<div dir="ltr"><div>Kudos to the folks behind this release !<br><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, Jun 24, 2014 at 4:20 PM, Niels de Vos <span dir="ltr"><<a href="mailto:ndevos@redhat.com" target="_blank">ndevos@redhat.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On Tue, Jun 24, 2014 at 03:15:58AM -0700, Gluster Build System wrote:<br>
><br>
><br>
> SRC: <a href="http://bits.gluster.org/pub/gluster/glusterfs/src/glusterfs-3.5.1.tar.gz" target="_blank">http://bits.gluster.org/pub/gluster/glusterfs/src/glusterfs-3.5.1.tar.gz</a><br>
><br>
> This release is made off jenkins-release-73<br>
<br>
</div>Many thanks to everyone how tested the glusterfs-3.5.1 beta releases and<br>
gave feedback. There were no regressions reported compared to the 3.5.0<br>
release.<br>
<br>
Many bugs have been fixed, and documentation for all new features in 3.5<br>
should be included now. Thanks to all the reporters, developers and<br>
testers for improving the 3.5 stable series.<br>
<br>
Below you will find the release notes in MarkDown format for<br>
glusterfs-3.5.1, these are included in the tar.gz as<br>
doc/release-notes/<a href="http://3.5.1.md" target="_blank">3.5.1.md</a>. The mirror repository on GitHub provides<br>
a nicely rendered version:<br>
- <a href="https://github.com/gluster/glusterfs/blob/v3.5.1/doc/release-notes/3.5.1.md" target="_blank">https://github.com/gluster/glusterfs/blob/v3.5.1/doc/release-notes/3.5.1.md</a><br>
<br>
Packages for different Linux distributions will follow shortly.<br>
Notifications are normally sent to this list when the packages are<br>
available for download, and/or have reached the distributions update<br>
infrastructure.<br>
<br>
Changes for a new 3.5.2 release are now being accepted. The list of<br>
proposed fixes is already growing:<br>
- <a href="https://bugzilla.redhat.com/showdependencytree.cgi?hide_resolved=0&id=glusterfs-3.5.2" target="_blank">https://bugzilla.redhat.com/showdependencytree.cgi?hide_resolved=0&id=glusterfs-3.5.2</a><br>
<br>
Anyone is free to request a bugfix or backport for the 3.5.2 release. In<br>
order to do so, file a bug and set the 'blocked' field to<br>
'glusterfs-3.5.2' so that we can track the requests. Use this link to<br>
make it a little easier for yourself:<br>
- <a href="https://bugzilla.redhat.com/enter_bug.cgi?product=GlusterFS&version=3.5.1&blocked=glusterfs-3.5.2" target="_blank">https://bugzilla.redhat.com/enter_bug.cgi?product=GlusterFS&version=3.5.1&blocked=glusterfs-3.5.2</a><br>
<br>
Cheers,<br>
Niels<br>
<br>
<br>
<br>
## Release Notes for GlusterFS 3.5.1<br>
<br>
This is mostly a bugfix release. The [Release Notes for 3.5.0](<a href="http://3.5.0.md" target="_blank">3.5.0.md</a>)<br>
contain a listing of all the new features that were added.<br>
<br>
There are two notable changes that are not only bug fixes, or documentation<br>
additions:<br>
<br>
1. a new volume option `server.manage-gids` has been added<br>
This option should be used when users of a volume are in more than<br>
approximately 93 groups (Bug [1096425](<a href="https://bugzilla.redhat.com/1096425" target="_blank">https://bugzilla.redhat.com/1096425</a>))<br>
2. Duplicate Request Cache for NFS has now been disabled by default, this may<br>
reduce performance for certain workloads, but improves the overall stability<br>
and memory footprint for most users<br>
<br>
### Bugs Fixed:<br>
<br>
* [765202](<a href="https://bugzilla.redhat.com/765202" target="_blank">https://bugzilla.redhat.com/765202</a>): lgetxattr called with invalid keys on the bricks<br>
* [833586](<a href="https://bugzilla.redhat.com/833586" target="_blank">https://bugzilla.redhat.com/833586</a>): inodelk hang from marker_rename_release_newp_lock<br>
* [859581](<a href="https://bugzilla.redhat.com/859581" target="_blank">https://bugzilla.redhat.com/859581</a>): self-heal process can sometimes create directories instead of symlinks for the root gfid file in .glusterfs<br>
* [986429](<a href="https://bugzilla.redhat.com/986429" target="_blank">https://bugzilla.redhat.com/986429</a>): Backupvolfile server option should work internal to GlusterFS framework<br>
* [1039544](<a href="https://bugzilla.redhat.com/1039544" target="_blank">https://bugzilla.redhat.com/1039544</a>): [FEAT] "gluster volume heal info" should list the entries that actually required to be healed.<br>
* [1046624](<a href="https://bugzilla.redhat.com/1046624" target="_blank">https://bugzilla.redhat.com/1046624</a>): Unable to heal symbolic Links<br>
* [1046853](<a href="https://bugzilla.redhat.com/1046853" target="_blank">https://bugzilla.redhat.com/1046853</a>): AFR : For every file self-heal there are warning messages reported in glustershd.log file<br>
* [1063190](<a href="https://bugzilla.redhat.com/1063190" target="_blank">https://bugzilla.redhat.com/1063190</a>): Volume was not accessible after server side quorum was met<br>
* [1064096](<a href="https://bugzilla.redhat.com/1064096" target="_blank">https://bugzilla.redhat.com/1064096</a>): The old Python Translator code (not Glupy) should be removed<br>
* [1066996](<a href="https://bugzilla.redhat.com/1066996" target="_blank">https://bugzilla.redhat.com/1066996</a>): Using sanlock on a gluster mount with replica 3 (quorum-type auto) leads to a split-brain<br>
* [1071191](<a href="https://bugzilla.redhat.com/1071191" target="_blank">https://bugzilla.redhat.com/1071191</a>): [3.5.1] Sporadic SIGBUS with mmap() on a sparse file created with open(), seek(), write()<br>
* [1078061](<a href="https://bugzilla.redhat.com/1078061" target="_blank">https://bugzilla.redhat.com/1078061</a>): Need ability to heal mismatching user extended attributes without any changelogs<br>
* [1078365](<a href="https://bugzilla.redhat.com/1078365" target="_blank">https://bugzilla.redhat.com/1078365</a>): New xlators are linked as versioned .so files, creating <xlator>.so.0.0.0<br>
* [1086743](<a href="https://bugzilla.redhat.com/1086743" target="_blank">https://bugzilla.redhat.com/1086743</a>): Add documentation for the Feature: RDMA-connection manager (RDMA-CM)<br>
* [1086748](<a href="https://bugzilla.redhat.com/1086748" target="_blank">https://bugzilla.redhat.com/1086748</a>): Add documentation for the Feature: AFR CLI enhancements<br>
* [1086749](<a href="https://bugzilla.redhat.com/1086749" target="_blank">https://bugzilla.redhat.com/1086749</a>): Add documentation for the Feature: Exposing Volume Capabilities<br>
* [1086750](<a href="https://bugzilla.redhat.com/1086750" target="_blank">https://bugzilla.redhat.com/1086750</a>): Add documentation for the Feature: File Snapshots in GlusterFS<br>
* [1086751](<a href="https://bugzilla.redhat.com/1086751" target="_blank">https://bugzilla.redhat.com/1086751</a>): Add documentation for the Feature: gfid-access<br>
* [1086752](<a href="https://bugzilla.redhat.com/1086752" target="_blank">https://bugzilla.redhat.com/1086752</a>): Add documentation for the Feature: On-Wire Compression/Decompression<br>
* [1086754](<a href="https://bugzilla.redhat.com/1086754" target="_blank">https://bugzilla.redhat.com/1086754</a>): Add documentation for the Feature: Quota Scalability<br>
* [1086755](<a href="https://bugzilla.redhat.com/1086755" target="_blank">https://bugzilla.redhat.com/1086755</a>): Add documentation for the Feature: readdir-ahead<br>
* [1086756](<a href="https://bugzilla.redhat.com/1086756" target="_blank">https://bugzilla.redhat.com/1086756</a>): Add documentation for the Feature: zerofill API for GlusterFS<br>
* [1086758](<a href="https://bugzilla.redhat.com/1086758" target="_blank">https://bugzilla.redhat.com/1086758</a>): Add documentation for the Feature: Changelog based parallel geo-replication<br>
* [1086760](<a href="https://bugzilla.redhat.com/1086760" target="_blank">https://bugzilla.redhat.com/1086760</a>): Add documentation for the Feature: Write Once Read Many (WORM) volume<br>
* [1086762](<a href="https://bugzilla.redhat.com/1086762" target="_blank">https://bugzilla.redhat.com/1086762</a>): Add documentation for the Feature: BD Xlator - Block Device translator<br>
* [1086766](<a href="https://bugzilla.redhat.com/1086766" target="_blank">https://bugzilla.redhat.com/1086766</a>): Add documentation for the Feature: Libgfapi<br>
* [1086774](<a href="https://bugzilla.redhat.com/1086774" target="_blank">https://bugzilla.redhat.com/1086774</a>): Add documentation for the Feature: Access Control List - Version 3 support for Gluster NFS<br>
* [1086781](<a href="https://bugzilla.redhat.com/1086781" target="_blank">https://bugzilla.redhat.com/1086781</a>): Add documentation for the Feature: Eager locking<br>
* [1086782](<a href="https://bugzilla.redhat.com/1086782" target="_blank">https://bugzilla.redhat.com/1086782</a>): Add documentation for the Feature: glusterfs and oVirt integration<br>
* [1086783](<a href="https://bugzilla.redhat.com/1086783" target="_blank">https://bugzilla.redhat.com/1086783</a>): Add documentation for the Feature: qemu 1.3 - libgfapi integration<br>
* [1088848](<a href="https://bugzilla.redhat.com/1088848" target="_blank">https://bugzilla.redhat.com/1088848</a>): Spelling errors in rpc/rpc-transport/rdma/src/rdma.c<br>
* [1089054](<a href="https://bugzilla.redhat.com/1089054" target="_blank">https://bugzilla.redhat.com/1089054</a>): gf-error-codes.h is missing from source tarball<br>
* [1089470](<a href="https://bugzilla.redhat.com/1089470" target="_blank">https://bugzilla.redhat.com/1089470</a>): SMB: Crash on brick process during compile kernel.<br>
* [1089934](<a href="https://bugzilla.redhat.com/1089934" target="_blank">https://bugzilla.redhat.com/1089934</a>): list dir with more than N files results in Input/output error<br>
* [1091340](<a href="https://bugzilla.redhat.com/1091340" target="_blank">https://bugzilla.redhat.com/1091340</a>): Doc: Add glfs_fini known issue to release notes 3.5<br>
* [1091392](<a href="https://bugzilla.redhat.com/1091392" target="_blank">https://bugzilla.redhat.com/1091392</a>): <a href="http://glusterfs.spec.in" target="_blank">glusterfs.spec.in</a>: minor/nit changes to sync with Fedora spec<br>
* [1095256](<a href="https://bugzilla.redhat.com/1095256" target="_blank">https://bugzilla.redhat.com/1095256</a>): Excessive logging from self-heal daemon, and bricks<br>
* [1095595](<a href="https://bugzilla.redhat.com/1095595" target="_blank">https://bugzilla.redhat.com/1095595</a>): Stick to IANA standard while allocating brick ports<br>
* [1095775](<a href="https://bugzilla.redhat.com/1095775" target="_blank">https://bugzilla.redhat.com/1095775</a>): Add support in libgfapi to fetch volume info from glusterd.<br>
* [1095971](<a href="https://bugzilla.redhat.com/1095971" target="_blank">https://bugzilla.redhat.com/1095971</a>): Stopping/Starting a Gluster volume resets ownership<br>
* [1096040](<a href="https://bugzilla.redhat.com/1096040" target="_blank">https://bugzilla.redhat.com/1096040</a>): AFR : self-heal-daemon not clearing the change-logs of all the sources after self-heal<br>
* [1096425](<a href="https://bugzilla.redhat.com/1096425" target="_blank">https://bugzilla.redhat.com/1096425</a>): i/o error when one user tries to access RHS volume over NFS with 100+ GIDs<br>
* [1099878](<a href="https://bugzilla.redhat.com/1099878" target="_blank">https://bugzilla.redhat.com/1099878</a>): Need support for handle based Ops to fetch/modify extended attributes of a file<br>
* [1101647](<a href="https://bugzilla.redhat.com/1101647" target="_blank">https://bugzilla.redhat.com/1101647</a>): gluster volume heal volname statistics heal-count not giving desired output.<br>
* [1102306](<a href="https://bugzilla.redhat.com/1102306" target="_blank">https://bugzilla.redhat.com/1102306</a>): license: xlators/features/glupy dual license GPLv2 and LGPLv3+<br>
* [1103413](<a href="https://bugzilla.redhat.com/1103413" target="_blank">https://bugzilla.redhat.com/1103413</a>): Failure in gf_log_init reopening stderr<br>
* [1104592](<a href="https://bugzilla.redhat.com/1104592" target="_blank">https://bugzilla.redhat.com/1104592</a>): heal info may give Success instead of transport end point not connected when a brick is down.<br>
* [1104915](<a href="https://bugzilla.redhat.com/1104915" target="_blank">https://bugzilla.redhat.com/1104915</a>): glusterfsd crashes while doing stress tests<br>
* [1104919](<a href="https://bugzilla.redhat.com/1104919" target="_blank">https://bugzilla.redhat.com/1104919</a>): Fix memory leaks in gfid-access xlator.<br>
* [1104959](<a href="https://bugzilla.redhat.com/1104959" target="_blank">https://bugzilla.redhat.com/1104959</a>): Dist-geo-rep : some of the files not accessible on slave after the geo-rep sync from master to slave.<br>
* [1105188](<a href="https://bugzilla.redhat.com/1105188" target="_blank">https://bugzilla.redhat.com/1105188</a>): Two instances each, of brick processes, glusterfs-nfs and quotad seen after glusterd restart<br>
* [1105524](<a href="https://bugzilla.redhat.com/1105524" target="_blank">https://bugzilla.redhat.com/1105524</a>): Disable nfs.drc by default<br>
* [1107937](<a href="https://bugzilla.redhat.com/1107937" target="_blank">https://bugzilla.redhat.com/1107937</a>): quota-anon-fd-nfs.t fails spuriously<br>
* [1109832](<a href="https://bugzilla.redhat.com/1109832" target="_blank">https://bugzilla.redhat.com/1109832</a>): I/O fails for for glusterfs 3.4 AFR clients accessing servers upgraded to glusterfs 3.5<br>
* [1110777](<a href="https://bugzilla.redhat.com/1110777" target="_blank">https://bugzilla.redhat.com/1110777</a>): glusterfsd OOM - using all memory when quota is enabled<br>
<br>
### Known Issues:<br>
<br>
- The following configuration changes are necessary for qemu and samba<br>
integration with libgfapi to work seamlessly:<br>
<br>
1. `gluster volume set <volname> server.allow-insecure on`<br>
2. restarting the volume is necessary<br>
~~~<br>
gluster volume stop <volname><br>
gluster volume start <volname><br>
~~~<br>
3. Edit `/etc/glusterfs/glusterd.vol` to contain this line:<br>
~~~<br>
option rpc-auth-allow-insecure on<br>
~~~<br>
4. restarting glusterd is necessary<br>
~~~<br>
service glusterd restart<br>
~~~<br>
<br>
More details are also documented in the Gluster Wiki on the [Libgfapi with qemu libvirt](<a href="http://www.gluster.org/community/documentation/index.php/Libgfapi_with_qemu_libvirt" target="_blank">http://www.gluster.org/community/documentation/index.php/Libgfapi_with_qemu_libvirt</a>) page.<br>
<br>
- For Block Device translator based volumes open-behind translator at the client side needs to be disabled.<br>
<br>
- libgfapi clients calling `glfs_fini` before a successfull `glfs_init` will cause the client to<br>
hang has been [reported by QEMU developers](<a href="https://bugs.launchpad.net/bugs/1308542" target="_blank">https://bugs.launchpad.net/bugs/1308542</a>).<br>
The workaround is NOT to call `glfs_fini` for error cases encountered before a successfull<br>
`glfs_init`. Follow [Bug 1091335](<a href="https://bugzilla.redhat.com/1091335" target="_blank">https://bugzilla.redhat.com/1091335</a>) to get informed when a<br>
release is made available that contains a final fix.<br>
<br>
- After enabling `server.manage-gids`, the volume needs to be stopped and<br>
started again to have the option enabled in the brick processes<br>
~~~<br>
gluster volume stop <volname><br>
gluster volume start <volname><br>
~~~<br>
<br>
_______________________________________________<br>
Gluster-devel mailing list<br>
<a href="mailto:Gluster-devel@gluster.org">Gluster-devel@gluster.org</a><br>
<a href="http://supercolony.gluster.org/mailman/listinfo/gluster-devel" target="_blank">http://supercolony.gluster.org/mailman/listinfo/gluster-devel</a><br>
</blockquote></div><br></div>