BUG: 918437 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4658 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-on: http://review.gluster.org/4663
* cluster/distribute: Ignore non-participating subvols for layout checksshishir gowda2013-04-112-20/+88
| | | | | | | | | | | | | | | | | | | | | | | | | Backporting fix http://review.gluster.org/#/c/4668/ When subvols-per-directory is < available subvols, then there are layouts which are not populated. This leads to incorrect identification of holes or overlaps. We need to ignore layouts, which have err == 0, and start == stop. In the current scenario (start == stop == 0). Additionally, in layout-merge, treat missing xattrs as err = 0. In case of missing layouts, anomalies will reset them. For any other valid subvoles, err != 0 in case of layouts being zeroed out. Also reverted back dht_selfheal_dir_xattr, which does layout calculation only on subvols which have errors. BUG: 921408 Change-Id: I75a8edcb92af5b53b3253c9addd7a812e9242836 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4800 Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* dht: improve transform/detransform of d_off (and be ext4 safe)shishir gowda2013-04-111-5/+82
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Backporting Avati's fix http://review.gluster.org/4711 The scheme to encode brick d_off and brick id into global d_off has two approaches. Since both brick d_off and global d_off are both 64-bit wide, we need to be careful about how the brick id is encoded. Filesystems like XFS always give a d_off which fits within 32bits. So we have another 32bits (actually 31, in this scheme, as seen ahead) to encode the brick id - which is typically plenty. Filesystems like the recent EXT4 utilize the upto 63 low bits in d_off, as the d_off is calculated based on a hash function value. This leaves us no "unused" bits to encode the brick id. However both these filesystmes (EXT4 more importantly) are "tolerant" in terms of the accuracy of the value presented back in seekdir(). i.e, a seekdir(val) actually seeks to the entry which has the "closest" true offset. This "two-prong" scheme exploits this behavior - which seems to be the best middle ground amongst various approaches and has all the advantages of the old approach: - Works against XFS and EXT4, the two most common filesystems out there. (which wasn't an "advantage" of the old approach as it is borken against EXT4) - Probably works against most of the others as well. The ones which would NOT work are those which return HUGE d_offs _and_ NOT tolerant to seekdir() to "closest" true offset. - Nothing to "remember in memory" or evict "old entries". - Works fine across NFS server reboots and also NFS head failover. - Tolerant to seekdir() to arbitrary locations. Algorithm: Each d_off can be encoded in either of the two schemes. There is no requirement to encode all d_offs of a directory or a reply-set in the same scheme. The topmost bit of the 64 bits is used to specify the "type" of encoding of this particular d_off. If the topmost bit (bit-63) is 1, it indicates that the encoding scheme holds a HUGE d_off. If the topmost bit is is 0, it indicates that the "small" d_off encoding scheme is used. The goal of the "small" d_off encoding is to stay as dense as possible towards the lower bits even in the global d_off. The goal of the HUGE d_off encoding is to stay as accurate (close) as possible to the "true" d_off after a round of encoding and decoding. If DHT has N subvolumes, we need ROOF(Log2(N)) "bits" to encode the brick ID (call it "n"). SMALL d_off =========== Encoding -------- If the top n + 1 bits are free in a brick offset, then we leave the top bit as 0 and set the remaining bits based on the old formula: hi_mask = 0xffffffffffffffff hi_mask = ~(hi_mask >> (n + 1)) if ((hi_mask & d_off_brick) != 0) do_large_d_off_encoding () d_off_global = (d_off_brick * N) + brick_id Decoding -------- If the top bit in the global offset is 0, it indicates that this is the encoding formula used. So decoding such a global offset will be like the old formula: if ((d_off_global & 0x8000000000000000) != 0) do_large_d_off_decoding() d_off_brick = (d_off_global % N) brick_id = d_off_global / N HUGE d_off ========== Encoding -------- If the top n + 1 bits are NOT free in a given brick offset, then we set the top bit as 1 in the global offset. The low n bits are replaced by brick_id. low_mask = 0xffffffffffffffff << n // where n is ROOF(Log2(N)) d_off_global = (0x8000000000000000 | d_off_brick & low_mask) + brick_id if (d_off_global == 0xffffffffffffffff) discard_entry(); Decoding -------- If the top bit in the global offset is set 1, it indicates that the encoding formula used is above. So decoding would look like: hi_mask = (0xffffffffffffffff << n) low_mask = ~(hi_mask) d_off_brick = (global_d_off & hi_mask & 0x7fffffffffffffff) brick_id = global_d_off & low_mask If "losing" the low n bits in this decoding of d_off_brick looks "scary", we need to realize that till recently EXT4 used to only return what can now be expressed as (d_off_global >> 32). The extra 31 bits of hash added by EXT recently, only decreases the probability of a collision, and not eliminate it completely, anyways. In a way, the "lost" n bits are made up by decreasing the probability of collision by sharding the files into N bricks / EXT directories -- call it "hash hedging", if you will :-) Change-Id: I9551c581c3f3d4c9e719764881036d554f60c557 Thanks-to: Zach Brown <zab@redhat.com> BUG: 838784 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4799 Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/distribute: Fix layout overlaps due to spread-count in selfheal pathshishir gowda2013-03-091-50/+12
| | | | | | | | | | | | | | | We needed to zero out the layout range, before we re-calculate the range. When spread-count is issued, we would end up with stale ranges in the layout. Replaced dht_selfheal_dir_xattr with dht_fix_dir_xattr, which correctly resets the un-used (after re-cal) layouts. Change-Id: I1a900d15df07335f59356bd23182ccec34381ab2 BUG: 884455 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4648 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: do complete split-brain check in all the fd based fopsRaghavendra Bhat2013-03-054-19/+33
| | | | | | | | | | | | | | | | | fd based operations such as readv checked only for data split brain instead of complete split-brain (i.e both data + metadata) assuming that open would have done the complete split-brain check. However open-behind would have unwound open, without winding to afr thus preventing the complete split-brain check and some appliations will be able to read the contents of the file even though the file has metadata split-brain. So let all the fd based fops do a defensive check of complete split-brain. Change-Id: I0ea52f782b371ce73e8e1c61f9def438fce1bd28 BUG: 846240 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/4620 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/distribute: Reopen fds in migration internally as root:rootshishir gowda2013-03-041-1/+10
| | | | | | | | | | | | | | Though linkfile_create and rebalance dst file create sent a setattr with correct ownership, there is still a race window where the linkfile open (client open due to migration) will fail, as its ownership will be root:root. BUG: 884597 Change-Id: Iba73681eae4f280d39ee6c9a40009e195768bee7 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4612 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/distribute: Prevent spurious multiple defrag crawlsshishir gowda2013-03-041-9/+14
| | | | | | | | | | | | | | | | | In dht_notify, we used to create a thread to start defrag crawls after we had heard from all child subvols. This was in-correct, as a later event, could also trigger the crawl again(due to the fact that all subvols had responded). The fix is to make sure, the thread is started only once after all subvols have responded the first time BUG: 916449 Change-Id: I1619344fbb1cb51d5e1db38d8a29821fa870fa8b Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4610 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/distribute: Preserve file size during rebalance migrationshishir gowda2013-03-041-0/+6
| | | | | | | | | | | | | | | | | | If holes are encountered, then we do not write these to the dst, which sometimes causes file size to be lesser than src. Data is not corrupted, as when non-zero reads are received, we do write that data. Calling a truncrate to give file size to prevent it from being truncated to less than src in case the file end has holes. Thanks to Brian Foster for providing the test case BUG: 915554 Change-Id: I7e1e0c475118b073c3ebb87e93220c1ec22e8b7d Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4609 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/distribute: Remove suprious fd_unref callshishir gowda2013-03-041-2/+0
| | | | | | | | | | | | After fix http://review.gluster.org/4282 (libglusterfsterfs/syncop: do not hold ref on the fd in cbk) was pushed, syncop_open does not take a ref anymore. BUG: 910661 Change-Id: Idedff91270966e6e70e71ee83785c0228e238d31 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4608 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/dht: Create linkfile with file uid/gidshishir gowda2013-03-044-4/+101
| | | | | | | | | | | | | | | | Currently, linkfile creation happens as root. use uid/gid returned from _cbk (link/rename) to set the correct ownership of the link files. Also added test/dht.rc to implement common dht functions BUG: 884597 Change-Id: I6bc0e04f62d4716fc033681e5678e852a1be7a2f Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4607 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/dht: pathinfo xattr changes for directoriesVenky Shankar2013-02-082-92/+224
| | | | | | | | | | | | | | | | | | | | Since directories have presence on all subvolumes there is no definite meaning of ->hashed_subvol or ->cached_subvol. getxattr() code path chooses ->cached_subvol for pathinfo extended attribute. While this makes sense of files, it makes less sense for directories. Further if a hashed or a cached subvolume is down, and there's a getxattr request for a directory, we return with an errno. This patch changes pathinfo extended attribute contents by aggregating information from all subvolumes that are up. Change-Id: I58adb741d63ccfd1d0239af75eb65f26f0fb384d Signed-off-by: Venky Shankar <vshankar@redhat.com> BUG: 856455 Reviewed-on: http://review.gluster.org/4047 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* Use proper libtool option -avoid-version instead of bogus -avoidversionAnand Avati2013-02-075-8/+8
| | | | | | | | | | Change-Id: I1c9541058c7d07786539a3266ca125a6a15287d8 BUG: 859835 Signed-off-by: Anand Avati <avati@redhat.com> Original-author: Kacper Kowalik (Xarthisius) <xarthisius.kk@gmail.com> Signed-off-by: Kacper Kowalik (Xarthisius) <xarthisius.kk@gmail.com> Reviewed-on: http://review.gluster.org/3967 Tested-by: Gluster Build System <jenkins@build.gluster.com>
* afr: serialize modification of {entrylk,inodelk}_lock_countAnand Avati2013-02-071-53/+54
| | | | | | | | | | | | | | Typically this lock was not needed in practice, but with http://review.gluster.org/3842, this code gets executed in multiple threads for different servers and we lose a count. This results in leaked lock and a hang for a future transaction. Change-Id: I377ed20e44f2a45cff522289dfef181f0653eca2 BUG: 765564 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4480 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* dht: better layout-optimization algorithmJeff Darcy2013-02-072-22/+76
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This method deals with the case where swapping might gain a bigger overlap for the xlator currently under consideration, but sacrifices even more from the xlator we're swapping with. For example: A = 0x00000000 - 0x44444443 (new 0x00000000 - 0x55555554) B = 0x44444444 - 0x77777776 (new 0x55555555 - 0xaaaaaaa9) C = 0x77777777 - 0xffffffff (new 0xaaaaaaaa - 0xffffffff) Here, the new range for B has a bigger overlap with the old C than with the old B (0x33333333 vs. 0x22222222 to be precise) so looking only at that might lead us to swap. However, such a swap turns the new C's overlap from 0x55555556 (vs. old C) to *zero* (vs. old B). In other words, we've gained 0x11111111 for B but lost 0x55555556 for C, so it's a bad idea. The new algorithm accounts for all effects of the swap, so it not only avoids bad swaps but can make some good ones that would have been missed previously. For example, if swapping a range X with a later range Y would not increase the overlap for X we would previously have skipped it even if the swap would increase Y's overlap without affecting X's. This is the normal case when we're adding a new brick (which initially has zero overlap with any old range) so finding more good swaps is probably even more important than avoiding bad ones. Also, the logic in dht_overlap_calc was completely broken before, causing integer overflows instead of providing correct values, so no matter what higher-level algorithm was in place the GIGO effect would have resulted in bad decisions. Change-Id: If61ed513cfcb931916c6b51da293e3efbaaf385f BUG: 853258 Signed-off-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-on: http://review.gluster.org/3908 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Avoid priv->eager_lock value update racePranith Kumar K2013-02-064-4/+6
| | | | | | | | | | Change-Id: I7049c0c64e36a9dfa4cc0e0b34de7ec111d2f6c1 BUG: 908302 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4076 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Perform wakeup just before fopPranith Kumar K2013-02-062-13/+21
| | | | | | | | | | | | | | | There is no necessity for the delayed-post-op to wait until the next fop phase on the fd completes. Change-log, locks are inherited by the time next fop phase is attempted so the wakeup can happen just before the fop phase is started. Change-Id: I0b8e591f591b0f7565eb55265ab51f476ed2b165 BUG: 908302 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4073 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/dht: Correct min_free_disk behaviourRaghavendra Talur2013-02-042-27/+89
| | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Files were being created in subvol which had less than min_free_disk available even in the cases where other subvols with more space were available. Solution: Changed the logic to look for subvol which has more space available. In cases where all the subvols have lesser than Min_free_disk available , the one with max space and atleast one inode is available. Known Issue: Cannot ensure that first file that is created right after min-free-value is crossed on a brick will get created in other brick because disk usage stat takes some time to update in glusterprocess. Will fix that as part of another bug. Change-Id: If3ae0bf5a44f8739ce35b3ee3f191009ddd44455 BUG: 858488 Signed-off-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-on: http://review.gluster.org/4420 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/stripe: Mount issues with Stripe xlatorVarun Shastry2013-02-033-20/+46
| | | | | | | | | | | | | | | | | | | Problem: * 'CONNECTING' is taken as CHILD_UP. * Sending notifications (default_notify()) for all the events individually while mounting. Solution: * Consider Child up only after the event CHILD_UP is received. * Send a single notification for all the children's events only while mounting. Change-Id: I1b7de127e12f5bfb8f80702dbdce02019e138bc8 BUG: 885072 Signed-off-by: Varun Shastry <vshastry@redhat.com> Reviewed-on: http://review.gluster.org/4356 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/dht: ignore EEXIST error in mkdir to avoid GFID mismatchAnand Avati2013-02-031-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In dht_mkdir_cbk, EEXIST error is treated like a true error. Because of this the following sequence of events can happen, eventually resulting in GFID mismatch and (and possibly leaked locks and hang, in the presence of replicate.) The issue exists when many clients concurrently attempt creation of directory and subdirectory (e.g mkdir -p /mnt/gluster/dir1/subdir) 0. First mkdir happens by one client on the hashed subvolume. Only one client wins the race. Others racing mkdirs get EEXIST. Yet other "laggers" in the race encounter the just-created directory in lookup() on the hash dir. 1. At least one "lagger" lookup() notices that there are missing directories on other subvolumes (which the "winner" mkdir is yet to create), and starts off self-heal of the directory. 2. At least on some subvolumes, self-heal's mkdir wins the race against the "winner" mkdir and creates the directory first. This causes the "winner" mkdir to experience EEXIST error on those subvolumes. 3. On other subvolumes where "winner" mkdir won the race, self-heal experiences EEXIST error, but self-heal is properly translating that into a success (but mkdir code path is not -- which is the bug.) 4. Both mkdir and self-heal assign hash layouts to the just created directory. But self-heal distributes hash range across N (total) subvolumes, whereas mkdir distributes hash range across N - M (where M is the number of subvolumes where mkdir lost the race). Both the clients "cache" their respective layouts in the near future for all future creates inside them (evidence in logs) 5. During the creation of the subdirectory, two clients race again. Ideally winner performs mkdir() on the hashed subvolume and proceeds to create other dirs, loser experiences EEXIST error on the hashed subvolume and backs off. But in this case, because the two clients have different layout views of the parent directory (because of different hash splits and assignements), the hashed subvolumes for the new directory can end up being different. Therefore, both clients now win the race (they were never fighting against each other on a common server), assigning different GFIDs to the directory on their respective (different) subvolumes. Some of the remaining subvolumes get GFID1, others GFID2. Conclusion/Fix: Making mkdir translate EEXIST error as success (just the way self-heal is already rightly doing) will bring back truth to the design claim that concurrent mkdir/self-heals perform deterministic + idempotent operations. This will prevent the differing "hash views" by different clients and thereby also avoid GFID mismatch by forcing all clients to have a "fair race", because the hashed subvolume for all will be the same (and thereby avoiding leaked locks and hangs.) Change-Id: I84592fb9b8a3f739a07e2afb23b33758a0a9a157 BUG: 907072 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4459 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com>
* cluster/afr: added logging of changelog for split-brain in glustershd.log fileVenkatesh Somyajula2013-02-034-10/+68
| | | | | | | | | | Change-Id: Iaf119f839cb2113b8f8efb7bf7636d471b6541bf BUG: 866440 Signed-off-by: Venkatesh Somyajula <vsomyaju@redhat.com> Reviewed-on: http://review.gluster.org/4385 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/dht: stack wind with cookiev3.4.0qa8Varun Shastry2013-01-313-29/+45
| | | | | | | | | | | | | | | Default_fops uses stack_wind_tail. It winds without creating the frame leading into wrong subvol return in the cookie. To avoid the problem caused by the same, we're getting the subvol by passing the cookie. Change-Id: I51ee79b22c89e4fb0b89e9a0bc3ac96c5b469f8f BUG: 893338 Signed-off-by: Varun Shastry <vshastry@redhat.com> Reviewed-on: http://review.gluster.org/4388 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com> Tested-by: Anand Avati <avati@redhat.com>
generated by cgit (git 2.34.1) at 2024-12-29 11:11:38 +0000