summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
* posix: Convert several posix_private members to gf_atomicMohit Agrawal2019-01-155-48/+18
| | | | | | Change-Id: I629698d8ddf6f15428880bdc1501d36bc37b8ebb fixes: bz#1657607 Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
* core: Resolve dict_leak at the time of destroying graphMohit Agrawal2019-01-141-4/+0
| | | | | | | | | | | | Problem: In gluster code some of the places it call's get_new_dict to create a dictionary without taking reference so at the time of dict_unref it has become a leak Solution: To resolve the same call dict_new instead of get_new_dict updates bz#1650403 Change-Id: I3ccbbf5af07079a4fa09aad2cd0458c8625b2f06 Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
* cluster/afr: fix zerofill transaction.startXiubo Li2019-01-141-1/+1
| | | | | | | | This maybe one mistake when coding. Fixes: bz#1665332 Change-Id: Ia8f8dadf4a71579240ff9950b141ca528bd342b3 Signed-off-by: Xiubo Li <xiubli@redhat.com>
* glusterd: fix crashSanju Rakonde2019-01-131-3/+2
| | | | | | | | | | | | | | | | | | Problem: running "gluster get-state glusterd odir /get-state" resulted in glusterd crash. Cause: In the above command output directory has been specified without "/" at the end. If "/" is not given at the end, "/" will be added to path using "strcat", so the added character "/" is not having memory allocated. When tried to free, glusterd will crash as"/" has no memory allocated. Solution: Instead of concatenating "/" to output directory, add it to output filename. Change-Id: I5dc00a71e46fbef4d07fe99ae23b36fb60dec1c2 fixes: bz#1665038 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* features/index: Correct incorrect definition in headerVijay Bellur2019-01-111-2/+2
| | | | | | | | | s/QUIESCE/INDEX/ fixes: bz#1665363 Change-Id: I6dc4fde682cedeaa10d870267b8909af1a9449c0 Signed-off-by: Vijay Bellur <vbellur@redhat.com>
* fix 32-bit-build-smoke warningsIraj Jamali2019-01-1112-27/+31
| | | | | | | fixes: bz#1622665 Change-Id: I777d67b1b62c284c62a02277238ad7538eef001e Signed-off-by: Iraj Jamali <ijamali@redhat.com>
* features/shard: Fix launch of multiple synctasks for background deletionKrutika Dhananjay2019-01-112-71/+128
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PROBLEM: When multiple sharded files are deleted in quick succession, multiple issues were observed: 1. misleading logs corresponding to a sharded file where while one log message said the shards corresponding to the file were deleted successfully, this was followed by multiple logs suggesting the very same operation failed. This was because of multiple synctasks attempting to clean up shards of the same file and only one of them succeeding (the one that gets ENTRYLK successfully), and the rest of them logging failure. 2. multiple synctasks to do background deletion would be launched, one for each deleted file but all of them could readdir entries from .remove_me at the same time could potentially contend for ENTRYLK on .shard for each of the entry names. This is undesirable and wasteful. FIX: Background deletion will now follow a state machine. In the event that there are multiple attempts to launch synctask for background deletion, one for each file deleted, only the first task is launched. And if while this task is doing the cleanup, more attempts are made to delete other files, the state of the synctask is adjusted so that it restarts the crawl even after reaching end-of-directory to pick up any files it may have missed in the previous iteration. This patch also fixes uninitialized lk-owner during syncop_entrylk() which was leading to multiple background deletion synctasks entering the critical section at the same time and leading to illegal memory access of base inode in the second syntcask after it was destroyed post shard deletion by the first synctask. Change-Id: Ib33773d27fb4be463c7a8a5a6a4b63689705324e updates: bz#1662368 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
* core: brick process is crashed at the time of spawn threadMohit Agrawal2019-01-113-18/+35
| | | | | | | | | | | | | Problem: brick is getting crashed at the time of calling pthread_detach after just call gf_thread_create.If sufficient resources are not available on the system pthread_create returns EAGAIN (non-negative) but the caller function expects negative error code in case of failure Solution: Change the condition in caller function to avoid the crash Change-Id: Ifeaa49f809957eb6c33aa9792f5af1b55566756d fixes: bz#1662906
* performance/md-cache: Fix a crash when statfs caching is enabledVijay Bellur2019-01-111-2/+2
| | | | | | | | | | | | | mem_put() in STACK_UNWIND_STRICT causes a crash if frame->local is not null as md-cache obtains local from CALLOC. Changed two occurrences of STACK_UNWIND_STRICT to MDC_STACK_UNWIND as the latter macro does not rely on STACK_UNWIND_STRICT for cleaning up frame->local. fixes: bz#1632503 Change-Id: I1b3edcb9372a164ef73119e99a49e747765d7166 Signed-off-by: Vijay Bellur <vbellur@redhat.com>
* afr : fix memory leakSunny Kumar2019-01-111-23/+15
| | | | | | | | | | | | | | | This patch fixes memory leak reported by ASan. The fix was first merged by https://review.gluster.org/#/c/glusterfs/+/21805. But later change was reverted due to this patch https://review.gluster.org/#/c/glusterfs/+/21178/. updates: bz#1633930 Change-Id: I1febe121e0be33a637397a0b54d6b78391692b0d Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
* cluster/afr: Disable client side heals in AFR by default.Sunil Kumar Acharya2019-01-101-3/+3
| | | | | | | | | With this changeset, default value for the AFR client side heal volume option is set to "off" fixes: bz#1663102 Change-Id: Ie4016932339c4896487e3e7cb5caca68739b7ba2 Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
* dht: Add NULL check for stbuf in dht_rmdir_lookup_cbkSusant Palai2019-01-101-2/+1
| | | | | | Change-Id: I2ced288113a369cc6497a77ac1871007df434da4 fixes: bz#1664647 Signed-off-by: Susant Palai <spalai@redhat.com>
* features/locks: Dump connection_id even for posix locksKrutika Dhananjay2019-01-091-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | ... in statedump for a better debugging experience. BEFORE: posixlk.posixlk[0](ACTIVE)=type=WRITE, whence=0, start=0, len=0, pid = 13635, owner=2dd2c3a11706dc8c, client=0x7f159012b000, connection-id=(null), granted at 2018-12-31 14:20:42 connection-id is null above. AFTER: posixlk.posixlk[0](ACTIVE)=type=WRITE, whence=0, start=0, len=0, pid = 10977, owner=b485e33df21bdaa2, client=0x7fa24c01ab90, connection-id=CTX_ID:68e12340-eed2-4386-bf5e-1f43cf8693d9-GRAPH_ID:0- PID:10901-HOST:dhcp35-215.lab.eng.blr.redhat.com-PC_NAME:patchy-client-0- RECON_NO:-0, granted at 2018-12-31 14:33:50 Change-Id: I4608994bacabb558a3be8c1634ee6b1d2d3022e2 fixes: bz#1662679 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
* features/shard: Assign fop id during background deletion to prevent ↵Krutika Dhananjay2019-01-081-0/+1
| | | | | | | | | | | | | | | | | | | | | | excessive logging ... of the kind "[2018-12-26 05:22:44.195019] E [MSGID: 133010] [shard.c:2253:shard_common_lookup_shards_cbk] 0-volume1-shard: Lookup on shard 785 failed. Base file gfid = cd938e64-bf06-476f-a5d4-d580a0d37416 [No such file or directory]" shard_common_lookup_shards_cbk() has a specific check to ignore ENOENT error without logging them during specific fops. But because background deletion is done in a new frame (with local->fop being GF_FOP_NULL), the ENOENT check is skipped and the absence of shards gets logged everytime. To fix this, local->fop is initialized to GF_FOP_UNLINK during background deletion. Change-Id: I0ca8d3b3bfbcd354b4a555eee520eb0479bcda35 updates: bz#1662368 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
* Revert "iobuf: Get rid of pre allocated iobuf_pool and use per thread mem pool"Amar Tumballi2019-01-081-1/+2
| | | | | | | | | This reverts commit b87c397091bac6a4a6dec4e45a7671fad4a11770. There seems to be some performance regression with the patch and hence recommended to have it reverted. Updates: #325 Change-Id: Id85d6203173a44fad6cf51d39b3e96f37afcec09
* features/bit-rot: do not send version and signature keys in dictRaghavendra Bhat2019-01-082-10/+53
| | | | | | | | | | | In lookup, if the file has been marked as bad, then bit-rot-stub was sending the version and signature xattr values as well in the response dictinary. This is not needed. Only bad file marker has to be sent. Change-Id: Id59c02e9857577c60849fd28ef657f71e0b15207 fixes: bz#1664122 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* performance/io-threads: Improve debuggability in statedumpVijay Bellur2019-01-071-6/+23
| | | | | | | | | | | | | | | | | | | | | | statedump from io-threads lacked information to understand the number of running threads & number of requests in each priority queue. This patch addresses that. Sample statedump output w/ this patch: current_high_priority_threads=7 current_normal_priority_threads=9 current_low_priority_threads=0 current_least_priority_threads=0 fast_priority_queue_length=32 normal_priority_queue_length=45 Also, changed the wording for least priority queue in iot_get_pri_meaning(). Change-Id: Ic5f6391a15cc28884383f5185fce1cb52e0d10a5 fixes: bz#1664124 Signed-off-by: Vijay Bellur <vbellur@redhat.com>
* gfapi: update returned/callback pre/post attributes to glfs_statShyamsundarR2019-01-071-2/+4
| | | | | | | Change-Id: Ie0fe971e694101aa011d66aa496d0644669c2c5a Updates: #389 Signed-off-by: Kinglong Mee <mijinlong@open-fs.com> Signed-off-by: ShyamsundarR <srangana@redhat.com>
* leases: Reset lease_ctx->timer post deletionSoumya Koduri2019-01-071-0/+1
| | | | | | | | | To avoid use_after_free, reset lease_ctx->timer back to NULL after the structure has been freed. Change-Id: Icd213ec809b8af934afdb519c335a4680a1d6cdc updates: bz#1648768 Signed-off-by: Soumya Koduri <skoduri@redhat.com>
* glusterd: rebalance status should display information of localhostSanju Rakonde2019-01-041-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: https://review.gluster.org/#/c/glusterfs/+/21762/ has migrated rebalance commands from op-sm framework to mgmt v3 framework. In a heterogenous cluster, if rebalance commands follow op-sm framework, localhost information is not displayed in the output of "gluster v rebalance <volname> status". Cause: Previously without https://review.gluster.org/#/c/glusterfs/+/21762/ rebalance commands were following op-sm framework. In glusterd_volume_rebalance_use_rsp_dict() current_index variable keeps track of number/count of peers in trusted storage pool. In op-sm, glusterd_volume_rebalance_use_rsp_dict() will be called only for the peers. So the current index should start from 2 assuming local host as node 1. With the above patch, rebalance commands are following mgmt v3 framework. In mgmt v3, glusterd_volume_rebalance_use_rsp_dict() is called for all nodes. For localhost it is called from brick-op function and for peers it is called from brick-op call back function. So the current index value should start from 1. https://review.gluster.org/#/c/glusterfs/+/21762/ has changed the value of current index to 1. Because of this, In heterogenous cluster, local host's information is overwritten by one of the peers information. And rebalance status will not display localhost's information in the output. Solution: assign a value to current index based on a op-version check. Change-Id: I2dfba1f007e908cf160acc4a4a5d8ef672572e4d fixes: bz#1663243 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* glusterd: aggregate rsp from peers for profile commandSanju Rakonde2019-01-031-1/+2
| | | | | | | | | | | | | | | | | When we run profile info command, it should display statistics of all the bricks of the volume. To display information of bricks which are hosted on peers, we need to aggregate the response from peers. For profile info command, all the statistics will be added into the dictionary in brick-op phase. To aggregate the information from peers, we need to call glusterd_syncop_aggr_rsp_dict() in brick-op call back function. fixes: bz#1663223 Change-Id: I5f5890c3d01974747f829128ab74be6071f4aa30 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* glusterd: fix memory leak in handshakeZhang Huan2019-01-031-0/+3
| | | | | | | | | Add missing unref to req_dict to fix memory leak in handle of handshake. Change-Id: I0d8573fc3668c1a0ccc9030e3a096bbe20ed5c36 fixes: bz#1663077 Signed-off-by: Zhang Huan <zhanghuan@open-fs.com>
* posix: buffer_size_warning coverity fixHarpreet Kaur2018-12-311-1/+1
| | | | | | | | | | | | | Defect: CID 1398469- Calling strncpy with a maximum size argument of 4096 bytes on destination array key of size 4096 bytes might leave the destination string unterminated. Fix: Using snprintf instead of strncpy. updates: bz#789278 Change-Id: I4fdcd0cbf3af8b2ded94603d92d1ceb4112284c4 Signed-off-by: Harpreet Kaur <hlalwani@redhat.com>
* multiple-files: clang-scan fixesAmar Tumballi2018-12-314-4/+20
| | | | | | updates: bz#1622665 Change-Id: I9f3a75ed9be3d90f37843a140563c356830ef945 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* glusterd: NULL pointer dereferencing clang fixSheetal Pamecha2018-12-291-1/+1
| | | | | | | | | Added ternary operator to avoid this issue Updates: bz#1622665 Change-Id: I163d0628304a0d61249d1d97a4a3d3bee4ba4927 Signed-off-by: Sheetal Pamecha <sheetal.pamecha08@gmail.com>
* mgmt/glusterd: fix clang warningSheetal Pamecha2018-12-291-1/+0
| | | | | | | | Attempt to free rsp.dict.dict_val twice Change-Id: I5dbc50430f59ca8d0c739b0fbe95d71981852889 Updates: bz#1622665 Signed-off-by: Sheetal Pamecha <sheetal.pamecha08@gmail.com>
* cluster/afr: Refactor internal locking code to allow multiple inodelksPranith Kumar K2018-12-286-798/+366
| | | | | | | | | | | | | | | | For implementing copy_file_range fop, AFR needs to perform two inodelks in the same transaction. This patch brings in the necessary structure to make it easier to do so. Entry-locks in AFR were already taking multiple entry-locks on different inodes with the respective basenames. This patch extends the logic in inodelks to use the same lockee_t structure. This lead to removal of quite a lot of duplicate code present in afr-lk-common.c as both the locks are doing same things except 'winding' part. updates: #536 Change-Id: Ibfce7e3f260bb27b18645152ec680c33866fe0ae Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* barrier: replace boolean-switch statement with if/elseNiels de Vos2018-12-281-43/+23
| | | | | | | | | | | | | | | | | | | | | | Squash some ugly warnings, and make the code a little bit simpler by removing some unneeded goto jumps. On Ubuntu 16.04 the following warnings were reported by Amudhan: CC barrier.lo barrier.c: In function ‘notify’: barrier.c:499:33: warning: switch condition has boolean value [-Wswitch-bool] switch (past) { ^ barrier.c: In function ‘reconfigure’: barrier.c:565:25: warning: switch condition has boolean value [-Wswitch-bool] switch (past) { ^ Change-Id: Ifb6b75058dff8c789b729c76530a1358d391f4d1 Updates: bz#1193929 Reported-by: Amudhan P <amudhan83@gmail.com> Signed-off-by: Niels de Vos <ndevos@redhat.com>
* glusterd: coverity fixesSanju Rakonde2018-12-272-0/+3
| | | | | | | | | | | | This patch addresses coverity issues with CID 1398470 and 1398475 1398470 - Missing unlock - False positive, Added a annotation to make coverity happy 1398475 - Unused value Change-Id: I1bb3df0b716690fad8fc52c393c8b2b6c41f7860 updates: bz#789278 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* core: Fixed typos in nl-cache and logging-guidelines.mdN Balachandran2018-12-261-2/+2
| | | | | | | | Replaced "recieve" with "receive". Change-Id: I58a3d3d4a0093df4743de9fae4d8ff152d4b216c fixes: bz#1662089 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cluster/dht: refactor dht_lookup_cbkN Balachandran2018-12-261-93/+101
| | | | | | | | | | Rearrange the dht_lookup_cbk code to make it easier to understand. Corrected a message in dht_linkfile_create_lookup_cbk Change-Id: Id41db9ef901732f0410f1c007807362c630218ff fixes: bz#1590385 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* posix: stack-buffer-overflow reported by asanHarpreet Kaur2018-12-262-2/+5
| | | | | | | | | | | | This patch fixes buffer overflow in $SRC/xlators/storage/posix/src/posix-inode-fd-ops.c Memory access at offset 432 overflows "md5_checksum" variable. SUMMARY: AddressSanitizer: stack-buffer-overflow (/lib64/libasan.so.5+0xb825a) updates: bz#1633930 Change-Id: I46010a09161d02cdf0c69679a334ec1d3d49cffb Signed-off-by: Harpreet Kaur <hlalwani@redhat.com>
* meta : fix coverity issueSunny Kumar2018-12-231-1/+1
| | | | | | | | This patch fixes coverity issue possible null dereference. Change-Id: I93c0847c3d93b29a1e001ed044a63e908c670167 updates: bz#789278 Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
* all: handle string-overflow warnings of coverityAmar Tumballi2018-12-202-29/+46
| | | | | | updates: bz#789278 Change-Id: I7de800b90a614e3666e965b0cafc70026a844b2d Signed-off-by: Amar Tumballi <amarts@redhat.com>
* all: handle USE_AFTER_FREE warningsAmar Tumballi2018-12-204-13/+52
| | | | | | | | | | | | | | | | | | | * we shouldn't be using 'local' after DHT_STACK_UNWIND() as it frees the content of local. Add a 'goto out' or similar logic to handle the situation. * fix possible overlook of unref(dict), instead of unref(xdata). * make coverity happy by re-ordering unref in meta-defaults. * gfid-access: re-order dictionary allocation so we don't have to do a extra unref. * other obvious errors reported. updates: bz#789278 Change-Id: If05961ee946b0c4868df19861d7e4a927a2a2489 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* posix: use synctask for janitorPoornima G2018-12-198-82/+196
| | | | | | | | | | | | | | With brick mux, the number of threads increases as the number of bricks increases. As an initiative to reduce the number of threads in brick mux scenario, replacing janitor thread to use synctask infra. Now close() and closedir() handle by separate janitor thread which is linked with glusterfs_ctx. Updates #475 Change-Id: I0c4aaf728125ab7264442fde59f3d08542785f73 Signed-off-by: Poornima G <pgurusid@redhat.com>
* cluster/afr: Allow lookup on root if it is from ADD_REPLICA_MOUNTkarthik-us2018-12-188-31/+79
| | | | | | | | | | | | | | | | | | | | | Problem: When trying to convert a plain distribute volume to replica-3 or arbiter type it is failing with ENOTCONN error as the lookup on the root will fail as there is no quorum. Fix: Allow lookup on root if it is coming from the ADD_REPLICA_MOUNT which is used while adding bricks to a volume. It will try to set the pending xattrs for the newly added bricks to allow the heal to happen in the right direction and avoid data loss scenarios. Note: This fix will solve the problem of type conversion only in the case where the volume was mounted at least once. The conversion of non mounted volumes will still fail since the dht selfheal tries to set the directory layout will fail as they do that with the PID GF_CLIENT_PID_NO_ROOT_SQUASH set in the frame->root. Change-Id: Ic511939981dad118cc946754341318b164954b3b fixes: bz#1655854 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* iobuf: Get rid of pre allocated iobuf_pool and use per thread mem poolPoornima G2018-12-181-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current implementation of iobuf_pool has two problems: - prealloc of 12.5MB memory, this limits the scale factor of the gluster processes due to RAM requirements - lock contention, as the current implementation has one global iobuf_pool lock. Credits for debugging and addressing the same goes to Krutika Dhananjay <kdhananj@redhat.com>. Issue: #410 Hence changing the iobuf implementation to use per thread mem pool. This may theoritically appear to cause perf dip as there is no preallocation. But per thread mem pool will not have significant perf impact as the last allocated memory is kept alive for subsequent allocs, for some time. The worst case would be if iobufs requested are of random sizes each time. The best case is, if we get iobuf request of the same size. From the perf tests, this patch did not seem to cause any perf decrease. Note that, with this patch, the rdma performance is going to degrade drastically. In one of the previous patchsets we had fixes to not degrade rdma perf, but rdma is not supported and also not tested [1]. Hence the decision was to not have code in rdma that is not tested and not supported. [1] https://lists.gluster.org/pipermail/gluster-users.old/2018-July/034400.html Updates: #325 Change-Id: Ic2ef3bd498f9250dea25f25ba0c01fde19584b27 Signed-off-by: Poornima G <pgurusid@redhat.com>
* performance/io-cache: update pages with write dataRaghavendra Gowdappa2018-12-182-4/+90
| | | | | | | | | | | | Currently io-cache invalidate pages falling in the range of write. But instead it can update pages with same data so that reads can make use of the cache. credits: Xavi Hernandez <xhernandez@redhat.com> Change-Id: I932bd3da97ddfd464187f3009b1013eb334f00a7 Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com> updates: bz#1659869
* performance/ob: make open-behind as a child of quick-readRaghavendra Gowdappa2018-12-181-9/+7
| | | | | | | | | | | | | | | | | With read-after-open being set to yes by default, if open-behind sees any reads, it'll do an open on backend (and hence flush/release later). This means with the current order of quick-read and open-behind, open-behind sees all reads and hence also does open bringing down performance for small file reads. Since for small files, reads are absorbed by quick-read, if quick-read is made a parent of open-behind, ob doesn't witness any reads. For read-only workloads, this means ob doen't do any opens (even with read-after-open yes and use-anonymous-fd no). Change-Id: I138a42b006d104cff43ee6f07829e39c36f6f234 Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com> Fixes: bz#1659327
* xlators/cluster/afr/src/afr-self-heal-common.c: remove a variable array.Yaniv Kaul2018-12-181-10/+6
| | | | | | | | | | | | | Added '-Wvla' and saw this - gcc doesn't like variable arrays. There are plenty of others in the EC code, but this seems OK to remove: there is no use for the array members (I hope - that was from reading the code). Compile-tested only! updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com> Change-Id: I350f4520e52b86c8bbcd60eea1b27ef99cd119aa
* glusterd: migrating rebalance commands to mgmt_v3 frameworkSanju Rakonde2018-12-188-21/+630
| | | | | | | | | Current rebalance commands use the op_state machine framework. Porting it to use the mgmt_v3 framework. Change-Id: I6faf4a6335c2e2f3d54bbde79908a7749e4613e7 fixes: bz#1655827 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* Don't depend on string options to be valid alwaysPranith Kumar K2018-12-1711-67/+93
| | | | | | updates bz#1650403 Change-Id: Ib5a11e691599ce4bd93c1ed5aca6060592893961 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* features/snapview-client: access priv->path inside lockRaghavendra Bhat2018-12-173-75/+360
| | | | | | | | | | To handle the race condition of a fop or a function accessing priv->path and a reconfigure changing priv->path (because entry point directory changed), the private structure's path is guarded by the lock. updates bz#1650403 Change-Id: I61c539da06d68d38eafcf2155699c7702f31323e Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* AFR xlator: use dict_{setn|getn|deln|get_int32n|set_int32n|set_strn}Yaniv Kaul2018-12-1712-203/+282
| | | | | | | | | | | | | | | | | | | | In a previous patch (https://review.gluster.org/20769) we've added the key length to be passed to dict_* funcs, to remove the need to strlen() it. This patch moves some xlators to use it. - In some cases, moved strlen() of the key length outside of locks, which is usually a good thing. Please verify it's safe to do so. - In some cases, created a prefix for the keys, replacing something like "%d-%d" with a "%s" in snprintf(). Not sure it adds value, but improves readability. Please review carefully. Compile-tested only! Change-Id: I04f2a1eb2ecfc3283d849d150d10d088ae7aa7f1 updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* selinux/glusterd : add "features.selinux" to glusterd-volume-set.cJiffin Tony Thottan2018-12-171-0/+9
| | | | | | Fixes: bz#1659868 Change-Id: I38675ba4d47c8ba7f94cfb4734692683ddb3dcfd Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
* cluster/afr: Fix mem leak reported by ASANKotresh HR2018-12-171-4/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Traceback: Direct leak of 765 byte(s) in 9 object(s) allocated from: #0 0x7ffb9cad2c48 in malloc (/lib64/libasan.so.5+0xeec48) #1 0x7ffb9c5f8949 in __gf_malloc ./libglusterfs/src/mem-pool.c:136 #2 0x7ffb9c5f91bb in gf_vasprintf ./libglusterfs/src/mem-pool.c:236 #3 0x7ffb9c5f938a in gf_asprintf ./libglusterfs/src/mem-pool.c:256 #4 0x7ffb826714ab in afr_get_heal_info ./xlators/cluster/afr/src/afr-common.c:6204 #5 0x7ffb825765e5 in afr_handle_heal_xattrs ./xlators/cluster/afr/src/afr-inode-read.c:1481 #6 0x7ffb825765e5 in afr_getxattr ./xlators/cluster/afr/src/afr-inode-read.c:1571 #7 0x7ffb9c635af7 in syncop_getxattr ./libglusterfs/src/syncop.c:1680 #8 0x406c78 in glfsh_process_entries ./heal/src/glfs-heal.c:810 #9 0x408555 in glfsh_crawl_directory ./heal/src/glfs-heal.c:898 #10 0x408cc0 in glfsh_print_pending_heals_type ./heal/src/glfs-heal.c:970 #11 0x408fc5 in glfsh_print_pending_heals ./heal/src/glfs-heal.c:1012 #12 0x409546 in glfsh_gather_heal_info ./heal/src/glfs-heal.c:1154 #13 0x403e96 in main ./heal/src/glfs-heal.c:1745 #14 0x7ffb99bc411a in __libc_start_main ../csu/libc-start.c:308 The dictionary is referenced by caller to print the status. So set it as dynstr, the last unref of dictionary will free it. updates: bz#1633930 Change-Id: Ib5a7cb891e6f7d90560859aaf6239e52ff5477d0 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* dht: Fix clang warnings in dht-common.cShyamsundarR2018-12-161-20/+37
| | | | | | Change-Id: I0894d62edd68e13d123aaa5ca1827b98283f0d3e Updates: bz#1622665 Signed-off-by: ShyamsundarR <srangana@redhat.com>
* fuse: SETLKW interruptCsaba Henk2018-12-141-0/+130
| | | | | | | | | Use the (f)getxattr based clearlocks interface to interrupt a pending lock request. updates: #465 Change-Id: I4e91a4d8791fc688fed400a02de4c53487e61be2 Signed-off-by: Csaba Henk <csaba@redhat.com>
* fuse: add --lru-limit optionAmar Tumballi2018-12-143-51/+86
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The inode LRU mechanism is moot in fuse xlator (ie. there is no limit for the LRU list), as fuse inodes are referenced from kernel context, and thus they can only be dropped on request of the kernel. This might results in a high number of passive inodes which are useless for the glusterfs client, causing a significant memory overhead. This change tries to remedy this by extending the LRU semantics and allowing to set a finite limit on the fuse inode LRU. A brief history of problem: When gluster's inode table was designed, fuse didn't have any 'invalidate' method, which means, userspace application could never ask kernel to send a 'forget()' fop, instead had to wait for kernel to send it based on kernel's parameters. Inode table remembers the number of times kernel has cached the inode based on the 'nlookup' parameter. And 'nlookup' field is not used by no other entry points (like server-protocol, gfapi etc). Hence the inode_table of fuse module always has to have lru-limit as '0', which means no limit. GlusterFS always had to keep all inodes in memory as kernel would have had a reference to it. Again, the reason for this is, kernel's glusterfs inode reference was pointer of 'inode_t' structure in glusterfs. As it is a pointer, we could never free it (to prevent segfault, or memory corruption). Solution: In the inode table, handle the prune case of inodes with 'nlookup' differently, and call a 'invalidator' method, which in this case is fuse_invalidate(), and it sends the request to kernel for getting the forget request. When the kernel sends the forget, it means, it has dropped all the reference to the inode, and it will send the forget with the 'nlookup' parameter too. We just need to make sure to reduce the 'nlookup' value we have when we get forget. That automatically cause the relevant prune to happen. Credits: Csaba Henk, Xavier Hernandez, Raghavendra Gowdappa, Nithya B fixes: bz#1560969 Change-Id: Ifee0737b23b12b1426c224ec5b8f591f487d83a2 Signed-off-by: Amar Tumballi <amarts@redhat.com>
generated by cgit (git 2.34.1) at 2024-05-21 10:23:54 +0000