summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
* glusterd: validate performance.nfs.* option values during volume set stageKrutika Dhananjay2013-04-181-2/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PROBLEM: performance.nfs.* option values (which are of type boolean) are not validated during the stage phase of 'volume set'. The result - nfs graph generation fails during commit phase, AFTER the option and its (invalid) value have been placed in volinfo->dict. CAUSE: nfsperfxl_option_handler() - the function that validates the values of performance.nfs.* options - never receives the (key,value) pair that needs to be set, for validation during 'volume set' stage. FIX: In build_nfs_graph(), copy the (mod_)dict containing the (option,value) parameters into set_dict before attempting to build the client graph for the volume on which the operation is being performed. Of course, an easier way out would be to simply do a 'volume reset' and pretend nothing wrong happened! Change-Id: I56b17d0239d58a9e0b7798933a3c8451e2675b69 BUG: 949930 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/4814 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: Avoided deadlock in single node cluster, glusterd restartKrishnan Parthasarathi2013-04-161-0/+3
| | | | | | | | | | | | | | | | | | | In a single node cluster, it is possible to deadlock on the "big lock", while restarting bricks. In glusterd_restart_bricks, we perform a glusterd_brick_connect, where we release the big lock in anticipation that glusterd_brick_rpc_notify could run in the same C stack (and deadlocking). So, in the restart code path, we could unlock before we have performed a lock on the big lock. To fix this, we need to take the big lock in the glusterd_launch_synctask 'thread' as well. Change-Id: I1abea1ca82b55c784b8a810a8194f254b32b1dcc BUG: 948686 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/4837 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* glusterd: big lock - a coarse-grained locking to prevent racesKrishnan Parthasarathi2013-04-1216-106/+656
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are primarily three lists that are part of glusterd process, that are concurrently accessed. Namely, priv->volumes, priv->peers and volinfo->bricks_list. Big-lock approach ----------------- WHAT IS IT? Big lock is a coarse-grained lock which protects all three lists, mentioned above, from racy access. HOW DOES IT WORK? At any given point in time, glusterd's thread(s) are in execution _iff_ there is a preceding, inbound network event. Of course, the sigwaiter thread and timer thread are exceptions. A network event is an external trigger to glusterd, via the epoll thread, in the form of POLLIN and POLLERR. As long as we take the big-lock at all such entry points and yield it when we are done, we are guaranteed that all the network events, accessing the global lists, are serialised. This amounts to holding the big lock at - all the handlers of all the actors in glusterd. (POLLIN) - all the cbks in glusterd. (POLLIN) - rpc_notify (DISCONNECT event), if we access/modify one of the three lists. (POLLERR) In the case of synctask'ized volume operations, we must remember that, if we held the big lock for the entire duration of the handler, we may block other non-synctask rpc actors from executing. For eg, volume-start would block in PMAP SIGNIN, if done incorrectly. To prevent this, we need to yield the big lock, when we yield the synctask, and reacquire on waking up of the synctask. Change-Id: Ib929f9905b55fb6c3fc27fefb497a26dba058e4f BUG: 948686 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/4784 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* license: xlators/protocol/server dual license GPLv2 and LGPLv3+Kaleb S. KEITHLEY2013-04-1210-145/+57
| | | | | | | | | BUG: 951549 Change-Id: I3de5bd86d4238a60a0a85ba2e15d9c131969b210 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: http://review.gluster.org/4816 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: Fixed spurious wakeups in glusterd syncopsKrishnan Parthasarathi2013-04-122-15/+26
| | | | | | | | | | | | | glusterd syncops perform a barrier_wake whenever rpc_clnt_submit returned -1. This is based on the wrong assumption that the cbkfn wasn't called. This would result in one more wakeup than there ought to be. Change-Id: I591e67c267f0e26d1145bf8fb5feeb2c13a751a1 BUG: 948686 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/4802 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* glusterd: changes in 'volume create' behaviourKrutika Dhananjay2013-04-095-20/+137
| | | | | | | | | | | | | This patch incorporates all the changes suggested on the behaviour of 'volume create' command in http://review.gluster.org/#change,4214 (comment #14, to be precise). Change-Id: Iaac524a59738b177415595b18aa8a136090d3d25 BUG: 948729 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/4740 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* mgmt/glusterd: enable valgrind usage even in non DEBUG buildRaghavendra Bhat2013-04-094-24/+13
| | | | | | | | | | | | * Till now running glusterfs processes were allowed to run in valgrind mode only when built with debug mode enabled. Change-Id: I11e07ea2a4da4f82f70cdded6258a22d65d6db64 BUG: 922877 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.org/4688 Reviewed-by: Anand Avati <avati@redhat.com> Tested-by: Anand Avati <avati@redhat.com>
* cluster/distribute: Ignore non-participating subvols for layout checksshishir gowda2013-04-092-20/+88
| | | | | | | | | | | | | | | | | | | | | | | When subvols-per-directory is < available subvols, then there are layouts which are not populated. This leads to incorrect identification of holes or overlaps. We need to ignore layouts, which have err == 0, and start == stop. In the current scenario (start == stop == 0). Additionally, in layout-merge, treat missing xattrs as err = 0. In case of missing layouts, anomalies will reset them. For any other valid subvoles, err != 0 in case of layouts being zeroed out. Also reverted back dht_selfheal_dir_xattr, which does layout calculation only on subvols which have errors. Change-Id: I9f57062722c9e8a26285e10675c31a78921115a1 BUG: 921408 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4668 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* afr: let eager-locking do its own overlap checksAnand Avati2013-04-054-56/+87
| | | | | | | | | | | | | | | | | | | | | | | | Today there is a non-obvious dependence of eager-locking on write-behind. The reason is that eager-locking works as long as the inheriting transaction has no overlaps with any of the transactions already in progress. While write-behind provides non-overlapping writes as a side-effect most of times (and only guarantees it when strict-write-ordering option is enabled, which is not on by default) eager-lock needs the behavior as a guarantee. This is leading to complex and unwanted checks for the presence of write-behind in the graph, for the simple task of checking for overlaps. This patch removes the interdependence between eager-locking and write-behind by making eager-locking do its own overlap checks with in-progress writes. Change-Id: Iccba1185aeb5f1e7f060089c895a62840787133f BUG: 912581 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4782 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* storage/posix: introduce node-uuid-pathinfoVenky Shankar2013-04-053-2/+39
| | | | | | | | | | | | | | | enabling this option has an effect on pathinfo xattr request returning <node-uuid>:<path> instead of the default - which is <hostname>:<path>. Change-Id: Ice1b38abf8e5df1568bab6d79ec0d53dfa520332 BUG: 765380 Signed-off-by: Venky Shankar <vshankar@redhat.com> Reviewed-on: http://review.gluster.org/4567 Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* gsync: Display additional information in status commandsarvotham s pai2013-04-042-4/+52
| | | | | | | | | | | | | | | | | | | | | | | | | | Added code to display extra information when status command is executed. Information shown now are 1 Number of files synced 2 crawl time 3 total sync time 4 bytes synced bytes synced is taken from rsync output . --stats option of rsync gives extra infor mation about the sync.In stats output there is a field called Total transferred file size which states the ammount of bytes synced . This information is parsed from stdout output using regular expressions.Bytes synced information can be used to calculate throughput. Change-Id: Id9bba9fff45ee7049bb8257c6fd918e5237e05b1 BUG: 947774 Signed-off-by: sarvotham s pai <spai@redhat.com> Reviewed-on: http://review.gluster.org/4749 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Treat all dir fop failure as success in changelogPranith Kumar K2013-04-033-2/+35
| | | | | | | | | | | | | | | For example: If a new entry creation fop fails with EEXIST or a delete entry fop fails with ENOENT, on all the subvols the fop is wound, then no change took place to the directory. So we can treat that case as no change happened to the directory. Change-Id: I3b3a7931954da2166a9cba19ff9f76f37739d751 BUG: 860210 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4626 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* posix: fix dangerous "sharing" of fd in readdir between two requestsAnand Avati2013-04-031-2/+17
| | | | | | | | | | | | | | | | | | | | | | | posix_fill_readdir() is a multi-step function which performs many readdir() calls, and expects the directory cursor to have not "seeked away" elsewhere between two successive iterations. Usually this is not a problem as each opendir() from an application has its own backend fd, and there is nobody else to "seek away" the directory cursor. However in case of NFS's use of anonymous fd, the same fd_t is shared between all NFS readdir requests, and two readdir loops can be executing in parallel on the same dir dragging away the cursor in a chaotic manner. The fix in this patch is to lock on the fd around the loop. Another approach could be to reimplement posix_fill_readdir() with a single getdents() call, but that's for another day. Change-Id: Ia42e9c7fbcde43af4c0d08c20cc0f7419b98bd3f BUG: 948086 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4774 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/afr: Made afr_sh_purge_entry_common message log more clear.Venkatesh Somyajulu2013-04-031-3/+1
| | | | | | | | | | | | | | | FIX: In missing entry self heal, once the source directories are determined after the lookup and if file is not present on any of the brick which contains the souce directory, the entry is removed from the directory. So log message should give information of "Purging of entry". Change-Id: I4d3deb602e0812dc1c9c8ba0a466716d81dede7e BUG: 947312 Signed-off-by: Venkatesh Somyajulu <vsomyaju@redhat.com> Reviewed-on: http://review.gluster.org/4753 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* pump: Set self-heal readdir size in pumpPranith Kumar K2013-04-031-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: In Pump entry self-heal happens for each directory during the first opendir using conservative merge. But in entry-self-heal readdir is issued with '0' size. So entry self-heal is not creating any files. After pump thinks entry self-heal is complete it proceeds to heal each of the file in the directory it just healed. Fortunately most of the times it chooses source-brick in pump as read-child for readdir. This happens because readchild is the subvolume on which lookup succeeds first. In pump lookup succeeds faster in local process than on the destination brick process most of the times. For all the entries pump finds in readdir it does a lookup. During this lookup the entry on the destination brick is created and healed. This is the reason why replace-brick succeeds whenever read-child for the directory is chosen as the source-brick. Which is most of the times. When read-child is chosen as the destination brick, readdir returns no entries so replace-brick completes without syncing the whole data. Fix: Set readdir-size in pump so that entry self-heal happens with 64k size. This ensures that entry self-heal triggered from opendir actually creates the files on the destination brick. Change-Id: I65ea45d3c2735a9578f3aa34eff771b6563241ca BUG: 909800 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4712 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* protocol/client: Print valid loc identifiersPranith Kumar K2013-04-031-35/+27
| | | | | | | | | | Change-Id: I45f91105862a2484b8906a7a63b98ab4aaf80d05 BUG: 924643 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4683 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* dht: make nufa/switch call dht's init/finiJeff Darcy2013-04-036-1032/+786
| | | | | | | | | | | | | These functions keep changing as new functionality is added, so copying and pasting the code is not a good solution. This way ensures that all fields get initialized properly no matter how much new stuff we throw in. Change-Id: I9e9b043d2d305d31e80cf5689465555b70312756 BUG: 924488 Signed-off-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-on: http://review.gluster.org/4710 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/distribute: Start rebalance with option readdir-optimize onshishir gowda2013-04-031-0/+1
| | | | | | | | | | | | | | With readdir-optimize set to on, we instruct the posix layer to ignore directory entries from not first subvolume. DHT discards directories returned from non first subvolume. By making posix itself ignore it, we are making directory crawls faster Change-Id: Ia1faf2dedec0c615c0632c3c063e846f5742ede6 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/4613 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: add more specific log messagesBala.FA2013-04-021-8/+8
| | | | | | | | | Change-Id: I57fbdd83f3098e64886c3dd690a1ae04fc37442d BUG: 928648 Signed-off-by: Bala.FA <barumuga@redhat.com> Reviewed-on: http://review.gluster.org/4739 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: detect in-progress creation in lookup and return ENOENTPranith Kumar K2013-04-021-0/+57
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | if any subvol returned ENOENT while parent entrylk lock was held, yield and return ENOENT for the entire lookup. This is how the issue happens: Multiple clients A, B and C are attempting 'mkdir -p /mnt/a/b/c' 1 Client A is in the middle of mkdir(/a). It has acquired lock. It has performed mkdir(/a) on one subvol, and second one is still in progress 2 Client B performs a lookup, sees directory /a on one, ENOENT on the other, succeeds lookup. 3 Client B performs lookup on /a/b on both subvols, both return ENOENT (one subvol because /a/b does not exist, another because /a itself does not exist) 4 Client B proceeds to mkdir /a/b. It obtains entrylk on inode=/a with basename=b on one subvol, but fails on other subvol as /a is yet to be created by Client A. 5 Client A finishes mkdir of /a on other subvol 6 Client C also attempts to create /a/b, lookup returns ENOENT on both subvols. 7 Client C tries to obtain entrylk on on inode=/a with basename=b, obtains on one subvol (where B had failed), and waits for B to unlock on other subvol. 8 Client B finishes mkdir() on one subvol with GFID-1 and completes transaction and unlocks 9 Client C gets the lock on the second subvol, At this stage second subvol already has /a/b created from Client B, but Client C does not check that in the middle of mkdir transaction 10 Client C attempts mkdir /a/b on both subvols. It succeeds on ONLY ONE (where Client B could not get lock because of missing parent /a dir) with GFID-2, and gets EEXIST from ONE subvol. This way we have /a/b in GFID mismatch. One subvol got GFID-1 because Client B performed transaction on only one subvol (because entrylk() could not be obtained on second subvol because of missing parent dir -- caused by premature/speculative succeeding of lookup() on /a when locks are detected). Other subvol gets GFID-2 from Client C because while it was waiting for entrylk() on both subvols, Client B was in the middle of creating mkdir() on only one subvol, and Client C does not "expect" this when it is between lock() and pre-op()/op() phase of the transaction. Original-author: Anand Avati <avati@redhat.com> Change-Id: Idca475dbbc2a51e09da6fa0f9e1e37148caef208 BUG: 860210 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4625 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* rpc/nfs: cleanup legacy code of general optionsRajesh Amaravathi2013-04-021-2/+1
| | | | | | | | | | | | | | | | | | Removing the code which handles "general" options. Since it is no longer possible to set general options which apply for all volumes by default, this was redundant. This cleanup of general options code also solves a bug wherein with nfs.addr-namelookup on, nfs.rpc-auth-reject wouldn't work on ip addresses Change-Id: Iba066e32f9a0255287c322ef85ad1d04b325d739 BUG: 921072 Signed-off-by: Rajesh Amaravathi <rajesh@redhat.com> Reviewed-on: http://review.gluster.org/4691 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/afr: sync xattrs removed on source to sink(s)Venky Shankar2013-04-023-12/+125
| | | | | | | | | | | | | xattrs are first removed from sink followed by setting source xattrs. Change-Id: I181cb5b785b667bbfc6e40787a2183a8f45de06b BUG: 906646 Signed-off-by: Venky Shankar <vshankar@redhat.com> Reviewed-on: http://review.gluster.org/4656 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* cluster/afr: prevent piggyback on stale pre_opPranith Kumar K2013-04-021-33/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here are the logs of a file on which we saw EIO because of size mismatch: [root@lizzie ~]# grep 38f18204 /var/log/glusterfs/mnt-x-.log Reporting Unstable write for 38f18204-2840-408e-ae65-c01f4106b8c4 for offset: 0, len: 7680 Cleared unstable write flag for 38f18204-2840-408e-ae65-c01f4106b8c4: offset 0 length 7680 Reporting Unstable write for 38f18204-2840-408e-ae65-c01f4106b8c4 for offset: 7680, len: 71680 Reporting Unstable write for 38f18204-2840-408e-ae65-c01f4106b8c4 for offset: 79360, len: 15716 fsync completed on 38f18204-2840-408e-ae65-c01f4106b8c4 for offset 0 length 7680 with changelog status: -1 -1 According to these logs fsync did not happen after writev with offset: 79360, len: 15716. Which is the reason for this problem. In total 3 writes came. lets call them w1, w2, w3 w1 does pre_op so pre_op_done[0], pre_op_done[1] counts become 1 and 1 then is_piggyback_post_op() is called for w1 and it returns *false* w1's fsync is fired Now w2 and w3 come and see that pre_op_done[0], pre_op_done[1] are both 1, so pre_op_piggyback[0] and pre_op_piggyback[1] are both incremented twice, once by w2, one more time by w3 and become 2, 2 ------- Step-A Now fsync of w1 is complete and it goes ahead with post op and decrements pre_op_done[0], pre_op_done[1] to 0, 0 Now w2, w3 writevs complete and is_piggyback_post_op will return *true* for both w2, w3. So fsync is not fired for both w2, w3 this patch prevents Step-A from happening. Change-Id: I8b6af1f1875b2cf5f718caa3c16ee7ff3dc96b5c BUG: 927146 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4752 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* dht: improve transform/detransform of d_off (and be ext4 safe)Anand Avati2013-04-011-5/+82
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The scheme to encode brick d_off and brick id into global d_off has two approaches. Since both brick d_off and global d_off are both 64-bit wide, we need to be careful about how the brick id is encoded. Filesystems like XFS always give a d_off which fits within 32bits. So we have another 32bits (actually 31, in this scheme, as seen ahead) to encode the brick id - which is typically plenty. Filesystems like the recent EXT4 utilize the upto 63 low bits in d_off, as the d_off is calculated based on a hash function value. This leaves us no "unused" bits to encode the brick id. However both these filesystmes (EXT4 more importantly) are "tolerant" in terms of the accuracy of the value presented back in seekdir(). i.e, a seekdir(val) actually seeks to the entry which has the "closest" true offset. This "two-prong" scheme exploits this behavior - which seems to be the best middle ground amongst various approaches and has all the advantages of the old approach: - Works against XFS and EXT4, the two most common filesystems out there. (which wasn't an "advantage" of the old approach as it is borken against EXT4) - Probably works against most of the others as well. The ones which would NOT work are those which return HUGE d_offs _and_ NOT tolerant to seekdir() to "closest" true offset. - Nothing to "remember in memory" or evict "old entries". - Works fine across NFS server reboots and also NFS head failover. - Tolerant to seekdir() to arbitrary locations. Algorithm: Each d_off can be encoded in either of the two schemes. There is no requirement to encode all d_offs of a directory or a reply-set in the same scheme. The topmost bit of the 64 bits is used to specify the "type" of encoding of this particular d_off. If the topmost bit (bit-63) is 1, it indicates that the encoding scheme holds a HUGE d_off. If the topmost bit is is 0, it indicates that the "small" d_off encoding scheme is used. The goal of the "small" d_off encoding is to stay as dense as possible towards the lower bits even in the global d_off. The goal of the HUGE d_off encoding is to stay as accurate (close) as possible to the "true" d_off after a round of encoding and decoding. If DHT has N subvolumes, we need ROOF(Log2(N)) "bits" to encode the brick ID (call it "n"). SMALL d_off =========== Encoding -------- If the top n + 1 bits are free in a brick offset, then we leave the top bit as 0 and set the remaining bits based on the old formula: hi_mask = 0xffffffffffffffff hi_mask = ~(hi_mask >> (n + 1)) if ((hi_mask & d_off_brick) != 0) do_large_d_off_encoding () d_off_global = (d_off_brick * N) + brick_id Decoding -------- If the top bit in the global offset is 0, it indicates that this is the encoding formula used. So decoding such a global offset will be like the old formula: if ((d_off_global & 0x8000000000000000) != 0) do_large_d_off_decoding() d_off_brick = (d_off_global % N) brick_id = d_off_global / N HUGE d_off ========== Encoding -------- If the top n + 1 bits are NOT free in a given brick offset, then we set the top bit as 1 in the global offset. The low n bits are replaced by brick_id. low_mask = 0xffffffffffffffff << n // where n is ROOF(Log2(N)) d_off_global = (0x8000000000000000 | d_off_brick & low_mask) + brick_id if (d_off_global == 0xffffffffffffffff) discard_entry(); Decoding -------- If the top bit in the global offset is set 1, it indicates that the encoding formula used is above. So decoding would look like: hi_mask = (0xffffffffffffffff << n) low_mask = ~(hi_mask) d_off_brick = (global_d_off & hi_mask & 0x7fffffffffffffff) brick_id = global_d_off & low_mask If "losing" the low n bits in this decoding of d_off_brick looks "scary", we need to realize that till recently EXT4 used to only return what can now be expressed as (d_off_global >> 32). The extra 31 bits of hash added by EXT recently, only decreases the probability of a collision, and not eliminate it completely, anyways. In a way, the "lost" n bits are made up by decreasing the probability of collision by sharding the files into N bricks / EXT directories -- call it "hash hedging", if you will :-) Thanks-to: Zach Brown <zab@redhat.com> Change-Id: Ieba9a7071829d51860b7c131982f12e0136b9855 BUG: 838784 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4711 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* mgmt/glusterd: Enable write-behind in nfsPranith Kumar K2013-04-011-1/+1
| | | | | | | | | | | | | We observed that the number of write requests thus inodelks are increasing very rapidly to thousands without write-behind in the graph. Change-Id: Id71c9c2b0a4c9601a4644a58a933221c62dab0c0 BUG: 928341 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4734 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* storage/posix: honor O_SYNC and O_DSYNC sent in @flags of writev()Anand Avati2013-03-292-5/+10
| | | | | | | | | | | | | | | | Historic bug - posix_writev() has been inspecting pfd->flushwrites for performing fsync() after write, instead of @flags for O_SYNC|O_DSYNC. pfd->flushwrites was never set anywhere and is unused completely. This is behavior from the time before anonymous FD where open() had @wbflags param. This is a leftover from that cleanup. Change-Id: Id9bfe562a60db4eb3bd0a7705bdba91f2df2f3ec BUG: 916372 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4738 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* cluster/afr: fix fd leak with unsafe call_resume()Anand Avati2013-03-282-1/+17
| | | | | | | | | | | | | | Introduce AFR_CALL_RESUME macro which cleans up frame->local, like how AFR_STACK_UNWIND etc. do. Therefore fix leak in afr_fsync() path. Change-Id: I3855d8e7e84dbc44e05f507563b7f722bf9621b8 BUG: 927146 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4745 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/afr: fsync before erase xattrs in data self-healPranith Kumar K2013-03-281-1/+75
| | | | | | | | | | | | Added extra fsync to data self-heal code to make sure the data reached disk before erasing the changelogs Change-Id: I9e7e6e55cdc49de2b991705d1638946464a9d4f9 BUG: 927146 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4744 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: piggyback and fsync resume changesPranith Kumar K2013-03-282-15/+18
| | | | | | | | | | | | | | 1) pre_op_piggyback should always be decremented. 2) Move fsync resume to just after post_op. 3) fsync stub should be created from afr's local not from the final response. Change-Id: I220bb532eb03bea584292f4dd2e816ad0c3e0cf7 BUG: 927146 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4741 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: fsync() guarantees POST-OP completionAnand Avati2013-03-273-10/+56
| | | | | | | | | | | | | | | | | | | | | | | | AFR now provides a stronger guarantee that fsync() returns only after completely finishing all the deferred/delayed POST-OP on that open file. To acheive this we make a stub out of the returning fsync and register it with the "delayed" frame in afr_changelog_wake_resume(). The delayed frame, after getting woken up and finishing the POST-OP will call_resume() the registered stub (which UNWINDs the fsync) at the time of frame destruction. This provides a guarantee that an application's (or FUSE) fsync() returns only after finishing up all the previous transactions, including delayed POST-OPs and UNLOCK. Change-Id: Iaa955457e2f25088a144fde37ad0444277b5cf49 BUG: 927146 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4737 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* cluster/afr: ensure DATA operations are made durable before POST-OPAnand Avati2013-03-274-22/+314
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The changelogging scheme of AFR stores information about the state of all replicas in all replicas (in the extended attribute of the respective files on each server) in the form of 'pending counts' of operations (effectively "dirty flags"). These xattrs are blindly trusted while performing self-heal, and therefore utmost care has to be taken while updating and maintaing them. The most critical updation is the clearing of the pending counts corresponding to the *other* server in the changelog of a given server. Before clearing the pending count, we need durability guarantee of the write which was performed on the other server. To obtain such a guarantee, it may be necessary to explicitly introduce an fsync() phase (if the file itself wasn't already opened with O_SYNC). This patch introduces the detection of unstable stable writes on a file and issues explicit fsync() on the servers before performing the POST-OP clearing of pending flags. Change-Id: I2171b86a74ec91e40e5877eef0a4e7379578ecf7 BUG: 927146 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4721 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* glusterd: Removed fd leaks in glusterfs_start utility functionKrishnan Parthasarathi2013-03-254-119/+79
| | | | | | | | | | | | | | | | | | | PROBLEM: The FILE* associated with the pidfile was leaked if pmap_registry_search on the brickinfo' path failed. FIX: Eliminates the use of the FILE* that was leaked. Uses glusterd_is_service_running utility function in place of the earlier attempt to check for the same. Change-Id: I94082bd5a94b8a6340f8cc11726d3264e364efe6 BUG: 916549 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/4596 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* config: better (i.e. more portable) test for libxml2Kaleb S. KEITHLEY2013-03-251-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Over the weekend I tried to build on MacOS X¹ and ran into the following issues: 1) The recent change to autogen.sh to test for pkg-config falls down. 2) After removing the pkg-config test in autogen.sh, w/o pkg-config the PKG_CHECK_MODULES macro invocation in configure[.ac] falls down. N.B. Solaris users run into this too, even through there's a (broken) pkg-config package that can be installed. 3) There are other problems in the code related to fuse that are beyond the scope of this. It seems that pkg-config is only a requirement for the definition of the PKG_CHECK_MODULES macro used to detect libxml2. Since this seems to be inherently unportable — at least to MacOS X and Solaris — I'd like to: A) Change the use of the PKG_CHECK_MODULES macro to the more portable AM_PATH_XML2 macro provided by the libxml2 package in /usr/.../share/aclocal/libxml.m4 2) Revisit the decision to add the check for pkg-config in autogen.sh in BZ 921817. For now this is just an rfc. If people are agreeable I'll reenter this change against BZ 921817. ¹Mountain Lion 10.8.3, XCode 4.6.1 Change-Id: I237b1ed8919088345b8fd943423b2a6ad289981b BUG: 921817 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: http://review.gluster.org/4720 Reviewed-by: Justin Clift <jclift@redhat.com> Tested-by: Justin Clift <jclift@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: Simplify glusterd_service_stop()Krishnan Parthasarathi2013-03-251-70/+11
| | | | | | | | | Change-Id: I396d250a3299ad1f7fce4bd14389b0c2756b6cb0 BUG: 764890 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/4718 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* mount: Added the xlator-option to mount.glusterfs script.Avra Sengupta2013-03-221-0/+6
| | | | | | | | | | | | | | | Now all xlator-options can be set from the mount command as well. Example : mount -t glusterfs Hostname:/Volume_Name Mount_Point -o "xlator-option=xyz=123, xlator-option=abc=999" Change-Id: If52d994986839d1c969e3e2e01b2e1a29a3140b7 BUG: 920583 Signed-off-by: Avra Sengupta <asengupt@redhat.com> Reviewed-on: http://review.gluster.org/4660 Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Shishir Gowda <sgowda@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* glusterd: Improve error logging when a brick from an old volume gets re-usedNiels de Vos2013-03-221-2/+7
| | | | | | | | | | | | | | | | | | | The error message when creating a volume that contains a brick with certain xatts set on a parent directory is unclear. Users do not understand '... or a prefix of it is already part of a volume'. Most users check the final directory that is used for a brick, but not its parents. It would be helpful to present the user with the actual directory that is preventing the volume to use the brick. BUG: 923917 Change-Id: I815ad32a992eb0e41ee8fca6ee9327400d042c45 Signed-off-by: Niels de Vos <ndevos@redhat.com> Reviewed-on: http://review.gluster.org/4701 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* nfs: ACCESS - reply only what was asked forAnand Avati2013-03-223-8/+12
| | | | | | | | | | | | | Set only those bits which were requested by the client. Some clients, like AIX, do not like the fact that we are returning the EXEC bit set in the ACCESS reply even though it only asked for LOOKUP bit. Change-Id: I3c2fd5dce030ea5ddae0511497cafa078c4d76d6 BUG: 924481 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4707 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* dht: make DHT xattr names configurableJeff Darcy2013-03-2112-125/+198
| | | | | | | | | | | | | | | | | | | | This is necessary to support "DHT over DHT" configurations, so that the upper and lower instances of DHT don't step all over each other. Why would we even consider such a thing? Because it gives us the ability to do data tiering and rack-aware placement, either by themselves or as complements to other functionality such as erasure codes or deduplication which save space but cost performance. By setting up the top-level DHT to place data into one of several lower-level DHT pools based on policy instead of pure elastic hashing, we get better performance for 90% of accesses and better storage efficiency for 90% of data, all for relatively low effort. Change-Id: I72e65c29edfc80babf39f7a2a00090f4588c4070 BUG: 924265 Signed-off-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-on: http://review.gluster.org/4694 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* features/marker: log error when unlinking timestamp fileVenky Shankar2013-03-211-4/+13
| | | | | | | | | | | | | | | | ... so it's easy to figure out errno caused it. As of now it's only due to ENOSPC. Logging is done in the error handling routine, so any further changes that require unlinking of the timestamp file due to some error condition(s) are logged. Change-Id: Ia59338e2e32b2adbbd1d56aa260018270f1abae9 BUG: 853911 Signed-off-by: Venky Shankar <vshankar@redhat.com> Reviewed-on: http://review.gluster.org/4649 Reviewed-by: Csaba Henk <csaba@redhat.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* performance/io-threads: Fix range-check for least-rate-limitPranith Kumar K2013-03-211-0/+1
| | | | | | | | | | | | | The issue could be fixed with .validate=GF_OPT_VALIDATE_MIN. But adding max value is more robust. Change-Id: Ia69c6f86855dbd34a26e20391e77bfa0f796a200 BUG: 923573 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4698 Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* nfs: handle stable write with @flags rather than fsync()Anand Avati2013-03-202-68/+27
| | | | | | | | | | | | | | | stable writes can be "made stable" by simply setting O_SYNC (or O_DSYNC, accordingly) in the write flags or fd->flags. Performing fsync() at the end of the write is extremely inefficient and completely messes up eager-locking logic in AFR. Change-Id: I4d954c133641e246b2ab4df874bad0282667561f BUG: 916372 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.org/4591 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* nfs, afr: Fail lookup only on split-brainPranith Kumar K2013-03-202-3/+7
| | | | | | | | | | | Change-Id: Icee9772f1f1bf5336eb82a4dc13e198424cd4a65 BUG: 921996 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4699 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* dht: fix a typoJulesWang2013-03-181-1/+1
| | | | | | | | | Change-Id: Id6f156957e58aad06bf2602f880c7e4102b80fd1 BUG: 764890 Signed-off-by: JulesWang <w.jq0722@gmail.com> Reviewed-on: http://review.gluster.org/4679 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* posix-acl: disable permission checks for fd based opsshishir gowda2013-03-141-4/+4
| | | | | | | | | Signed-off-by: shishir gowda <sgowda@redhat.com> Change-Id: I9d49537c2c7b51d5598b80627d61f060aaec8549 BUG: 921437 Reviewed-on: http://review.gluster.org/4671 Reviewed-by: Vijay Bellur <vbellur@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* geo-rep: retire old style ssh setupCsaba Henk2013-03-143-1/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | Users are still using geo-rep with the old, deprecated, insecure, unsupported ssh setup. Not their fault -- the implementation of the new method had the following charasteristics: - old method is possible, but with default settings it's not working - it can be made operational by fiddling with "remote-gsyncd" tunable - with default setting, an unhelpful, actually misleading error message is produced - the UI gave no hint to the changes in the ssh setup http://review.gluster.org/4392 tried to fix these; what it accomplished was unrestricted support to the bad practice (by making the default old setup operational). From this on: - we disable the old method by reserving the "remote-gsyncd" tunable - if the old method is attempted, give a hint what to do Change-Id: Icade94725d8d8d2d4c89cab992d4226351637b86 BUG: 895656 Signed-off-by: Csaba Henk <csaba@redhat.com> Reviewed-on: http://review.gluster.org/4602 Reviewed-by: Venky Shankar <vshankar@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* Storage/posix: Don't log at ERROR level for failed getxattr.Raghavendra Talur2013-03-121-0/+4
| | | | | | | | | | | | | | | | | | Problem: ENOATTR returned by getxattr -n <NotAnExistingAttribute> <file> was being logged at ERROR level. Solution: Moved logging to DEBUG level. Change-Id: I982a577a4c231faa958ea71abdb272f8d5ffd70c BUG: 918052 Signed-off-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-on: http://review.gluster.org/4628 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Preserve mtime in self-healPranith Kumar K2013-03-121-14/+25
| | | | | | | | | | | | | | | | | | | | | | | Problem: Data self-heal may choose sink iatt to set mtimes. This happens because after syncing of data is done self-heal does one more xattrops/fstat to determine sources sinks to set the inode-ctx. Since this is done after data syncing and erase of xattrs, old source and old sink are now sources, but the mtimes of them differ. Old code just takes the first source from the list and update mtimes, which could be sink before the self-heal started. Fix: Set mtime from 'sources before syncing'. Change-Id: Id769e1b99aa4f041eaee775f64cbf2c57b799723 BUG: 918437 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: http://review.gluster.org/4658 Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* glusterd: Mark vol as deleted by renaming voldir before cleaning up the storeKrutika Dhananjay2013-03-114-44/+78
| | | | | | | | | | | | | | | | | | | | | | | | | | PROBLEM: During 'volume delete', when glusterd fails to erase all information about a volume from the backend store (for instance because rmdir() failed on non-empty directories), not only does volume delete fail on that node, but also subsequent attempts to restart glusterd fail because the volume store is left in an inconsistent state. FIX: Rename the volume directory path to a new location <working-dir>/trash/<volume-id>.deleted, and then go on to clean up its contents. The volume is considered deleted once rename() succeeds, irrespective of whether the cleanup succeeds or not. Change-Id: Iaf18e1684f0b101808bd5e1cd53a5d55790541a8 BUG: 889630 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.org/4639 Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Kaushal M <kaushal@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* glusterd: Fixed volume-sync in synctask codepath.Krishnan Parthasarathi2013-03-101-9/+13
| | | | | | | | | | Change-Id: I2911d3ac80825310f84c5ba6bd7890e65e1ee219 BUG: 865700 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.org/4624 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* glusterd: fix segfault on volume status detailLars Ellenberg2013-03-081-9/+11
| | | | | | | | | | | | | | | | | | | | | | | If for some reason glusterd_get_brick_root() fails, it frees the gf_strdup'ed *mount_point in its own error path, and returns -1. Unfortunately it already had assigned that pointer value to the output argument, the caller function glusterd_add_brick_detail() sees a non-NULL pointer, and free() again: segfault. Could be fixed with a one-liner (*mount_point = NULL) in the error path, but I think glusterd_get_brick_root() should only assign to the output argument once all checks passed, so I use a local temporary pointer, which increases the patch a bit. Change-Id: I3f3035f01e80a5e9bdf2da895e4cf7baa3dfbd2f BUG: 919352 Signed-off-by: Lars Ellenberg <lars@linbit.com> Reviewed-on: http://review.gluster.org/4646 Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>