summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
* performance/readdir-ahead: indicate EOF for readdirpRaghavendra Bhat2014-12-171-2/+9
| | | | | | | | | | | | | | | | | | posix xlator sends op_errno as ENOENT and op_ret as 0, to indicate readdir has been completed. readdir-ahead should send that op_errno that it has saved in the fd context, when it serves the readdir requests. Otherwise some xlators sitting above performance xlators such as snapview-client, which checks for end of readdir operation by checking op_ret to 0 and op_errno to ENOENT will not be able to identify end of readdir. Change-Id: Ib0835136c61cb1e0d7df933226c479c7db703a71 BUG: 1174783 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/9283 Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* ec: Fix more mutex related coverity scan issuesXavier Hernandez2014-12-171-9/+5
| | | | | | | | | | | This patch solves CID 1257622. Change-Id: I95680c7de49cd84011d2ad38f02e5fad82e15c90 BUG: 1170254 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-on: http://review.gluster.org/9263 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* telldir()/seekdir() portability fixesEmmanuel Dreyfus2014-12-174-28/+87
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | POSIX says that an offset obtained from telldir() can only be used on the same DIR *. Linux is abls to reuse the offset accross closedir()/opendir() for a given directory, but this is not portable and such a behavior should be fixed. An incomplete fix for the posix xlator was merged in http://review.gluster.com/8926 This change set completes it. - Perform the same fix index xlator. - Use appropriate casts and variable types so that 32 bit signed offsets obtained by telldir() do not get clobbered when copied into 64 bit signed types. - modify glfs-heal.c and afr-self-heald.c so that they do not use anonymous fd, since this will cause closedir()/opendir() between each syncop_readdir(). On failure we fallback to anonymous fs only for Linux so that we can cope with updated client vs not updated brick. - Avoid sending an EINVAL when the client request for the EOF offset. Here we fix an error in previous fix for posix xlator: since we fill each directory entry with the offset of the next entry, we must consider as EOF the offset of the last entry, and not the value of telldir() after we read it. - Add checks in regression tests that we do not hit cases where offsets fed to seekdir() are wrong. Introduce log_newer() shell function to check for messages produced by the current script. This fix gather changes from http://review.gluster.org/9047 and http://review.gluster.org/8936 making them obsolete. BUG: 1129939 Change-Id: I59fb7f06a872c4f98987105792d648141c258c6a Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/9071 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Tested-by: Raghavendra Bhat <raghavendra@redhat.com>
* uss/gluster: In SVS, check for entry-point in dict only if inode_ctx isvmallika2014-12-161-17/+19
| | | | | | | | | | | | | | not available Change-Id: I990487003b712bf4aed8f54291417965f301655e BUG: 1172430 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/9265 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Sachin Pandit <spandit@redhat.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* features/snapview-client: handle fsync fopRaghavendra Bhat2014-12-121-0/+45
| | | | | | | | | | Change-Id: Idaf46bd7497266af837789b09a0c62698f56ee4e BUG: 1172262 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/9258 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Sachin Pandit <spandit@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* fuse: protect checking/modifying graph->used with a mutexNiels de Vos2014-12-121-3/+9
| | | | | | | | | | | | | | | | | | While reviewing http://review.gluster.org/#/c/9060/ we noticed that fuse checks/modifies graph->used in a unlocked way. There is a potential race in case when fuse_graph_setup() gets called from two different threads (currently unlikely, but poller/notify() might change in future?). In libgfapi we came to the conclusion that each glfs structure can be used to protect the checking/updating of graph->used. In fuse this would be the fuse_private_t with sync_mutex. BUG: 1170643 Change-Id: If5ab5468d22fdb92cfb24a469f538f63f12baf78 Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/9237 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* build: FreeBSD 11-Current causes libtool to fail with '-shared'Harshavardhana2014-12-123-4/+4
| | | | | | | | | | | | Thanks for Markiyan Kushnir <markiyan.kushnir@gmail.com> for reporting this Change-Id: I7f637295c7c2d54c33a4c16e29daf0b518874911 BUG: 1111774 Signed-off-by: Harshavardhana <harsha@harshavardhana.net> Reviewed-on: http://review.gluster.org/9251 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com>
* storage/posix: Set errno for xattrop failuresPranith Kumar K2014-12-101-0/+3
| | | | | | | | | | Change-Id: I4d44068c8da5257227d62906ec18ae16f6ed6c02 BUG: 1172477 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/9261 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Tested-by: Raghavendra Bhat <raghavendra@redhat.com>
* nfs: prevent potential NULL dereference in mnt3_readlink_cbk()Niels de Vos2014-12-101-2/+6
| | | | | | | | | | | | | | In case 'mres' is NULL, we jump to 'mnterr' where 'mres->req' will be used. This will cause a segmentation fault, which we really do not want. Change-Id: I01d6f3aa0343def54dcd4a5610001ff7d62c8834 CID: 1256179 BUG: 789278 Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/9257 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: jiffin tony Thottan <jthottan@redhat.com> Reviewed-by: Humble Devassy Chirammal <humble.devassy@gmail.com>
* features/snapview-client: handle readdir requests differently for sambaRaghavendra Bhat2014-12-095-9/+578
| | | | | | | | | | | * For samba export, the entry point is also added to the readdir response. Change-Id: I825c017e0f16db1f1890bb56e086f36e6558a1c2 BUG: 1168875 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/9218 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: Coverity fix for string_overflow overrunGauravKumarGarg2014-12-082-3/+3
| | | | | | | | | | | | | | | | | | | | In function glusterd_dump_peer() it is copying "input_key" into "key" buffer without checking the length which might cause string_overflow overrun. Similar problem with other coverity issue. With this fix it will copy "input_key" into "key" buffer by maximum length of buffer. Coverity CID: 1256171 Coverity CID: 1256172 Coverity CID: 1256174 Change-Id: I4e092309d9503bd79ff82cf83ed5e8d758743453 BUG: 1093692 Signed-off-by: Gaurav Kumar Garg ggarg <ggarg@redhat.com> Reviewed-on: http://review.gluster.org/9208 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Kaushal M <kaushal@redhat.com>
* ec: Fix return errors when not enough bricksXavier Hernandez2014-12-058-14/+81
| | | | | | | | | | | | | | | | | | | | | Changes introduced by this patch: * Fix an incorrect error propagation when the state of the life cycle of a fop returns an error. * Fix incorrect unlocking of failed locks. * Return ENOTCONN if there aren't enough bricks online. * In readdir(p) check that the fd has been successfully open by a previous opendir. Change-Id: Ib44f25a1297849ebcbab839332f3b6359f275ebe BUG: 1162805 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-on: http://review.gluster.org/9098 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* Fixed small typo in marker.cJan-Hendrik Zab2014-12-051-1/+1
| | | | | | | | | | Change-Id: Iec89e7141f6ff3216355d8a6b971617e16df679b BUG: 1075417 Signed-off-by: Jan-Hendrik Zab <zab@l3s.de> Reviewed-on: http://review.gluster.org/9238 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* uss/gluster: Send success on parent lookup of entry-pointvmallika2014-12-052-14/+51
| | | | | | | | | | | | | | | | | | When a lookup sent to snapview-server for entry-point directory protocol server first tries to resolve gfid of a parent directory. looking up the parent gfid from a latest snapshot can fail if the volume is a restored volume. As this gfid is already looked-up by snapview-client, we can return success for the parent gfid. Change-Id: Ic9b20561ef79b93032f07c3a81eae54a94e1747b BUG: 1162498 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/9229 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Eliminate locking in sh domain in metadata self-healKrutika Dhananjay2014-12-051-35/+2
| | | | | | | | | | Change-Id: I9ef25a17c9a43ba06fac2ad3f7c18cb47de91537 BUG: 1170913 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/9240 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* ec: Fix mutex related coverity scan issuesXavier Hernandez2014-12-041-2/+14
| | | | | | | | | | | | | | | | | | This patch solves 3 issues detected by coverity scan: CID1241484 Data race condition CID1241486 Data race condition CID1256173 Thread deadlock With this patch, inode lock is never acquired inside a region locked with fop->lock. Change-Id: I35c4633efd1b68b9f72b42661fa7c728b1f52c6a BUG: 1170254 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-on: http://review.gluster.org/9230 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* ec: Fix incorrect value of EC_MAX_NODESXavier Hernandez2014-12-042-1/+4
| | | | | | | | | | | | | EC_MAX_NODES was incorrectly calculated. Now the value if computed as the minimum between the theoretical maximum and the limit imposed by the Galois Field. Change-Id: I75a8345147f344f051923d66be2c10d405370c7b BUG: 1167419 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-on: http://review.gluster.org/9193 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* ec: Fix self-healing issues.Xavier Hernandez2014-12-0410-313/+548
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Three problems have been detected: 1. Self healing is executed in background, allowing the fop that detected the problem to continue without blocks nor delays. While this is quite interesting to avoid unnecessary delays, it can cause spurious failures of self-heal because it may try to recover a file inside a directory that a previous self-heal has not recovered yet, causing the file self-heal to fail. 2. When a partial self-heal is being executed on a directory, if a full self-heal is attempted, it won't be executed because another self-heal is already in process, so the directory won't be fully repaired. 3. Information contained in loc's of some fop's is not enough to do a complete self-heal. To solve these problems, I've made some changes: * Improved ec_loc_from_loc() to add all available information to a loc. * Before healing an entry, it's parent is checked and partially healed if necessary to avoid failures. * All heal requests received for the same inode while another self-heal is being processed are queued. When the first heal completes, all pending requests are answered using the results of the first heal (without full execution), unless the first heal was a partial heal. In this case all partial heals are answered, and the first full heal is processed normally. * An special virtual xattr (not physically stored on bricks) named 'trusted.ec.heal' has been created to allow synchronous self-heal of files. Now, the recommended way to heal an entire volume is this: find <mount> -d -exec getfattr -h -n trusted.ec.heal {} \; Some minor changes: * ec_loc_prepare() has been renamed to ec_loc_update(). * All loc management functions return 0 on success and -1 on error. * Do not delay fop unlocks if heal is needed. * Added basic ec xattrs initially on create, mkdir and mknod fops. * Some coding style changes Change-Id: I2a5fd9c57349a153710880d6ac4b1fa0c1475985 BUG: 1161588 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-on: http://review.gluster.org/9072 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* protocol/server: No root-squash checks for self-heal pidPranith Kumar K2014-12-041-0/+1
| | | | | | | | | | | | | | | | | | Problem: Self-heal pid used to be -1 which was colliding with gsyncd. Gsyncd was not checked for root-squash authentication. Recently self-heal pid changed to -6, but root-squash authentication is not disabled for this. Fix: disable root-squash authentication for self-heal Change-Id: I93233d4ae681cb936d166b22992eb47c658ea977 BUG: 1170407 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/9231 Reviewed-by: Vijay Bellur <vbellur@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* features/index: closedir() if fd_ctx_set() fails in __index_fd_ctx_get.Vijay Bellur2014-12-041-0/+1
| | | | | | | | | | | Change-Id: I9ab599cf4a90c2e285d63ec0f56af210709285b6 BUG: 1166284 Signed-off-by: Vijay Bellur <vbellur@redhat.com> Reviewed-on: http://review.gluster.org/9168 Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* Glusterfs/posix: Stack corruption in posix_handle_pumpNithya Balachandran2014-12-031-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | posix_handle_pump can corrupt the stack if the buffer passed to it is too small to hold the final path. Fix : Check if the buffer is sufficiently large to hold the new path component before modifying it. This will prevent the buffer overrun but the path returned will most likely have too many symbolic links causing subsequent file ops to fail with ELOOP. The callers of this function do not currently check the return value. The code needs to be modified to have all callers check the return value and take appropriate action in case of an error. Change-Id: I6d9589195a4b0d971a107514ded6e97381e5982e BUG: 1113960 Signed-off-by: Nithya Balachandran <nbalacha@redhat.com> Reviewed-on: http://review.gluster.org/8189 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* ec: Change licenseXavier Hernandez2014-12-0323-369/+138
| | | | | | | | | | Change-Id: Iae90ade2421898417b53dec0417a610cf306c44b BUG: 1168167 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-on: http://review.gluster.org/9201 Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* uss: CIFS gives error while accessing .snaps directoryRajesh Joseph2014-12-034-24/+70
| | | | | | | | | | | | | | | | | | | | | CIFS sends getxattr call while accessing .snaps directory. snapview server should return ENODATA if getxattr is called on .snaps folder. Also flush on .snaps folder should be handled. Windows client sends and getxattr call on the parent folder of .snaps with a special key to get the real filename of a file. This is used by samba to do case insensitive check of files. Fixed few FreeBSD compilation error. bug: 1168875 Change-Id: I74d5cb4419568c8ed8709ba6d1ddff0e41392204 Signed-off-by: Rajesh Joseph <rjoseph@redhat.com> Reviewed-on: http://review.gluster.org/9211 Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: use synclock_t for synchronizing concurrent '\op_sm\' invocationsKrishnan Parthasarathi2014-12-011-5/+7
| | | | | | | | | | | | | | | | | | | | | | | | | In glusterd_op_sm(), we lock and unlock the gd_op_sm_lock mutex. Unfortunately, locking and unlocking can happen in different threads (task swap will occur in handler call with use of synctasks). This case is explictely covered by POSIX: the behavior is undefined. http://pubs.opengroup.org/onlinepubs/9699919799/functions/pthread_mutex_lock.html When unlocking from a thread that is not owner, Linux seems to be fine (though you never know with unspecified operation), while NetBSD returns EPERM, causing a spurious error in tests/basic/pump. To fix this, we use synclock_t which was precisely meant for this. synclock is a pthread_mutex_t like synchronization object which uses the synctask handle for owner and is immune to the task being run on multiple threads during its lifetime. Change-Id: Idca15190d42f32a843088cc8236138f676377586 BUG: 1129939 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/9212 Reviewed-by: Kaushal M <kaushal@redhat.com> Tested-by: Kaushal M <kaushal@redhat.com>
* mgmt/glusterd: Out of bounds access to fs_info structPetr Medonos2014-12-011-1/+1
| | | | | | | | | | | Change-Id: Ifa0d4ac17f9da94660a7b7f567a0f07b5cec7aec BUG: 1164775 Signed-off-by: Petr Medonos <petr.medonos@etnetera.cz> Reviewed-on: http://review.gluster.org/9138 Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Kaushal M <kaushal@redhat.com>
* USS : fill proper uid and gid during a access call from nfsSachin Pandit2014-12-011-1/+6
| | | | | | | | | | | | | | | | | | | | Problem : when an user tries to access a file/folder for which he does not have a proper permission required then fuse gives out a proper error "Permission denied", but nfs does not give out that error, rather he can access the file/folder. The reason being uid and gid of call frame stack takes a default value of uid and gid which point to root permission. Solution : Set a proper uid and gid during a access call from nfs Change-Id: Ib060706fde66ec7e60f242fab1f3e59122ed2245 BUG: 1167580 Signed-off-by: Sachin Pandit <spandit@redhat.com> Reviewed-on: http://review.gluster.org/9194 Reviewed-by: Vijaikumar Mallikarjuna <vmallika@redhat.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd/uss: Create rebalance volfile.Avra Sengupta2014-11-305-21/+117
| | | | | | | | | | | | | | | | | | | | Create a new rebalance volfile, which will not contain snap-view client translators, irrespective of the status of USS. This volfile, will be created and regenerated everytime the fuse-volfile is generated, and will be consumed by the rebalance process. Change-Id: I514a8e88d06c0b8fb6949c3a3e6dc4dbe55e38af BUG: 1164711 Signed-off-by: Avra Sengupta <asengupt@redhat.com> Reviewed-on: http://review.gluster.org/9190 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* glusterd/uss: if snapd is not running, return success from ↵Atin Mukherjee2014-11-301-0/+3
| | | | | | | | | | | | | | | | | | glusterd_handle_snapd_option glusterd_handle_snapd_option was returning failure if snapd is not running because of which gluster commands were failing. Change-Id: I22286f4ecf28b57dfb6fb8ceb52ca8bdc66aec5d BUG: 1168803 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/9206 Reviewed-by: Kaushal M <kaushal@redhat.com> Reviewed-by: Avra Sengupta <asengupt@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijaikumar Mallikarjuna <vmallika@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* glusterd: logging improvement in txn_opinfo getter/setter functionAtin Mukherjee2014-11-301-11/+12
| | | | | | | | | | | | | | | | | | There is a code path (__glusterd_handle_stage_op) where glusterd_get_txn_opinfo may fail to get a valid transaction id if there is no volume name provided in the command, however if this function fails to get a txn id in op state machine then its a serious issue and op-sm is impacted. From debugability aspect gf_log () can never give the consumer of this function, so logging these failures with gf_log_calling_fn is must here. Change-Id: I4937a9fb20cc6a747fd30dcd9fd4936398d0602a BUG: 1168809 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/9207 Reviewed-by: Avra Sengupta <asengupt@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* posix: remove duplicate dirfd calls in posix_opendirAtin Mukherjee2014-11-301-1/+1
| | | | | | | | | BUG: 1168910 Change-Id: I285d352d20374bb3edee2db42d062d4724198425 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/9186 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* Prevent metadata corruption in the race conditions betweenEdward Shishkin2014-11-301-9/+23
| | | | | | | | | | | | | | | | | | | | FOP->open() and FOP->link(). Problem: crypt_open() modifies @local->format, which is used by crypt_link() to store the updated metadata string on disk. This results in metadata corruption. Fixup: Don't modify @local->format by FOP->open(). Instead modify a local copy, allocated in the low-level meta-data handler open_format_v1(). Change-Id: I046bb39ddefc33afe59c8d3b1a2fa798298f8499 BUG: 1157839 Signed-off-by: Edward Shishkin <edward@redhat.com> Reviewed-on: http://review.gluster.org/8982 Reviewed-by: Emmanuel Dreyfus <manu@netbsd.org> Tested-by: Emmanuel Dreyfus <manu@netbsd.org> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* posix: Fix buffer overrun in _handle_list_xattr()Emmanuel Dreyfus2014-11-281-4/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | In _handle_list_xattr() we test remaining_size > 0 to check that we do not overrun the buffer, but since that variable was unsigned (size_t), the condition would let us go beyond end of buffer if remaining_size became negative. This could happen if attribute list grew between the first sys_llistxattr() call that gets the size and the second sys_llistxattr() call that get the data. We fix the problem by making remaining_size signed (ssize_t). This also matches sys_llistxattr() return type. While there, we use the size returned by the second sys_llistxattr() call to parse the buffser, as it may also be smaller than the size obtained from first call, if attribute list shrank. This fixes a spurious crash in tests/basic/afr/resolve.t BUG: 1129939 Change-Id: Ifc5884dd0f39a50bf88aa51fefca8e2fa22ea913 Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/9204 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* gluster/uss: Handle notify in snapview-clientvmallika2014-11-281-0/+25
| | | | | | | | | | | | | | | As there are two subvolumes in snapview-client, there is a possibility that the regular subvolume is still down and snapd subvolume come up first. So if we don't handle this situation CHILD_UP event will be propagated upwards to fuse when regular subvolume is still down. This can cause data unavailable for the application Change-Id: I9e5166ed22c2cf637c15db0457c2b57ca044078e BUG: 1168643 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/9205 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* core: fix remaining *printf formation warnings on 32-bitKaleb S. KEITHLEY2014-11-265-17/+20
| | | | | | | | | | | | | | | | | | This fixes a few lingering size_t problems. Of particular note are some uses of off_t for size params in function calls. There is no correct, _portable_ way to correctly print an off_t. The best you can do is use a scratch int64_t/PRId64 or uint64_t/PRIu64. Change-Id: I86f3cf4678c7dbe5cad156ae8d540a66545f000d BUG: 1110916 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: http://review.gluster.org/8105 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Harshavardhana <harsha@harshavardhana.net> Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* gluster/uss: Handle ESTALE in snapview client when NFS server restartedvmallika2014-11-252-13/+30
| | | | | | | | | | | | | | | When NFS server is restarted inode-context is lost. Nameless lookup will be sent to regular volume. If the gfid is from virtual graph, lookup will fail with ESTALE. We need to send a lookup to snapview server Change-Id: I22920614f0d14cb90b53653fce95b6b70023eba6 BUG: 1166197 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/9153 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Sachin Pandit <spandit@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* core: fix Ubuntu code audit (cppcheck) resultsKaleb S. KEITHLEY2014-11-2510-32/+57
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | See also http://review.gluster.org/#/c/7693/, BZ 1091677 AFAICT these are false positives: [geo-replication/src/gsyncd.c:100]: (error) Memory leak: str [geo-replication/src/gsyncd.c:403]: (error) Memory leak: argv [xlators/nfs/server/src/nlm4.c:1201]: (error) Possible null pointer dereference: fde [xlators/cluster/afr/src/afr-self-heal-common.c:138]: (error) Possible null pointer dereference: __ptr [xlators/cluster/afr/src/afr-self-heal-common.c:140]: (error) Possible null pointer dereference: __ptr [xlators/cluster/afr/src/afr-self-heal-common.c:331]: (error) Possible null pointer dereference: __ptr Test program: [extras/test/test-ffop.c:27]: (error) Buffer overrun possible for long command line arguments. [tests/basic/fops-sanity.c:55]: (error) Buffer overrun possible for long command line arguments. the remainder are fixed with this change-set: [cli/src/cli-rpc-ops.c:8883]: (error) Possible null pointer dereference: local [cli/src/cli-rpc-ops.c:8886]: (error) Possible null pointer dereference: local [contrib/uuid/gen_uuid.c:369]: (warning) %ld in format string (no. 2) requires 'long *' but the argument type is 'unsigned long *'. [contrib/uuid/gen_uuid.c:369]: (warning) %ld in format string (no. 3) requires 'long *' but the argument type is 'unsigned long *'. [xlators/cluster/dht/src/dht-rebalance.c:1734]: (error) Possible null pointer dereference: ctx [xlators/cluster/stripe/src/stripe.c:4940]: (error) Possible null pointer dereference: local [xlators/mgmt/glusterd/src/glusterd-geo-rep.c:1718]: (error) Possible null pointer dereference: command [xlators/mgmt/glusterd/src/glusterd-replace-brick.c:942]: (error) Resource leak: file [xlators/mgmt/glusterd/src/glusterd-replace-brick.c:1026]: (error) Resource leak: file [xlators/mgmt/glusterd/src/glusterd-sm.c:249]: (error) Possible null pointer dereference: new_ev_ctx [xlators/mgmt/glusterd/src/glusterd-snapshot.c:6917]: (error) Possible null pointer dereference: volinfo [xlators/mgmt/glusterd/src/glusterd-utils.c:4517]: (error) Possible null pointer dereference: this [xlators/mgmt/glusterd/src/glusterd-utils.c:6662]: (error) Possible null pointer dereference: this [xlators/mgmt/glusterd/src/glusterd-utils.c:7708]: (error) Possible null pointer dereference: this [xlators/mount/fuse/src/fuse-bridge.c:4687]: (error) Uninitialized variable: finh [xlators/mount/fuse/src/fuse-bridge.c:3080]: (error) Possible null pointer dereference: state [xlators/nfs/server/src/nfs-common.c:89]: (error) Dangerous usage of 'volname' (strncpy doesn't always null-terminate it). [xlators/performance/quick-read/src/quick-read.c:586]: (error) Possible null pointer dereference: iobuf Rerunning cppcheck after fixing the above: As before, test program: [extras/test/test-ffop.c:27]: (error) Buffer overrun possible for long command line arguments. [tests/basic/fops-sanity.c:55]: (error) Buffer overrun possible for long command line arguments. As before, false positive: [geo-replication/src/gsyncd.c:100]: (error) Memory leak: str [geo-replication/src/gsyncd.c:403]: (error) Memory leak: argv [xlators/nfs/server/src/nlm4.c:1201]: (error) Possible null pointer dereference: fde [xlators/cluster/afr/src/afr-self-heal-common.c:138]: (error) Possible null pointer dereference: __ptr [xlators/cluster/afr/src/afr-self-heal-common.c:140]: (error) Possible null pointer dereference: __ptr [xlators/cluster/afr/src/afr-self-heal-common.c:331]: (error) Possible null pointer dereference: __ptr False positive after fix: [xlators/performance/quick-read/src/quick-read.c:584]: (error) Possible null pointer dereference: iobuf Change-Id: I20e0e3ac1d600b2f2120b8d8536cd6d9e17023e8 BUG: 1109180 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: http://review.gluster.org/8064 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cli: .cmd_log_history should not be hiddenAtin Mukherjee2014-11-241-1/+1
| | | | | | | | | Change-Id: I4513a2c260530855e09be64083e9344108c7a6c0 BUG: 1165996 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/9150 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaushal M <kaushal@redhat.com>
* mount:Handle -o transport option in mount.glusterfsMohammed Rafi KC2014-11-211-13/+5
| | | | | | | | | | | | | | | | | | | In current scenario ,when tcp transport type(default) specified for mounting,glusterfs mount script won't append '.tcp' to volume name.But to accommodate the change in http://review.gluster.org/#/c/9146/, we need to append ".tcp" with volfile-id if '-o transport=tcp' is given. Change-Id: I506edeb339de062087f820f10e17c7d67b8b63fe BUG: 1164079 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: http://review.gluster.org/9147 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: jiffin tony Thottan <jthottan@redhat.com> Reviewed-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com>
* glusterd: Add hostname/ip-address along with host's UUID in glusterd log messageGauravKumarGarg2014-11-201-2/+3
| | | | | | | | | | | | | | | | | Previously when host disconnected from cluster then glusterd logs identifies host using host's UUID. Now with this fix, UUID along with host's ip will be present in glusterd log message when one of the peer disconnected from cluster. So it will enhancement better readability of user from log file. Change-Id: I3b7eaf1b1a8963ef2096e67a78cf69f67d5d5166 BUG: 1101382 Signed-off-by: Gaurav Kumar Garg <ggarg@redhat.com> Reviewed-on: http://review.gluster.org/9136 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-by: Kaushal M <kaushal@redhat.com>
* core: use gf_time_fmt() instead of localtime()+strftime()Kaleb S. KEITHLEY2014-11-205-76/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | gf_time_fmt() has existed since 3.3; it provides consistent timestamps (i.e. UTC times) throughout the implementation. (BTW, the other name for UTC is GMT.) N.B. many (all?) commercial storage solutions use UTC time for logging. This makes for easier debugging across geographically distributed systems. Also adding a "%s" fmt for portably printing time as simple numeric value on systems regardless of whether 32-bit or 64-bit time_t. Plus a minor tweak to return a ptr to the dest-string to allow gf_time_fmt() to be passed as a param in a *printf(). Someday we should pick the "one true" timestamp format and revise all calls to gf_time_fmt() to use it instead of the five or six different formats. Change-Id: I78202ae14b7246fa424efeea56bf2463e14abfb0 BUG: 1109917 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: http://review.gluster.org/8085 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* rdma: Client volfile name change for supporting rdmaAnoop C S2014-11-192-4/+22
| | | | | | | | | | | | | | | | | | | For rdma only volumes, daemons like snapd, glustershd etc make use of tcp transport for their operations. This patch will introduce the support of rdma by default for those daemons in rdma only volumes. In order to accomodate this change we rename the tcp client volfile labels from <volname>-fuse.vol to <volname>.tcp-fuse.vol Change-Id: Id9727b97d00e62a4a1556b9c0c56653d45c8fe1d BUG: 1164079 Signed-off-by: Anoop C S <achiraya@redhat.com> Reviewed-on: http://review.gluster.org/9146 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaushal M <kaushal@redhat.com>
* md-cache: check mtime_nsec and ctime_nsec fields for changes tooNiels de Vos2014-11-191-2/+6
| | | | | | | | | | | | | | | | | When md-cache checks for an updated mtime/ctime in order to see if it should request a cache invalidation, it only looks at the whole-second "mtime" and "ctime" fields. If a file was modified a fraction of a second after the time of the cached data, md-cache won't notice the change. BUG: 1164506 Change-Id: Ieb7baa46f84489d2d7ea8d60a7fb2595564eb955 Original-author: Philip Spencer <pspencer@fields.utoronto.ca> Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/9131 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* md-cache: cache the iatt when readv() was successfulNiels de Vos2014-11-191-1/+1
| | | | | | | | | | | | | | | | | | | | | | When md-cache is handling a readv callback, its logic is to skip checking the file's iatt (to see if it has changed since the time of the last cache) if "op_ret != 0". This is a mistake, and the intent was to skip only on error; for other fops, op_ret is 0 on success and nonzero on error, and md-cache has the same code for them. For readv, though, op_ret is the number of bytes read, so it will normally be positive except on EOF. BUG: 1164503 Change-Id: Ifc0f2084b41b6f8d075a9260dc7cf69e97535eed Original-author: Philip Spencer <pspencer@fields.utoronto.ca> Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/9129 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* rdma :mount fails for nfs protocol in rdma volumesJiffin Tony Thottan2014-11-197-25/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we mount rdma only volume or tcp,rdma volume using newly peer probed IP's(nfs-server on new nodes) through nfs protocol, mount fails for rdma only volume and mount happens with help of tcp protocol in the case of tcp,rdma volumes. That is for newly added servers will always get transport type as "socket". This is due to nfs_transport_type is exported correctly and imported wrongly. This can be verified by the following , * Create a rdma only volume or tcp,rdma volume * Add a new server into the trusted pool. * Checkout the client transport type specified nfs-server volgraph.It will be always tcp(socket type) instead of rdma. * And also for rdma only volume in the nfs log, we can see 'connection refused' message for every reconnect between nfs server and glusterfsd. BUG: 1157381 Change-Id: I6bd4979e31adfc72af92c1da06a332557b6289e2 Author: Jiffin Tony Thottan <jthottan@redhat.com> Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com> Reviewed-on: http://review.gluster.org/8975 Reviewed-by: Meghana M <mmadhusu@redhat.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Tested-by: Niels de Vos <ndevos@redhat.com>
* rdma: Wrong volfile fetch on fuse mounting tcp,rdma volume via rdmaAnoop C S2014-11-184-82/+128
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As of now for both tcp only volumes and rdma only volumes, volfile names are in the format <volname>-fuse.vol. This patch will change the client volfile namings as shown below. * TCP mounts always use <volname>-fuse.vol * RDMA mounts always use <volname>.rdma-fuse.vol Following the above naming convention, for tcp,rdma volumes both volfiles will be present under /var/lib/glusterd/vols/<volname>/ such that rdma only volume can be mounted as mount -t glusterfs -o transport=rdma <server/ip>:/<volname> <mount-point> OR mount -t glusterfs <server/ip>:/<volname>.rdma <mount-point> The above command format can also be used to fuse mount a tcp,rdma volume via rdma transport. When we try to fuse mount a tcp,rdma volume with transport-type as rdma it silently mounts via tcp. This change will also make sure that it fetches the correct volfile based on the transport-type specified from client side. BUG: 1131502 Change-Id: I34da4b01ac813b69494a43188f51145457412923 Signed-off-by: Anoop C S <achiraya@redhat.com> Reviewed-on: http://review.gluster.org/8498 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com>
* rdma: client connection establishment takes more timeMohammed Rafi KC2014-11-185-36/+42
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For rdma type only volume client connection establishment with server takes more than three seconds. Because for tcp,rdma type volume, will have 2 ports one for tcp and one for rdma, tcp port is stored with brickname and rdma port is stored as "brickname.rdma" during pamap_sighin. During the handshake when trying to get the brick port for rdma clients, since we are not aware of server transport type, we will append '.rdma' with brick name. So for tcp,rdma volume there will be an entry with '.rdma', but it will fail for rdma type only volume. So we will try again, this time without appending '.rdma' using a flag variable need_different_port, and it will succeed, but the reconnection happens only after 3 seconds. In this patch for rdma only type volume we will append '.rdma' during the pmap_signin. So during the handshake we will get the correct port for first try itself. Since we don't need to retry , we can remove the need_different_port flag variable. Change-Id: Ie8e3a7f532d4104829dbe995e99b35e95571466c BUG: 1153569 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: http://review.gluster.org/8934 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com>
* rdma:rdma fuse mount hangs for tcp,rdma volumes if brick is down.Mohammed Rafi KC2014-11-172-7/+14
| | | | | | | | | | | | | | | | | | | | | | | | When we try to mount a tcp,rdma volume as rdma transport using FUSE protocol, then mount will hang if the brick is down. When we kill a process, signal will be received in glusterfsd process and it will call pmap_signout with port listening on tcp only. In case of the tcp,rdma there will be two ports, and port which is listening for rdma will not called for sign out. So the mount process will try to connect to a port which is not open and it will keep trying to connect. This patch will call pmap_signout for rdma port also, So when mount tries to get the brick port,it will fail. Change-Id: I23676f65f96eb90b69b76478f7a21412a6aba70f BUG: 1143886 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: http://review.gluster.org/8762 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com>
* changelog: correct mmap() error checkingNiels de Vos2014-11-171-8/+6
| | | | | | | | | | | | | | | Upon failure, mmap() returns MAP_FAILED, which is not equal to NULL. This small correction makes sure that the potential error gets caught and handled. BUG: 1138621 Change-Id: I0a88b5187afa7187dcaa8f7d2cb0f9bb775c929d Reported-by: Santosh Pradhan <santosh.pradhan@gmail.com> Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/9130 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Venky Shankar <vshankar@redhat.com>
* USS : Kill snapd during glusterd restart if USS is disabledSachin Pandit2014-11-171-5/+25
| | | | | | | | | | | | | | | | | | | | | | | Problem : When glusterd is down on one of the nodes and during that time if USS is disabled then snapd will still be running in the node where glusterd was down. Solution : during restart of glusterd check if USS is disabled, if so then issue a kill for snapd. NOTE : The test case which I wrote in my previous patchset is facing some spurious failures, hence I thought of removing that test case. I'll add the test case once the issue is resolved. Change-Id: I2870ebb4b257d863cdfc319e8485b19e932576e9 BUG: 1161015 Signed-off-by: Sachin Pandit <spandit@redhat.com> Reviewed-on: http://review.gluster.org/9062 Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Reviewed-by: Avra Sengupta <asengupt@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* gluster/uss: Don't send failure when accessing snap with namevmallika2014-11-171-14/+2
| | | | | | | | | | | | | | | | | | same as entry-point name In a scenario, if the snap name is same as the snap-directory than cd to snaps/snaps fails. Send a lookup to snap-view server instead of failing Change-Id: Ie7b811815ff30961500592bbc8cdb514a9d76ef5 BUG: 1164613 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/9135 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Rajesh Joseph <rjoseph@redhat.com> Reviewed-by: Avra Sengupta <asengupt@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
generated by cgit (git 2.34.1) at 2024-05-22 05:37:33 +0000