summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* features/index: Choose different base file on EMLINK errorPranith Kumar K2018-04-062-18/+61
| | | | | | | Change-Id: I4648816af908539efdc2528608aa2ebf7f0d0e2f fixes: bz#1559004 BUG: 1559004 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* doc: Update the admin guide linkVarsha Rao2018-04-061-1/+1
| | | | | | | | Update the existing admin guide link as it is incorrect. Change-Id: I05669192623aeac287dfa9002caa0f390ea79499 Updates: bz#1193929 Signed-off-by: Varsha Rao <varao@redhat.com>
* cluster/ec: Turn ON the stripe-cache option by defaultAshish Pandey2018-04-061-1/+1
| | | | | | Change-Id: I0a290396c30c635b13ee73004d20259efb76a954 fixes: bz#1563945 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
* gfapi: fix a couple of minor issuesKaleb S. KEITHLEY2018-04-053-6/+2
| | | | | | | | | | | | | | | | | | | | duplicatation of exported functions in gfapi.map. Only the newest one is needed. Both the legacy and current symbols are exported. glfs_io_cbk34 typedef should not be in a public header file. The old application was compiled with the original glfs_io_cbk. Outside of libgfapi, nothing now uses/needs this old typedef, move it into the C file that needs it. Similarly glfs_realpath34() decl should not be in glfs.h. Period. Old applications were compiled with the then glfs_realpath() decl and linked with glfs_realpath@@GFAPI_3_4.0. New applications should only call glfs_realpath() and it will be linked to the new/current glfs_realpath(). Change-Id: Icd5b0c9e9b68f0c133f14447b09ace35f33dbab2 fixes: bz#1564235 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
* glusterd: show brick online after port registrationAtin Mukherjee2018-04-051-2/+3
| | | | | | | | | | | | | | | | | gluster-block project needs a dependency check to see if all the bricks are online before bringing up the relevant gluster-block services. While the patch https://review.gluster.org/#/c/19785/ attempts to write the script but brick should be only marked as online only when the pmap_signin is completed. While this is perfectly fine for non brick multiplexing, but with brick multiplexing this patch still doesn't eliminate the race completely as the attach_req call is asynchrnous and glusterd immediately marks the port as registerd. Change-Id: I81db54b88f7315e1b24e0234beebe00de6429f9d Fixes: bz#1563273 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* afr: add quorum checks in pre-opRavishankar N2018-04-051-33/+31
| | | | | | | | | | | | | | | | | Problem: We seem to be winding the FOP if pre-op did not succeed on quorum bricks and then failing the FOP with EROFS since the fop did not meet quorum. This essentially masks the actual error due to which pre-op failed. (See BZ). Fix: Skip FOP phase if pre-op quorum is not met and go to post-op. Fixes: 1561129 Change-Id: Ie58a41e8fa1ad79aa06093706e96db8eef61b6d9 fixes: bz#1561129 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
* glusterd: mark port_registered to true for all running bricks with brick muxAtin Mukherjee2018-04-053-2/+16
| | | | | | | | | | | | | | | | | | | | | | | | glusterd maintains a boolean flag 'port_registered' which is used to determine if a brick has completed its portmap sign in process. This flag is (re)set in pmap_sigin and pmap_signout events. In case of brick multiplexing this flag is the identifier to determine if the very first brick with which the process is spawned up has completed its sign in process. However in case of glusterd restart when a brick is already identified as running, glusterd does a pmap_registry_bind to ensure its portmap table is updated but this flag isn't which is fine in case of non brick multiplex case but causes an issue if the very first brick which came as part of process is replaced and then the subsequent brick attach will fail. One of the way to validate this is to create and start a volume, remove the first brick and then add-brick a new one. Add-brick operation will take a very long time and post that the volume status will show all other brick status apart from the new brick as down. Solution is to set brickinfo->port_registered to true for all the running bricks when brick multiplexing is enabled. Change-Id: Ib0662d99d0fa66b1538947fd96b43f1cbc04e4ff Fixes: bz#1560957 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* features/changelog: Update option levelsAravinda VK2018-04-051-0/+7
| | | | | | | | Options levels for Changelog Xlator Change-Id: Idd246717e38096c44258a990a0939f82e5fc9654 Updates: #430 Signed-off-by: Aravinda VK <avishwan@redhat.com>
* cluster/dht: enable lookup-optimize by defaultN Balachandran2018-04-043-3/+5
| | | | | | | | | | | | | | Lookup-optimize has been shown to improve create performance. The code has been in the project for several years and is considered stable. Enabling this by default in order to test this in the upstream regression runs. Change-Id: Iab792979ee34f0af4713931e0b5b399c23f65313 updates: bz#1557435 BUG: 1557435 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* glusterd: fix txn_opinfo memory leakAtin Mukherjee2018-04-043-9/+25
| | | | | | | | | | | | | For transactions where there's no volname involved (eg : gluster v status), the originator node initiates with staging phase and what that means in op-sm there's no unlock event triggered which resulted into a txn_opinfo dictionary leak. Credits : cynthia.zhou@nokia-sbell.com Change-Id: I92fffbc2e8e1b010f489060f461be78aa2b86615 Fixes: bz#1550339 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* glusterd: honour localtime-logging for all the daemonsAtin Mukherjee2018-04-035-0/+30
| | | | | | Change-Id: I97a70d29365b0a454241ac5f5cae56d93eefd73a Fixes: bz#1563334 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* cluster/afr: Prevent ping-event handling on shdPranith Kumar K2018-04-031-0/+2
| | | | | | | | | On shd, we shouldn't treat any brick down based on latency, otherwise self-heal will never happen fixes: bz#1562717 Change-Id: Ica07fcc4fae91a6bfd9c9a670e2be464704d94b7 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* glusterd: setting mgmt_v3_timer->timer to NULL after deleting mgmt_v3_timerSanju Rakonde2018-04-021-1/+0
| | | | | | | | | | | We are setting mgmt_v3_timer->timer to NULL after mgmt_v3_timer is deleted which is unnecessary. So removing the statement. This issue is caught while running glusterd with ASAN. Change-Id: Ied1f91590a2c64ec1af36d4de9c3febd6cf94bb9 Fixes: bz#1562907 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* mount/fuse: Set default fuse reader thread count to 1Krutika Dhananjay2018-04-021-1/+1
| | | | | | | Updates #412 Change-Id: Ida53d8b630feabb856a3551fa888f92382ade768 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
* ocf: use glusterd-workdir for finding volume status filesNiels de Vos2018-04-021-1/+1
| | | | | | | | | The volume status files are located in the glusterd-workdir, not under /etc (sysconfdir). BUG: 1234873 Change-Id: Id7f7c83261bb4b5ac2fc104dcd6cb198d6a930aa Signed-off-by: Niels de Vos <ndevos@redhat.com>
* build: revert configure --without-ipv6-default behaviourKaleb S. KEITHLEY2018-04-021-14/+31
| | | | | | | | | | | | | | | | | | | | | | | | | patch https://review.gluster.org/19692 breaks gluster on systems with IPv6 enabled but don't have IPv6 reverse DNS. Also it defaulted to enabling ipv6-default regardless of whether --with-ipv6-default or --without-ipv6-default were specified in the options to configure. (Also the patch was merged without review.) Prefer libtirpc over glibc rpc. On newer linux with tirpc and without glibc rpc use tirpc (obviously) on less new linux with both tirpc and glibc rpc default to use tirpc, unless --without-tirpc is specified, in which case use glibc rpc On less new linux without tirpc fall back to glib rpc (obviously) ipv6-default requires libtirpc. It is off by default. It must be explicitly enabled with --with-ipv6-default. If --with-ipv6-default is specified, but tirpc is not available, disable it and issue a warning Change-Id: Ib96a230fafb83ec83a71948fe55af1215a7a6ffa BUG: 1562052 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
* cluster/dht: Update dht option levelsN Balachandran2018-04-021-2/+16
| | | | | | | | | Set the levels for DHT options based on https://review.gluster.org/#/c/19466/ Change-Id: I51b31a706a0b9517404e83224c89de145fd5d7e1 updates: #430 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* mount/fuse: Add support for multi-threaded fuse readersKrutika Dhananjay2018-04-028-83/+196
| | | | | | | | | | | | | | Usage: Use 'reader-thread-count=<NUM>' as command line option to set the thread count at the time of mounting the volume. Next task is to make these threads auto-scale based on the load, instead of having the user remount the volume everytime to change the thread count. Updates #412 Change-Id: I94aa1505e5ae6a133683d473e0e4e0edd139b76b Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
* cluster/dht: Update layout in inode only on successN Balachandran2018-04-022-4/+24
| | | | | | | | | | | | | | | | | | | | | With lookup-optimize enabled, gf_defrag_settle_hash in rebalance sometimes flips the on-disk layout on volume root post the migration of all files in the directory. This is sometimes seen when attempting to fix the layout of a directory multiple times before calling gf_defrag_settle_hash. dht_fix_layout_of_directory generates a new layout in memory but updates it in the inode ctx before it is set on disk. The layout may be different the second time around due to dht_selfheal_layout_maximize_overlap. If the layout is then not written to the disk, the inode now contains the wrong layout. gf_defrag_settle_hash does not check the correctness of the layout in the inode before updating the commit-hash and writing it to the disk thus changing the layout of the directory. Change-Id: Ie1407d92982518f2a0c40ec70ad370b34a87b4d4 updates: bz#1557435 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* Revert "glusterd: handling brick termination in brick-mux"Sanju Rakonde2018-03-297-147/+39
| | | | | | | | | | | | | This reverts commit a60fc2ddc03134fb23c5ed5c0bcb195e1649416b. This commit was causing multiple tests to time out when brick multiplexing is enabled. With further debugging, it's found that even though the volume stop transaction is converted into mgmt_v3 to allow the remote nodes to follow the synctask framework to process the command, there are other callers of glusterd_brick_stop () which are not synctask based. Change-Id: I7aee687abc6bfeaa70c7447031f55ed4ccd64693 updates: bz#1545048
* afr: add new value for read-hash-mode volume optionRavishankar N2018-03-297-32/+175
| | | | | | | | | | Updates: #363 This new value (3) will try to wind read requests to the child of AFR having the least amount of pending requests in its queue. Change-Id: If6bda2aac9bf7aec3fc39622f78659313c4b6508 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
* cluster/ec: send list-node-uuids request to all subvolumesXavi Hernandez2018-03-282-1/+2
| | | | | | | | | | | | The xattr trusted.glusterfs.list-node-uuids was only sent to a single subvolume. This was returning null uuids from the other subvolumes as if they were down. This fix forces that xattr to be requested from all subvolumes. Change-Id: If62eb39a6857258923ba625e153d4ad79018ea2f fixes: bz#1561406 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
* Fix gluster(8) formattingMichael Scherer2018-03-281-1/+0
| | | | | | | | | Looking at the man page show that "Snapshot command" wasn't aligned with the other section titles. Change-Id: I24bdb2e3728e03862fee57710cfe34b0607fe09a BUG: 1507230 Signed-off-by: Michael Scherer <misc@redhat.com>
* glusterd: changing the op-version of volume stop mgmt v3Kaleb S. KEITHLEY2018-03-281-3/+3
| | | | | | | | log message describe the actual test Change-Id: I1ea7300a6b186032a65236492d6d2a6eef0ab983 fixes: bz#1560441 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
* rpc: update tirpc registration to "force" unregister old mapping before ↵Shreyas Siravara2018-03-282-0/+8
| | | | | | | | | | | | re-registering > Reviewed-on: https://review.gluster.org/16849 > Reviewed-by: Shreyas Siravara <sshreyas@fb.com> Change-Id: I05ed6b7c715a71e5819fbe8116e7c3146010f836 BUG: 1521030 Signed-off-by: Kevin Vigor <kvigor@fb.com> Signed-off-by: Amar Tumballi <amarts@redhat.com>
* rpc: simplify parameters when a saved frame is forced to unwindZhang Huan2018-03-281-4/+2
| | | | | | | | | When a saved frame is to be forced unwind, there is no need to pass an empty iovector without any data pointed to. Change-Id: I6e858fb38644326e22239b83272b15db656035e5 BUG: 1523122 Signed-off-by: Zhang Huan <zhanghuan@open-fs.com>
* rpc: fix incorrect return value when xdr decode failsZhang Huan2018-03-281-1/+0
| | | | | | | | | | | | | xdr_replymsg is called to decode reply message, and it returns failure if the message is corrupted. However, retrieving return value from the global errno is 0 even xdr_replymsg fails. Fix this issue by simply returning a negative value if call to xdr_replymsg fails. Change-Id: I2b9a1dc97652fbb6cf6568ea617f120713784a55 BUG: 1523122 Signed-off-by: Zhang Huan <zhanghuan@open-fs.com>
* glusterd: handling brick termination in brick-muxSanju Rakonde2018-03-287-39/+147
| | | | | | | | | | | | | | | Problem: There's a race between the last glusterfs_handle_terminate() response sent to glusterd and the kill that happens immediately if the terminated brick is the last brick. Solution: When it is a last brick for the brick process, instead of glusterfsd killing itself, glusterd will kill the process in case of brick multiplexing. And also changing gf_attach utility accordingly. Change-Id: I386c19ca592536daa71294a13d9fc89a26d7e8c0 fixes: bz#1545048 BUG: 1545048 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* cluster/dht: ENOSPC will not fail rebalanceN Balachandran2018-03-281-6/+2
| | | | | | | | | ENOSPC returned by a file migration is no longer considered a rebalance failure. Change-Id: I21cf3a8acdc827bc478e138d6cb5db649d53a28c fixes: bz#1553598 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* Quota: heal directory on newly added bricks when quota limit is reachedSanoj Unnikrishnan2018-03-288-6/+281
| | | | | | | | | | | | | | | | | Problem: if a lookup is done on a newly added brick for a path on which limit has been reached, the lookup fails to heal the directory tree due to quota. Solution: Tag the lookup as an internal fop and ignore it in quota. Since marking internal fop does not usually give enough contextual information. Introducing new flags to pass the contextual info. Adding dict_check_flag and dict_set_flag to aid flag operations. A flag is a single bit in a bit array (currently limited to 256 bits). Change-Id: Ifb6a68bcaffedd425dd0f01f7db24edd5394c095 fixes: bz#1505355 BUG: 1505355 Signed-off-by: Sanoj Unnikrishnan <sunnikri@redhat.com>
* quick-read: Provide statistics to the monitorPoornima G2018-03-282-26/+89
| | | | | | | Updates: #425 Change-Id: Iea5198821f4eabc46bc63529afa4a92d4b4c2be0 Signed-off-by: Poornima G <pgurusid@redhat.com>
* glusterd: changing the op-version of volume stop mgmt v3Sanju Rakonde2018-03-271-1/+1
| | | | | | Change-Id: Iefc5a00d36436b23181871fa365f27b8d90cff0a fixes: bz#1560441 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* glfs heal binary change to accomodate socket file argumentsVishal Pandey2018-03-271-8/+18
| | | | | | | Change-Id: I755d6552decd015aec7859ad2cf99c76c8bee9dc fixes: bz#1558380 BUG: 1558380 Signed-off-by: Vishal Pandey <vpandey@redhat.com>
* glusterd: Implementing volume stop in mgmt v3Sanju Rakonde2018-03-262-1/+66
| | | | | | Change-Id: I8f9c594cf56331d54eb4884335699744685ef20d fixes: bz#1560441 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* tests: fix nl-cache.t failureAtin Mukherjee2018-03-261-1/+1
| | | | | | | | | | commit fef9293 changed network.inode-lru-limit from 50000 to 200000 in nl-cache group profile but the test wasn't changed to reflect it accordingly. Change-Id: Ibb5fb0a387f160f6b726246b161a9a7b33135755 fixes: bz#1560589 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
* .testignore: remove the group profile filesAmar Tumballi2018-03-261-4/+0
| | | | | | Change-Id: I6619827f1bf6fe9bd974537af6169164b19a0aa5 fixes: bz#1560393 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* glusterfind: Log remote stderr on `node_cmd` error. Fixes #1559130Niklas Hambüchen2018-03-261-1/+16
| | | | | | | | | The problem of lost stderr was introduced in commit feea851fad4f89b48bfe89fe3b75250cc7bd6501. Change-Id: Ic98f9bc9682ae3bd9c3ebea3855667fc8ba2843d BUG: 1559130 Signed-off-by: Niklas Hambüchen <mail@nh2.me>
* md-cache: fix ./tests/basic/md-cache/bug-1418249.tSusant Palai2018-03-261-1/+1
| | | | | | | | | inode table size is currently set to 200000. Hence the need of change in testcase which was expecting the old value 50000. Change-Id: I8e44b1d0a2da1e8100bebd25f48bb36e2897b4f8 fixes: bz#1560393 Signed-off-by: Susant Palai <spalai@redhat.com>
* extras/group: Change the server inode table size when upcall is onPoornima G2018-03-242-2/+2
| | | | | | | | | | | | | | | By default server inode table size is 16K, when upcall is enabled, there is going to be too many forgets sent on inodes as the brick can hold only 16K inodes in memory, so we increased this to 50K. This is still less than the client inode table size. We have seen performance improvement when server inode table size is set to 200000(almost as client inode table size). Hence changing the value to 200000. Increasing this increases the memory consumption by <1MB. BUG: 1559235 Change-Id: I931db965cd34bf33094328541bd5a633b3357805 Signed-off-by: Poornima G <pgurusid@redhat.com>
* nl-cache: Provide statistics to the monitorPoornima G2018-03-241-9/+61
| | | | | | | Updates: #429 Change-Id: Ic2e64422055f1838d5d453643c739ef1e9319cfe Signed-off-by: Poornima G <pgurusid@redhat.com>
* md-cache: Provide statistics to the monitorPoornima G2018-03-241-9/+57
| | | | | | | Updates: #427 Change-Id: Ib1f45016ac75d7bc2755db0dd4b68ce1d95d26c3 Signed-off-by: Poornima G <pgurusid@redhat.com>
* features/quota: Add new fields to translator options for GD2Sanoj Unnikrishnan2018-03-242-29/+51
| | | | | | | | | | | | alert-time, soft timeout, hard timeout, default soft limit and deem-statfs will be settable through volume set command. hence marked as settable. Other options are used only via quota commands. Updates #302 Change-Id: I02d258cc3aa7fe58ccbadd59441cce64cfd9ba6e Signed-off-by: Sanoj Unnikrishnan <sunnikri@redhat.com>
* build: Fix misleading TIRPC result in configure summaryJames Le Cuirot2018-03-241-5/+4
| | | | | | | | | | | | | | Requesting ipv6-default even if you explicitly disable libtirpc will then implicitly enable libtirpc because that is required. That is fine but the configure summary should not then show TIRPC as disabled when it is not. The result has also been made clearer by stating that TIRPC is "missing" when it has been tried but not found. BUG: 1553938 Change-Id: I945bd6859aaf3defa682b0d05ee34a9827b9c45f Signed-off-by: James Le Cuirot <chewi@gentoo.org>
* build: Fix configure --without-ipv6-default behaviourJames Le Cuirot2018-03-241-1/+1
| | | | | | | | | | | | | | | | The current behaviour disables ipv6-default when no switch is given at all but otherwise checks if libtirpc was requested, regardless of whether you have given --with-ipv6-default or --without-ipv6-default. I believe the intention was to enable when libtirpc is requested by default but otherwise respect the switch given. This is important because ipv6-default breaks Gluster for systems that have IPv6 disabled. BUG: 1553926 Change-Id: I76b91ae2699574b2e5b777453732bb5cbd79bbca Signed-off-by: James Le Cuirot <chewi@gentoo.org>
* libgfchangelog: Correct the log messageNiklas Hambüchen2018-03-241-1/+1
| | | | | | | | | | | | Provide correct error message for changelog end time check Updated error message to print "wrong result for end". Original patch by Keith Schincke <kschinck@redhat.com> from https://review.gluster.org/#/c/8121/ Change-Id: Ia3458cbac7784bfc71c05da67391a3f8259f18f0 BUG: 1559126 Signed-off-by: Niklas Hambüchen <mail@nh2.me>
* python: Remove all uses of find_library. Fixes #1450593Niklas Hambüchen2018-03-247-20/+9
| | | | | | | | `find_library()` doesn't consider LD_LIBRARY_PATH on Python < 3.6. Change-Id: Iee26085cb5d14061001f19f032c2664d69a378a8 BUG: 1450593 Signed-off-by: Niklas Hambüchen <mail@nh2.me>
* glusterfind: Show C function when raising ChangelogExceptionNiklas Hambüchen2018-03-241-9/+10
| | | | | | | Fixes: #432 Change-Id: I9ab031e098aff717e619d9deb6410281b96de14a Signed-off-by: Niklas Hambüchen <mail@nh2.me>
* rfc.sh: provide a unified way to update bugs or github issues IDAmar Tumballi2018-03-223-73/+91
| | | | | | Change-Id: Ie78d87b31512da6201ae26f3d391fa3f8e5b68d1 fixes: bz#1545891 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* client: make fuse direct I/O strategies explicitCsaba Henk2018-03-223-9/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | So far the --direct-io-mode option has been presented as of being Boolean valued. That is however not exact, as a third behavior is chosen if the option is not specified. We accept now the "auto" value as an explicit choice for the default heuristics, and indicate in the descriptions of the option (which occur in commandline help and in the gluterfs / mount.glusterfs man pages) that auto is the default. The default heuristics was briefly described in the commandline help. We are getting rid of that, because: - it's not the right place to provide such details; - there is no guarantee of keeping the current heuristics so it might go out of sync with reality; - that is already the case to some degree, because the description did not take into account that the default heuristics varies between platforms (on Mac, it's just "off"), and that xlators can also prescribe direct I/O for the file of their choice (see change I3fe3312cd96baa4eecfe1247ab7255b4f455f049). Change-Id: Ia83479c0c67fe66b7fc2e0e8db5b7792d9f44b28 Signed-off-by: Csaba Henk <csaba@redhat.com>
* rpcsvc: enable ownthread feature for glusterfs4_0_fop_progMilind Changire2018-03-221-0/+1
| | | | | | | | Ownthread feature needs enabling for glusterfs4_0_fop_prog Change-Id: Idce63eb094ae0fdfcddbd52d0dee25aa0e074926 BUG: 1559075 Signed-off-by: Milind Changire <mchangir@redhat.com>
generated by cgit (git 2.34.1) at 2024-11-15 13:24:58 +0000