summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
...
* features/shard: Take hole size into account while computing ia_sizeKrutika Dhananjay2015-05-032-2/+10
| | | | | | | | | | | Backport of: http://review.gluster.org/10446 Change-Id: Ic05e07801605c0d610545368a513b56d8df21bf4 BUG: 1214247 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/10493 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* features/shard: Add "is-directory" checks in stat/fstatKrutika Dhananjay2015-05-031-0/+12
| | | | | | | | | | | | | | | | | Backport of: http://review.gluster.org/10427 During mount, NFS directly calls stat on the root of the volume without sending a lookup on it. This was causing inode_ctx_get_block_size() to fail on /. A check is now added in [f]stat which would ensure no action is taken by shard xlator when the operation is on a directory. Change-Id: I8645b7fe58b2d44b5f527d50c1c7102de44acc00 BUG: 1214247 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/10509 Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* geo-rep: Re Enable ignore_deletes OptionAravinda VK2015-05-031-2/+6
| | | | | | | | | | | | | | | | | | | | | | | If this option is set, Deletes will not be propogated to Slave. This option is applicable for UNLINK and RMDIR. gluster volume geo-replication <MASTER> <SLAVEHOST>::<SLAVEVOL> \ config ignore_deletes true Default value is false. PS: Use this option with caution, If you create the file in master with same path then it fails to sync to slave. Old file in Slave will have different GFID compared to New. BUG: 1217929 Change-Id: I1f7816d1ea36460a654873739d3fb1b6c13e0f8d Signed-off-by: Aravinda VK <avishwan@redhat.com> Reviewed-on: http://review.gluster.org/9583 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kotresh HR <khiremat@redhat.com> Reviewed-on: http://review.gluster.org/10498 Tested-by: NetBSD Build System
* nfs: Enable throttling for rpcsvcVijay Bellur2015-05-031-0/+7
| | | | | | | | | | | | | | | | Busy nfs servers need throttling to prevent duplicate requests from casuing an increase in memory consumption. Change-Id: I855bb1e841ba06b523e55f86f7788a0f7a49d444 BUG: 1216310 Signed-off-by: Vijay Bellur <vbellur@redhat.com> Reviewed-on: http://review.gluster.org/10269 Reviewed-by: Niels de Vos <ndevos@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-on: http://review.gluster.org/10444 Tested-by: NetBSD Build System
* NFS-Ganesha: Handling CLI commands when NFS-Ganesha keys are setMeghana Madhusudhan2015-05-037-74/+238
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When ganesha.enable is set to on and features.ganesha is enabled, there are a few behaviour changes that should be seen in other volume operations. 1. ganesha.enable can be set to 'on' only when features.ganesha is set to 'enable' 2.When gluster vol is started, and if ganesha.enable key was set to 'on', it should automatically export the volume via NFS-Ganesha. 3.When ganesha.enable is set to 'on', and a volume is stopped, that volume should be unexported via NFS-Ganesha. 4. gluster vol reset <volname> If ganesha.enable was set to on, then unexport the volume via NFS-Ganesha. 5. gluster vol reset all If features.ganesha is set to enable, as part of reset all, set it to disable. This translates to teardown cluster. All the above problems are fixed by checking the global key and value, depending on the value, specific functions are called. And also, functions related to global commands are moved to cli-cmd-global.c Commit phase of features.ganesha enable/disable runs the ganesha-ha.sh setup/teardown respectively. Before the script begins, it is important that the NFS-Ganesha service starts on all the HA nodes. Having the start service commands in the commit phase could lead to problems. Moving the pre-requisite service start commands to the 'stage' phase. Change-Id: I9084d004c0aaf960f59ed8aa2d7f1c9309924658 BUG: 1217793 Signed-off-by: Meghana Madhusudhan <mmadhusu@redhat.com> Reviewed-on: http://review.gluster.org/10489 Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com>
* storage/posix: prevent NULL dereferencePranith Kumar K2015-05-021-3/+3
| | | | | | | | | | | | | | Backport of http://review.gluster.org/10383 filler->fd is never set but used. BUG: 1216303 Change-Id: I02dc346c526be5af4ea55ae13e8314316f127455 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/10437 Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* quota: Validate NULL inode from the entries received in readdirp_cbkvmallika2015-05-022-6/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a backport of http://review.gluster.org/#/c/10416/ > In quota readdirp_cbk, inode ctx filled for the all entries > received. > In marker readdirp_cbk, files/directories are inspected for > dirty > > There is no guarantee that entry->inode is populated. > If entry->inode is NULL, this needs to be treated as readdir > > Change-Id: Id2d17bb89e4770845ce1f13d73abc2b3c5826c06 > BUG: 1215550 > Signed-off-by: vmallika <vmallika@redhat.com> > Reviewed-on: http://review.gluster.org/10416 > Tested-by: NetBSD Build System > Tested-by: Gluster Build System <jenkins@build.gluster.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Tested-by: Raghavendra G <rgowdapp@redhat.com> Change-Id: I6e01b16bae1555efb6c20e49c9c473d81c6d46a0 BUG: 1217406 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/10468 Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/ec: Handle unhandled statesPranith Kumar K2015-05-021-0/+3
| | | | | | | | | | | | | | Backport of http://review.gluster.org/10390 This was leading to hangs when get_size_and_version fails BUG: 1216303 Change-Id: Iae455ee957b9377e1b0b711b0ef567d50d32c7cb Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/10436 Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr,dht: Fix memleak after syncop_readlinkPranith Kumar K2015-05-012-0/+2
| | | | | | | | | | | | Backport of http://review.gluster.org/10305 BUG: 1216302 Change-Id: Icb0f2d6bbff806e1c5827fabcbf46b9b7983491f Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/10441 Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* dht: tackle thread race in dht_getxattr_cbkSusant Palai2015-05-011-17/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | problem: 1. When two threads execute in parallel in dht_getxattr_cbk it may so happen that, both may find local->xattr to be NULL. As a result dht_aggregate_xattr may not get executed. 2. In dht_getxattr_cbk, thread1 thread2 T1 this_call_cnt = 2 -1 T2 this_call_cnt = 1 - 1 T3 fills local_xattr T4 DHT_STACK_UNWIND -> local_wipe T5 tries to dereference local which is already freed, leading to crash. Solution: for problem1: Execute critical section inside frame lock to resolve race. for problem2: Calculate this_call_count just before out section. BUG: 1217386 Change-Id: I14fdb0cb1825896721670d71f48c93053448be7b Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/10389 Reviewed-by: Anuradha Talur <atalur@redhat.com> Reviewed-by: N Balachandran <nbalacha@redhat.com> Reviewed-by: Kotresh HR <khiremat@redhat.com> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com> Signed-off-by: Susant Palai <spalai@redhat.com> Reviewed-on: http://review.gluster.org/10467 Tested-by: Gluster Build System <jenkins@build.gluster.com>
* features/shard: Implement rename() fopKrutika Dhananjay2015-04-302-44/+261
| | | | | | | | | | | | | Backport of: http://review.gluster.org/10373 Change-Id: I15867667d50b2b4aad0ee3738a29f7a410d61ef4 BUG: 1214247 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/10455 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* libglusterfs: Implementation of sync lock as recursive lock to avoid crash.anand2015-04-305-33/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | Problem : In glusterd,we are using big lock which is implemented based on sync task frame work for thread synchronization and rcu lock for data consistency. sync task frame work swap the threads if there is no worker poll threads available,due to this rcu lock and rcu unlock was happening in different threads (urcu-bp will not allow this),resulting into glusterd crash. fix : To avoid releasing the sync lock(big lock) in between rcu critical section,implemented sync lock as recursive lock. More details: link : http://www.spinics.net/lists/gluster-devel/msg14632.html Change-Id: I2b56c1caf3f0470f219b1adcaf62cce29cdc6b88 BUG: 1216942 Signed-off-by: anand <anekkunt@redhat.com> Reviewed-on: http://review.gluster.org/10285 Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System Reviewed-by: Vijay Bellur <vbellur@redhat.com> (cherry picked from commit ada6b3a8800867934af57a57d5312f5a5d8374f0) Reviewed-on: http://review.gluster.org/10432 Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Vijay Bellur <vbellur@redhat.com>
* quick-read: Do a null check before unrefRavishankar N2015-04-281-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Backport of http://review.gluster.org/10206 Commit 4ea5b8d2046b9e0bc7f24cdf1b2e72ab8b462c9e seems to have removed the check as a part of static analyis fixes but I'm seeing errors in the client log. --------------------- touch /mnt/fuse_mnt/zero-byte-file echo 3 > /proc/sys/vm/drop_caches cat /mnt/fuse_mnt/zero-byte-file mount log: [2015-04-13 05:52:21.683256] E [iobuf.c:790:iobuf_unref] (--> /usr/local/lib/libglusterfs.so.0(_gf_log_callingfn+0x232)[0x7feda12c0e24] (--> /usr/local/lib/libglusterfs.so.0(iobuf_unref+0x56)[0x7feda1304c8e] (--> /usr/local/lib/glusterfs/3.7dev/xlator/performance/quick-read.so(qr_readv_cached+0x466)[0x7fed95b7e2fc] (--> /usr/local/lib/glusterfs/3.7dev/xlator/performance/quick-read.so(qr_readv+0x70)[0x7fed95b7e385] (--> /usr/local/lib/libglusterfs.so.0(default_readv_resume+0x270)[0x7feda12d4401] ))))) 0-iobuf: invalid argument: iobuf --------------------- Change-Id: Ia6b29165cdef8783a716fe1e72d3d13bf0220aae BUG: 1214168 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reviewed-on: http://review.gluster.org/10329 Tested-by: NetBSD Build System Reviewed-by: Raghavendra Talur <rtalur@redhat.com> Tested-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* tier: relax libgfdb required version numberEmmanuel Dreyfus2015-04-281-1/+1
| | | | | | | | | | | | | | | | | | | | | When calling dlopen() for libgfdb, do not specify the library version number "libgfdb.so.0.0.1", since libtool will not always create libraries or link with that name with the full 3-digit version. For instance on NetBSD only up to the 2-digit version is available and "libgfdb.so.0.0.1" does not exist. Instead, just specify "libgfdb.so" and rely on smymlinks installed by libtool to find the relevant library. Backport of: I074b1009d3622a122fdaeb4b99658bca3277e211 BUG: 1212676 Change-Id: I334cb6be8508051105c393ce4bb350f9df014df5 Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/10408 Reviewed-by: Dan Lambright <dlambrig@redhat.com> Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com>
* quota: display error when inode-quota cmds executed with cluster ver < 3.7vmallika2015-04-281-23/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a backport of http://review.gluster.org/#/c/10261/ > In a heterogeneous cluster with op_version less than 3.7, inode quotas will > be accounted on those bricks which has glusterfs version 3.7 and this is > not available in older version. > This will have incorrect values displayed when user queries inode count > from CLI. > > This patch will display error when inode-quota commands > are executed with cluster version less than 3.7 > > Change-Id: Ia0e6d5635d1d8e7b2e2cfc3daa7b7f9e314a263a > BUG: 1212253 > Signed-off-by: vmallika <vmallika@redhat.com> > Reviewed-on: http://review.gluster.org/10261 > Tested-by: Gluster Build System <jenkins@build.gluster.com> > Reviewed-by: Atin Mukherjee <amukherj@redhat.com> > Reviewed-by: Vijay Bellur <vbellur@redhat.com> Change-Id: I1ddd605e5b87a248aa85f0eab14c404895751083 BUG: 1215907 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/10415 Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: Remove direct references to peerinfo in frame cookiesKaushal M2015-04-283-67/+158
| | | | | | | | | | | | | | | | | | | | | | RCU protection requires that we don't have direct references to protected data structures outside read-critical sections This change was developed on the git branch at [1]. This commit is a combination of the following commits on the development branch. 82ebfdd Remove direct references to peerinfo in frame cookies dec4bec Remove incorrect and unneeded code from gd_syncop_mgmt_v3_unlock_cbk_fn 7aced7b Use stack allocated uuid for frame cookie. 38e4124 Address comments from 10192/2 [1]: https://github.com/kshlm/glusterfs/tree/urcu Change-Id: Ic50e5fca0be72af5090f4cf318efa55d29075de9 BUG: 1205186 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.org/10399 Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* glusterd: incorrect rsp.op_ret handling in volume get in cliAtin Mukherjee2015-04-281-1/+0
| | | | | | | | | | | | | | | | | | | Cherry picked from commit 540fc2829bd63a2fa070c68ed105eb23145df406: > Change-Id: Iabe99c06166578fc90121e7cfdca4a6a3f5328ae > BUG: 1211132 > Signed-off-by: Atin Mukherjee <amukherj@redhat.com> > Reviewed-on: http://review.gluster.org/10229 > Reviewed-by: Niels de Vos <ndevos@redhat.com> > Tested-by: Gluster Build System <jenkins@build.gluster.com> > Reviewed-by: Kaushal M <kaushal@redhat.com> Change-Id: Iabe99c06166578fc90121e7cfdca4a6a3f5328ae BUG: 1215547 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/10398 Reviewed-by: Niels de Vos <ndevos@redhat.com> Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com>
* glusterd: initialize snapd svc at volume restore pathAtin Mukherjee2015-04-283-11/+9
| | | | | | | | | | | | | | | | | | | | | In restore path snapd svc was not initialized because of which any glusterd instance which went down and came back may have uninitialized snapd svc. The reason I used 'may' is because depending on the nodes in the cluster. In a single node cluster this wouldn't be a problem since glusterd_spawn_daemon takes care of initializing it. Backport of http://review.gluster.org/10304 Change-Id: I2da1e419a0506d3b2742c1cf39a3b9416eb3c305 BUG: 1215518 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/10304 Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System Reviewed-by: Kaushal M <kaushal@redhat.com> (cherry picked from commit 18fd2fdd60839d737ab0ac64f33a444b54bdeee4) Reviewed-on: http://review.gluster.org/10397 Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com>
* tier: fix off-by-one overrun in UUID stringEmmanuel Dreyfus2015-04-271-1/+1
| | | | | | | | | | | | | | | | | | UUID strings are UUID_CANONICAL_FORM_LEN (36) bytes long plus the trailing nul character that various function (e.g.: uuid_unparse) will add. As a consequence, UUID strings must be declared as UUID_CANONICAL_FORM_LEN+1 long, otherwise we get a off-by-one overrun that corrupts the next variable on stack. Backport of: I5837ad6ca06fa17cc7ab143eedd02d8099ecca2a BUG: 1212676 Change-Id: I27bd223bdd72cda679b7cdc861e09e0f367bd6e4 Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/10395 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* features/trash: Minor coverity fixesAnoop C S2015-04-261-4/+23
| | | | | | | | | | | | | | | | | | | | Backport of http://review.gluster.org/10315 CID 1288784 CID 1288785 CID 1288795 CID 1288796 CID 1288797 CID 1288802 Change-Id: I51dd7653a2dce3b7b6387e5d91c1c07eb157a04b BUG: 1215026 Signed-off-by: Anoop C S <achiraya@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: jiffin tony Thottan <jthottan@redhat.com> Reviewed-by: jiffin tony Thottan <jthottan@redhat.com> Reviewed-on: http://review.gluster.org/10361 Tested-by: NetBSD Build System
* features/shard: Implement unlink fopKrutika Dhananjay2015-04-262-233/+495
| | | | | | | | | | | | Backport of: http://review.gluster.org/10249 Change-Id: I01761721224c4efbbc5e4992e70ecf68b3868d63 BUG: 1214247 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/10377 Tested-by: Gluster Build System <jenkins@build.gluster.com> Tested-by: NetBSD Build System Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* tiering: skip CTR-xlator and config options when disabledNiels de Vos2015-04-262-0/+7
| | | | | | | | | | | | | | | | | | | | | When compiling with --disable-tiering, GlusterD should not contain any references to the ChangeTimeRecorder xlator or any of the data tiering options. Cherry picked from commit a675ab96b917fc48fc3d7ca035590ebd7cf102bd: > BUG: 1213125 > Change-Id: Idb46fb80f0ca8b66115e06841d9ec15ba14c24a0 > Signed-off-by: Niels de Vos <ndevos@redhat.com> > Reviewed-on: http://review.gluster.org/10296 > Reviewed-by: Dan Lambright <dlambrig@redhat.com> > Tested-by: Gluster Build System <jenkins@build.gluster.com> BUG: 1215382 Change-Id: Idb46fb80f0ca8b66115e06841d9ec15ba14c24a0 Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/10378 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Dan Lambright <dlambrig@redhat.com>
* features/shard: Consume size and block count in metadata read opsKrutika Dhananjay2015-04-262-98/+271
| | | | | | | | | | | | | | | | | | Backport of : http://review.gluster.org/10098 Metadata read fops like lookup, stat etc will now fetch the xattr that holds the size and block count information, extract the size and block count fields and set them in respective stbuf before unwinding the resultant iatt to the parent xlator. Change-Id: If7d2c4af886f8d70dc69d7cb09f1f66be391f198 BUG: 1214248 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/10331 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* quota/marker: fix inode quota healing after glusterfs upgradevmallika2015-04-241-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a backport of http://review.gluster.org/10239 > There is a problem during upgrade where, inode quotas are not healed in > the contri xattrs. > Healing happens if contri xattrs are missing. > But healing doesn't happen if contri xattrs are present and inode quota > values are missing in the contri xattrs. > > This patch fixes the problem > > Change-Id: I6c88b74b5bb333a97c5419e24cc4ada82839f474 > BUG: 1211808 > Signed-off-by: vmallika <vmallika@redhat.com> > Reviewed-on: http://review.gluster.org/10239 > Tested-by: NetBSD Build System > Reviewed-by: Sachin Pandit <spandit@redhat.com> > Tested-by: Gluster Build System <jenkins@build.gluster.com> > Reviewed-by: Raghavendra G <rgowdapp@redhat.com> > Tested-by: Raghavendra G <rgowdapp@redhat.com> Change-Id: I3245ee62fd69b60ec56ebe28043c273dbb58dd58 BUG: 1212348 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/10265 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* features/trash : fixing trash dir optionJiffin Tony Thottan2015-04-231-3/+14
| | | | | | | | | | | | | | | | Backport of http://review.gluster.org/#/c/10215/ Previously, problem was caused due to buffer overflow of variable used in the code. This patch fix the same. Change-Id: I3df5e06044470022f9475d93d33447db35384da2 BUG: 1214563 Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com> Reviewed-on: http://review.gluster.org/10345 Tested-by: NetBSD Build System Reviewed-by: Emmanuel Dreyfus <manu@netbsd.org> Tested-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* Fix case mistake for MKDIR_P in MakefilesEmmanuel Dreyfus2015-04-221-1/+1
| | | | | | | | | | | | | | | | | | | | | Some makefiles used $(mkdir_p) instead of the corectly defined $(MKDIR_P). The former is substituted as an empty string, leading to possible failures depending of the user shell tolerance. NetBSD's /bin/sh seems to choke more easily than Linux's /bin/bash, but if the later does not fail, it does not created the intended directories anyway. Resubmit after rebase in hope that regression test will have less spurious failures. Backport of: I8caed4000f3c91cb3a685453848fb854793945ed BUG: 1212676 Change-Id: I393ffda8aa24bffce86f7b1f19b999ffa30c3562 Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/10279 Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: Store ping-timeout in glusterd conf in memoryAtin Mukherjee2015-04-141-0/+5
| | | | | | | | | | | | | As of now glusterd_conf_t's ping-timeout was of no use and hence statedump was not reflecting correct ping-timeout value configured in glusterd.vol Change-Id: I76ba7b721264565751538c7fcdcb801a28317581 BUG: 1210627 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: http://review.gluster.org/10191 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* quota/disperse: handle inode quotas in xattr aggregatevmallika2015-04-141-13/+22
| | | | | | | | | | | | | | | | | | | | with the inode quota feature, quota size is now increased from 64bit to 192bits which contains values of 'file size', 'file count' and 'dir count' This change in quota size xattr needs to be handled in disperse xattr aggregation Signed-off-by: vmallika <vmallika@redhat.com> Change-Id: I5fd28aa9f5b8b6cba83a98360236417a97ac16ee BUG: 1207967 Reviewed-on: http://review.gluster.org/10112 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Sachin Pandit <spandit@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Tested-by: Raghavendra G <rgowdapp@redhat.com>
* cluster/ec: Entry self-heal fixesPranith Kumar K2015-04-131-79/+27
| | | | | | | | | | | | | | | | - Directory deletion should always happen with 'rm -rf' flag, otherwise the call may fail with ENOTEMPTY. - Instead of doing an explicit 'link' call, perform mknod call with GLUSTERFS_INTERNAL_FOP_KEY which acts as 'link' if the gfid already exists. Change-Id: I8826f92170421db37efb67dfc00afad4ab695907 BUG: 1207085 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/10045 Tested-by: NetBSD Build System Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
* NFS-Ganesha : Fixing HA script invocation and othersMeghana Madhusudhan2015-04-132-4/+27
| | | | | | | | | | | | | | gluster features.ganesha disable failed invariably. And also, there were problems in unexporting volumes dynamically.Fixed the above problems. Change-Id: I29aa289dc8dc7b39fe0fd9d3098a02097ca8ca0c BUG: 1207629 Signed-off-by: Meghana Madhusudhan <mmadhusu@redhat.com> Reviewed-on: http://review.gluster.org/10199 Reviewed-by: jiffin tony Thottan <jthottan@redhat.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Tested-by: NetBSD Build System
* glusterd: Restarting glusterd should not reset bitrot tunable value to defaultGaurav Kumar Garg2015-04-132-3/+26
| | | | | | | | | | | | | | | | | | | Previously when user restart glusterd then bitrot tunable value for option "scrub-throttle", "scrub-frequency", "scrub" cleared. glusterd was not storing these bitrot tunable value properly. With this fix it will store scrub-throttle, scrub-frequency, scrub, bitrot tunable value in /var/lib/glusterd/vols/<volname>/info file and it will resume these value when glusterd restart. Change-Id: I58e756e14fbb8513a6f878f808ea148e7aa33cb0 BUG: 1209751 Signed-off-by: Gaurav Kumar Garg <ggarg@redhat.com> Reviewed-on: http://review.gluster.org/10159 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Venky Shankar <vshankar@redhat.com>
* bitrot/glusterd: Scrubber should not crawl bricks if bitrot is not enableGaurav Kumar Garg2015-04-131-12/+11
| | | | | | | | | | | | | | | | | | If user create multiple volume and user enable bitrot only for few volume then scrubber should not crawl bricks of volume on which bitrot is not enable. Fix is while volfile generation check whether bitrot is enable on that particular volume or not. If bitrot is enable then only it add bricks of that volume in scrubber volfile. Change-Id: I53687c8e1acb39f8e4a712028b73f0b7006122b9 BUG: 1207547 Signed-off-by: Gaurav Kumar Garg <ggarg@redhat.com> Reviewed-on: http://review.gluster.org/10190 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Venky Shankar <vshankar@redhat.com>
* storage/posix: Introduce xattr-fill on fdsKrutika Dhananjay2015-04-134-85/+148
| | | | | | | | | | | ... with some of the code borrowed from http://review.gluster.org/#/c/3904/ Change-Id: I4901ef14d6f843d8d69f102d43d21b60ba298092 BUG: 1207603 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/10180 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* glusterd: Replace transaction peers listsKaushal M2015-04-1312-318/+404
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Transaction peer lists were used in GlusterD to peers belonging to a transaction. This was needed to prevent newly added peers performing partial transactions, which could be incorrect. This was accomplished by creating a seperate transaction peers list at the beginning of every transaction. A transaction peers list referenced the peerinfo data structures of the peers which were present at the beginning of the transaction. RCU protection of peerinfos referenced by the transaction peers list is a hard problem and difficult to do correctly. To have proper RCU protection of peerinfos, the transaction peers lists have been replaced by an alternative method to identify peers that belong to a transaction. The alternative method is to the global peers list along with generation numbers to identify peers that should belong to a transaction. This change introduces a global peer list generation number, and a generation number for each peerinfo object. Whenever a peerinfo object is created, the global generation number is bumped, and the peerinfos generation number is set to the bumped global generation. With the above changes, the algorithm to identify peers belonging to a transaction with RCU protection is as follows, - At the beginning of a transaction, the current global generation number is saved - To identify if a peers belonging to the transaction, - Start a RCU read critical section - For each peer in the global peers list, - If the peers generation number is not greater than the saved generation number, continue with the action on the peer - End the RCU read critical section The above algorithm guarantees that, - The peer list is not modified when a transaction is iterating through it - The transaction actions are only done on peers that were present when the transaction started But, as a transaction could iterate over the peers list multiple times, the algorithm cannot guarantee that same set of peers will be selected every time. A peer could get deleted between two iterations of the list within a transaction. This problem existed with transaction peers list as well, but unlike before now it will not lead to invalid memory access and potential crashes. This problem will be addressed seprately. This change was developed on the git branch at [1]. This commit is a combination of the following commits on the development branch. 52ded5b Add timespec_cmp 44aedd8 Add create timestamp to peerinfo 7bcbea5 Fix some silly mistakes 13e3241 Add start time to opinfo 17a6727 Use timestamp comparisions to identify xaction peers instead of a xaction peer list 3be05b6 Correct check for peerinfo age 70d5b58 Use read-critical sections for peer list iteration ba4dbca Use peerinfo timestamp checks in op-sm instead of xaction peer list d63f811 Add more peer status checks when iterating peers list in glusterd-syncop 1998a2a Timestamp based peer list traversal of mgmtv3 xactions f3c1a42 Remove transaction peer lists b8b08ee Remove unused labels 32e5f5b Remove 'npeers' usage a075fb7 Remove 'npeers' from mgmt-v3 framework 12c9df2 Use generation number instead of timestamps. 9723021 Remove timespec_cmp 80ae2c6 Remove timespec.h include a9479b0 Address review comments on 10147/4 [1]: https://github.com/kshlm/glusterfs/tree/urcu Change-Id: I9be1033525c0a89276f5b5d83dc2eb061918b97f BUG: 1205186 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.org/10147 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Anand Nekkunti <anekkunt@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* nfs: do not overwrite the return value and errno of resolve operationRaghavendra Bhat2015-04-131-3/+6
| | | | | | | | | | | | | | * In auth status check which is done in the resume function, the op_ret and op_errno values saved in nfs3_call_state are overwritten by the return value of the auth status check function. Change-Id: Id4682ddd399c78a1cef6313a534892ef309c57a6 BUG: 1210338 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/10179 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Venky Shankar <vshankar@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com>
* crypt xlator: Fix memory coruption after freeEmmanuel Dreyfus2015-04-112-1/+2
| | | | | | | | | | | | | | | | | | | | | | crypt xlator allocated local memory through mem_get0(), but code called through STACK_DESTROY()/FRAME_DESTROY() does not expect memory to be allocated that way: it will use GF_FREE() even for data allocated by mem_get0(), which should be given the mem_put() treatment. As a result, allocating using mem_get0(), while relying on FRAME_DESTROY() cleanup led to memory corruption. Using GF_CALLOC() instead of mem_get0() sets memory allocation on par with cleanup code, and crypt.t can pass on NetBSD. The initial patch was crafted by Raghavendra Talur. BUG: 1129939 Change-Id: Ib71b4b57f8d1bb782f950e3c8fa74a4f7e10946e Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/10109 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/ec: Fix readdir de-itransformPranith Kumar K2015-04-113-8/+73
| | | | | | | | | | | | | | | | | | | Problem: gf_deitransform returns the glbal client-id in the complete graph. So except for the first disperse subvolume under dht, all the other disperse subvolumes will return a client-id greater than ec->nodes, so readdir will always error out in those subvolumes. Fix: Get the client subvolume whose client-id matches the client-id returned by gf_deitransform of offset. Change-Id: I26aa17504352d48d7ff14b390b62f49d7ab2d699 BUG: 1209113 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/10165 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
* features/bit-rot: Mark versioning fsetxattr as internal fopKotresh HR2015-04-112-5/+10
| | | | | | | | | | | | | | | | | Changelog xlator was capturing bitrot-stub's fsetxattr sent for versioning. Since it was using the same frame as of the create fop, there was inconsistency in fop number and gfid of capturing metadata. So fix is to mark fsetxattr used for versioning as internal and add internal fop filter in changelog_fsetxattr. Change-Id: I51ff468995139838b22bf293a59a0713a92ee7a5 BUG: 1170075 Signed-off-by: Kotresh HR <khiremat@redhat.com> Reviewed-on: http://review.gluster.org/10148 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Venky Shankar <vshankar@redhat.com>
* build: make contrib/uuid dependency optionalNiels de Vos2015-04-1019-28/+20
| | | | | | | | | | | | | | | | | | | On Linux systems we should use the libuuid from the distribution and not bundle and statically link the contrib/uuid/ bits. libglusterfs/src/compat-uuid.h has been introduced and should become an abstraction layer for different UUID APIs. Non-Linux operating systems should implement their compatibility layer there. Once all operating systems have an implementation in compat-uuid.h, we can remove contrib/uuid/ from the repository completely. Change-Id: I345e5357644be2521685e00358bb8c83c4ea0577 BUG: 1206587 Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/10129 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* io-stats: Fixing dereference after null check.arao2015-04-101-1/+2
| | | | | | | | | | | | | CID: 1124492 If 'this' pointer fails to exist then the component is hard-coded in the gf_log if not it is dereferenced. Change-Id: I988137d4f5ac4c9aedef7cef0c75b167a8a5c59f BUG: 789278 Signed-off-by: arao <arao@redhat.com> Reviewed-on: http://review.gluster.org/9639 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* quota/cli: validate quota hard-limit optionvmallika2015-04-101-16/+12
| | | | | | | | | | | | | | | | | | | Quota hard-limit is supported only upto: 9223372036854775807 (int 64) In CLI, it is allowed to set the value upto 16384PB (unsigned int 64), this is not a valid value as the xattrop for quota accounting and the quota enforcer operates on a signed int64 limit value. This patches fixes the problem in CLI and allows user to set the hard-limit value only from range 0 - 9223372036854775807 Change-Id: Ifce6e509e1832ef21d3278bacfa5bd71040c8cba BUG: 1206432 Signed-off-by: vmallika <vmallika@redhat.com> Reviewed-on: http://review.gluster.org/10022 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaushal M <kaushal@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/ec: Ignore volume-mark key for comparing dictsPranith Kumar K2015-04-101-0/+1
| | | | | | | | | Change-Id: Id60107e9fb96588d24fa2f3be85c764b7f08e3d1 BUG: 1207712 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/10077 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
* cluster/ec: Fix iobuf mem leakRaghavendra Talur2015-04-101-2/+2
| | | | | | | | | | | | | | | | iobuf_get and iobref_add implicitly ref the iobuf. Hence, it is necessary to unref iobuf before setting it to NULL. Change-Id: Icadd8925574cf04fe708d8090868e49356653a8e Signed-off-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-on: http://review.gluster.org/9818 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: bitd daemon should not start on the node which dont have any brickGauravKumarGarg2015-04-104-22/+36
| | | | | | | | | | | | | | | | | | If user enable bitrot from node1 which have brick then glusterd starting bitd daemon on node1 as well as glusterd starting bitd deamon on another node2 which does not have any brick (node1 and node2 are part of cluster). With this fix glusterd will not start bitd daemon on the node which don't have brick. Change-Id: Ic1c68d204221d369d89d628487cdd5957964792e BUG: 1207029 Signed-off-by: Gaurav Kumar Garg <ggarg@redhat.com> Reviewed-on: http://review.gluster.org/10071 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Venky Shankar <vshankar@redhat.com>
* snapshot:Fetching wrong peer data during probeMohammed Rafi KC2015-04-101-5/+6
| | | | | | | | | | | | | | | | | | When adding a new friend to the cluster, the snap volfile are populating with wrong information for reconfigured option. For snap volumes, reconfigured data's are filling from the regular volumes data. This is because wrong dictionary key is used here. Change-Id: I659ebdc48c33419a2b825f26ce1f174abc8ea7dd BUG: 1204636 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: http://review.gluster.org/9969 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Avra Sengupta <asengupt@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* glusterd/snapshot : While snapshot restore, compute quota checksum.Sachin Pandit2015-04-104-32/+30
| | | | | | | | | | | | | | | | | | Problem : During snapshot restore we anyways copy the quota conf file after that we need to compute the checksum for that. If not, there might be a checksum mismatch during glusterd handshake. Solution : Compute a checksum file for quota conf file if its present. Change-Id: Ic4a6567c6ede9923443abf4ca59380679be88094 BUG: 1202436 Signed-off-by: Sachin Pandit <spandit@redhat.com> Reviewed-on: http://review.gluster.org/9901 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Avra Sengupta <asengupt@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* features/quota: Fixing Logically dead codearao2015-04-101-5/+0
| | | | | | | | | | | | | CID: 1134007 The code never reaches the condition check on retlen in ret label, hence removing the dead code. Change-Id: Ia0108b69489bb78a2561ff8da6e00685f472ae82 BUG: 789278 Signed-off-by: arao <arao@redhat.com> Reviewed-on: http://review.gluster.org/9644 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: remove-brick status/stop should not show output for non-existing brickGaurav Kumar Garg2015-04-091-17/+59
| | | | | | | | | | | | | | | | | | | Previously when user start remove-brick operation on a volume then by giving non-existing brick for remove-brick status/stop command it was showing remove-brick status/stoping remove-brick operation on a volume. With this fix it will validate bricks which user have given for remove-brick status/stop command and if bricks are part of volume then it will show statistics of remove-brick operation otherwise it will show error "Incorrect brick <brick_name> for <volume_name>". Change-Id: I151284ef78c25f52d1b39cdbd71ebfb9eb4b8471 BUG: 1121584 Signed-off-by: Gaurav Kumar Garg <ggarg@redhat.com> Reviewed-on: http://review.gluster.org/9681 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
* Snapshot: Remove brick details on snapshot create errorMohammed Rafi KC2015-04-091-6/+8
| | | | | | | | | | | | | | Since we are not providing complete details of stopped bricks on snapshot create error, it is better to remove the incomplete brick details. BUG: 1205596 Change-Id: I73386059c0d29a6cbe7f2bb6834d8b8fb393db0a Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: http://review.gluster.org/9999 Reviewed-by: Avra Sengupta <asengupt@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: Better GlusterD handshake request validationKaushal M2015-04-091-13/+49
| | | | | | | | | | | | | | | | | | | | | | | | | | | | To allow handshake requests to be validated correctly in a multi network environment, the request validation process has been improved. The handshake request initiator will add it's peer ID the request. The handshake request reciever will allow a request (as before) if, - it has no peers, or - the request came from a known peer Identifying the known peer is done as follows. - If the request contains a peer ID, it is matched against the IDs in the peer list. If a match is found, the request is allowed. - The address of the incoming request is matched against the peer addresses in the peer list. If a match is found, the request is allowed. - Otherwise, the request if disallowed Change-Id: I9eabe2935d16276bb147dfeebf8c8beb08e01411 BUG: 1207611 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.org/10122 Reviewed-by: Atin Mukherjee <amukherj@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>
generated by cgit (git 2.34.1) at 2025-07-14 09:51:33 +0000