summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
* gluster rebalance: give option to split the commandAmar Tumballi2011-03-013-65/+65
| | | | | | | | | | | | | | | | | the 'gluster volume rebalance <VOLNAME> start' is enhanced with two more options: * 'gluster volume rebalance <VOLNAME> fix-layout start' (for fixing layout only) * 'gluster volume rebalance <VOLNAME> migrate-data start' (for migrating data only) Also the old way of running rebalance in one shot will still work fine * 'gluster volume rebalance <VOLNAME> start' Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2258 (enhance gluster volume rebalance) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2258
* gluster rebalance: get the proper/exact error msg to cliAmar Tumballi2011-03-013-84/+314
| | | | | | | | | | | introduce a new field in XDR structure, and hence change the version of program. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1922 (Volume not present wrong message displayed on command line) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1922
* CLI : NFS disable option through volume set.Gaurav2011-03-012-0/+23
| | | | | | | | Signed-off-by: Gaurav <gaurav@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2094 (Need option to turn off NFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2094
* CLI : NFS persistent local/global options.Gaurav2011-03-016-162/+597
| | | | | | | | Signed-off-by: Gaurav <gaurav@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2064 (NFS options are removed upon glusterd restart) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2064
* rpcsvc: safe mem_put of req in nfs_rpcsvc_submit_vectorsAnand Avati2011-03-011-1/+4
| | | | | | | | Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2481 (Crash in NFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2481
* rpcsvc: perform mem_put at the very end for safe unrefsAnand Avati2011-03-011-1/+2
| | | | | | | | Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2481 (Crash in NFS) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2481
* NFS : Send error reply on rpc decoding failure.Gaurav2011-03-011-0/+1
| | | | | | | | Signed-off-by: Gaurav <gaurav@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2436 (Not able to send error reply in nfs_rpcsvc) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2436
* glusterd: remove rpc code from internals of glusterdAmar Tumballi2011-03-014-1003/+664
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2333 (make glusterd more rpc friendly) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2333
* glusterd: separate out cli specific programs and mgmt specific programsAmar Tumballi2011-03-015-5/+106
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2333 (make glusterd more rpc friendly) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2333
* glusterd: dependency on 'priv->mgmt' completely removedAmar Tumballi2011-03-015-396/+392
| | | | | | | | | | using 'peerinfo->mgmt' instead. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2333 (make glusterd more rpc friendly) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2333
* glusterd: keep mgmt program peerinfo specificAmar Tumballi2011-03-015-70/+208
| | | | | | | | | | | | | With different version of glusterd in different machines, its not possible to support using just one mgmt program. Instead each peerinfo structure should have its own mgmt programs, so one glusterd can support multiple versions Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2333 (make glusterd more rpc friendly) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2333
* v2 Prevent removal of trusted.gfid xattrshishir gowda2011-03-011-0/+6
| | | | | | | | Signed-off-by: shishir gowda <shishirng@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2461 (remove xattr of trusted.gfid succeeds) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2461
* mgmt/Glusterd: while resetting the volume options, the previous data ↵v3.1.3qa1Kaushik BV2011-03-011-7/+27
| | | | | | | | | | structures w.r.t auth. must be deleted Signed-off-by: Kaushik BV <kaushikbv@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2374 ([glusterfs 3.1.2] gluster volume set auth.reject command failure on nfs) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2374
* check for inode and path being NULL before copying themRaghavendra Bhat2011-02-281-24/+28
| | | | | | | | Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2463 ([7b07d444a77526f27f860210930bf1d4c7fbea9b]: ./autogen.sh crashes glusterfs in fuse_getattr or fuse_setattr) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2463
* mgmt/glusterd: prevent spurious cmd unlocksPranith K2011-02-283-571/+318
| | | | | | | | Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2460 ([glusterfs-3.2.0-gsync1]: gluserd crashed trying to access freed request) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2460
* performance/io-cache: Whitespace cleanupVijay Bellur2011-02-281-205/+201
| | | | | | | | Signed-off-by: Vijay Bellur <vijay@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2346 (Log message enhancements in GlusterFS - phase 1) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346
* cluster/afr: Re-factoring code for alignmentVijay Bellur2011-02-281-62/+62
| | | | | | | | Signed-off-by: Vijay Bellur <vijay@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2346 (Log message enhancements in GlusterFS - phase 1) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346
* cluster/afr: Whitespace cleanupVijay Bellur2011-02-282-124/+126
| | | | | | | | Signed-off-by: Vijay Bellur <vijay@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2346 (Log message enhancements in GlusterFS - phase 1) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346
* dht: log filename on failed reavalidateAnand Avati2011-02-251-2/+3
| | | | | | | | | | The previous log message is pretty much useless without the path Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2346 (Log message enhancements in GlusterFS - phase 1) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2346
* check the op_ret for less than zero in sp_create_cbk since posix can send ↵Raghavendra Bhat2011-02-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | negative values other than -1 posix can send -ve values other than -1 in create: op_ret = setgid_override (this, real_path, &gid); if (op_ret < 0) { goto out; } In stat-prefetch we check op_ret only for -1 which results in function proceeding and op_ret may become 0 due to other function calls in sp_create_cbk, hence crashes in fuse. Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2409 (crash in stat prefetch) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2409
* storage/posix: set op_ret to -1 when setgid_override fails.Raghavendra G2011-02-221-5/+18
| | | | | | | | Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2409 (crash in stat prefetch) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2409
* send the CHILD_DOWN event also to fuseAmar Tumballi2011-02-221-31/+11
| | | | | | | | | | and start the fuse thread in CHILD_DOWN event too. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2005 (Mounting Gluster volume with RO bricks hangs) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2005
* log the total replace brick command receivedRaghavendra Bhat2011-02-221-2/+1
| | | | | | | | | | | Log the full replace brick command which says whether the command is start/pause/status/abort/commit. Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1404 (need a dump of all the op/mgmt commands) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1404
* glusterd/cli: option added to create volume with both transportsAmar Tumballi2011-02-224-6/+80
| | | | | | | | | | | | | | | to avail the option, enter the volume create command with arguments 'transport tcp,rdma' and while mounting, on the mountpoints which works on rdma, do, mount -t glusterfs <IP>:/<VOLNAME>-rdma <MOUNT-POINT> Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2294 (Currently there is no way through cli to make a volume listen on both the transports (socket/rdma)) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2294
* debug/trace: updated the logs to print gfid instead of 'ino' valuesAmar Tumballi2011-02-221-413/+562
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2332 (update 'debug/trace' to log gfid instead of 'ia_ino') URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2332
* glusterd: make it more RPC friendlyAmar Tumballi2011-02-222-196/+282
| | | | | | | | | | | make sure each procedure has a actor associated with it. that way proper versioning can be implemented in modules Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2333 (make glusterd more rpc friendly) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2333
* fuse-resolve: correction in resolve logicAmar Tumballi2011-02-224-91/+286
| | | | | | | | | | | | * bring in soft (gfid based) and hard (path based) resolving * 'fd' resolving to new graph is not yet done. * fuse-resolve works similar to server-resolve Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2281 (I/O operations exit when add-brick is done) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2281
* dht: do not send revalidates to all subvolsAnand Avati2011-02-221-10/+6
| | | | | | | | Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2450 (scalability enhancements) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2450
* core: have framework for GF_EVENT_CHILD_MODIFIED notifcation eventAnand Avati2011-02-222-1/+12
| | | | | | | | | | | | | If anything changes anywhere in the graph, all xlators know that something has changed. Previously dht and afr would mask certain events from propagating up. Now they forward GF_EVENT_CHILD_MODIFIED event in those situations Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2450 (scalability enhancements) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2450
* mgmt/glusterd: In store-retrieve exit with error message instead of crashing.Pranith K2011-02-224-29/+159
| | | | | | | | Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2066 (glusterd crashed while trying to restore volumes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2066
* rpc/rpc-clnt: don't use iobref passed from caller to hold iobufs required ↵Raghavendra G2011-02-222-32/+55
| | | | | | | | | | | | | | | | for building program and rpc headers. - clustering translators like replicate assume iobrefs will be used as read-only by lower layers and there by use the same iobref while winding calls to multiple children. Since iobrefs can hold only limited number of iobufs, this limits the number of children these translators can have without overflowing iobrefs. Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2157 (glusterfsd crash) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2157
* Solaris build fixesshishir gowda2011-02-225-7/+6
| | | | | | | | | | | | | | Disable geosync build for non-linux platforms. And fix solaris related build failures.The symbol getxattr was not available on solaris. The build on linux passed as getxattr symbol was available in the library. The argument was not being used, hence removed it. Make 64 bit build default in CFLAGS option by passing -m64 option Signed-off-by: Shishir Gowda <shishirng@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2440 (Fix solaris build issues seen on latest git) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2440
* replicate: set read-child to the subvolume which replied first (fastest)Anand Avati2011-02-221-20/+0
| | | | | | | | | | | | In the old logic read-child was set to the last subvolume which replied. Now it is updated to the first subvolume. This way we can ensure that local servers are preferred for reads _most_ of the times. Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1172 (ls -lh on NFS mount of 2-mirror replicate gives incorrect file size) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1172
* io-threads: use slow/normal/fast classification of fopsAnand Avati2011-02-221-168/+75
| | | | | | | | Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2241 (GlusterFs Stat Actions Degrade During I/O) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2241
* io-threads: implement bulk and priority queuesAnand Avati2011-02-222-12/+55
| | | | | | | | Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2241 (GlusterFs Stat Actions Degrade During I/O) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2241
* io-threads: whitespace cleanupAnand Avati2011-02-221-12/+9
| | | | | | | | Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2241 (GlusterFs Stat Actions Degrade During I/O) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2241
* performance/quick-read: disable caching for fds opened with GF_OPEN_NOWB flags.Raghavendra G2011-02-181-2/+4
| | | | | | | | Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2240 (Solaris client hangs on file read operations) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2240
* cluster/dht : implement a fsetxattr() commandAmar Tumballi2011-02-182-0/+45
| | | | | | | | | | Thanks to Jeff Darcy <jdarcy(at)redhat.com> for the patch Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2319 (DHT uses wrong subvolume for fsetxattr) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2319
* rpc/xdr: update .x filesAmar Tumballi2011-02-182-2/+2
| | | | | | | | | | * now, 'rpcgen $filename-xdr.x' file will result in $filename-xdr.h Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2153 (glusterfs3.x does not match "generated" source) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2153
* cluster/afr: stop spawning self-heal loops when the self-heal failsPranith K2011-02-171-12/+24
| | | | | | | | | | | | | With the current model of self-heal, any loop wont resume if a self-heal failure happens. I have added the fix to stop spawning the initial loops on self-heal failure. Fixed the invalid read shown in valgrind in diff-self-heal. Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1174 (Replicate spawns read loops even after destination fails) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1174
* Typo fixes in log and cli messages.Sachidananda2011-02-171-1/+1
| | | | | | | | Signed-off-by: Sachidananda Urs <sac@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2321 (Typo in logs and cli messages) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2321
* glusterd-volgen: fix a issue of wrongly closing '0' fd.Amar Tumballi2011-02-151-3/+6
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2421 ([glusterfs-3.1.2gsync10]: epoll_create returning 0) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2421
* volgen: avoid fclose() on cosed FILE *Anand Avati2011-02-151-0/+1
| | | | | | | | | | | | It can so happen that a previously freed up FILE * structure has 0'ed out memory and a second fclose() on it can result in close(0). This results in bad bad things happening. Signed-off-by: Anand Avati <avati@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2421 ([glusterfs-3.1.2gsync10]: epoll_create returning 0) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2421
* syncdaemon: fix RePCe version queryCsaba Henk2011-02-152-8/+11
| | | | | | | | | | | RepceClient.repce_version fetched its own RePCe version, instead of querying peer, so lame. Signed-off-by: Csaba Henk <csaba@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 2381 (RePCe versioning) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2381
* syncdaemon: Checking for the existence of gsyncd by trying to aquire a lock ↵Mohammed Junaid Ahmed2011-02-151-5/+6
| | | | | | | | | | | | | on the pidfile. If the locking succeeds then gsyncd is not running and if it fails then it means that gsyncd is running. Signed-off-by: Junaid <junaid@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1570 (geosync related changes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1570
* syncdaemon: change pidfile handling approachCsaba Henk2011-02-153-28/+48
| | | | | | | | | | | | | | Signal handling of the python interpreter is a bit messy, so we cannot rely on executing a final clause upon termination. Switch over fcntl lock based pidfile handling which can provide reliable info about status of the process. (Due to aforementioned reason, pidfile as such is not guaranteed to be cleaned up, but lock acquisition is a reliable measure.) Signed-off-by: Csaba Henk <csaba@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1570 (geosync related changes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1570
* syncdaemon: Check for ctx context in the glusterd_op_stage_gsyc_set().Mohammed Junaid Ahmed2011-02-151-0/+8
| | | | | | | | Signed-off-by: Junaid <junaid@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1570 (geosync related changes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1570
* syncdaemon: set the default value of timeout to 120secKaushik BV2011-02-151-1/+1
| | | | | | | | Signed-off-by: Kaushik BV <kaushikbv@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1570 (geosync related changes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1570
* syncdaemon: Moved the gsync start code to cli.Mohammed Junaid Ahmed2011-02-152-141/+93
| | | | | | | | Signed-off-by: Junaid <junaid@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1570 (geosync related changes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1570
* syncdaemon: fix some pidfile issuesCsaba Henk2011-02-141-17/+18
| | | | | | | | Signed-off-by: Csaba Henk <csaba@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 1570 (geosync related changes) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1570
generated by cgit (git 2.34.1) at 2024-04-20 02:52:21 +0000