Joe Julian
2015-Mar-27 14:51 UTC
[Gluster-users] Can not mount volumes after upgrade to Gluster 3.6.6
We hit this in IRC recently. Somewhere during the upgrade process your .vol files were deleted (or renamed). Recreate them by stopping glusterd then running: glusterd --xlator-option "*.upgrade=on" -N On March 27, 2015 6:02:47 AM PDT, Kal Black <kaloblak at gmail.com> wrote:>Hello, >After upgrading from Glusterfs 3.5 to Glusterfs 3.6.2, on Ubuntu 12.04, >I >am unable to mount gluster volumes any more. Found the same issue >reported >already and the answer suggested that a path to the volume is being >used, >during the mount attempt, instead of the volume name. This however >seems >not to be the case according to what I am seeing. Any ideas? > ># gluster volume info >Volume Name: dbs >Type: Replicate >Volume ID: 84984920-2dde-4b40-8041-5d6d514adf37 >Status: Started >Number of Bricks: 1 x 2 = 2 >Transport-type: tcp >Bricks: >Brick1: 10.0.1.103:/data/exports/dbs_content >Brick2: 10.0.1.104:/data/exports/dbs_content > >When I try to mount I got the following: > ># mount -t glusterfs 10.0.1.103:/dbs /dbs_content >WARNING: getfattr not found, certain checks will be skipped.. >Mount failed. Please check the log file for more details. > >In the log I am getting: > >[2015-03-26 20:34:31.582757] I [MSGID: 100030] [glusterfsd.c:2018:main] >0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version >3.6.2 >(args: /usr/sbin/glusterfs --volfile-server=10.0.1.103 >--volfile-id=/dbs >/dbs_content) >[2015-03-26 20:34:31.587971] E >[glusterfsd-mgmt.c:1494:mgmt_getspec_cbk] >0-glusterfs: failed to get the 'volume file' from server >[2015-03-26 20:34:31.588053] E >[glusterfsd-mgmt.c:1596:mgmt_getspec_cbk] >0-mgmt: failed to fetch volume file (key:/dbs) >[2015-03-26 20:34:31.588330] W [glusterfsd.c:1194:cleanup_and_exit] >(--> >0-: received signum (0), shutting down >[2015-03-26 20:34:31.588373] I [fuse-bridge.c:5599:fini] 0-fuse: >Unmounting >'/dbs_content'. > > >Best, > > >------------------------------------------------------------------------ > >_______________________________________________ >Gluster-users mailing list >Gluster-users at gluster.org >http://www.gluster.org/mailman/listinfo/gluster-users-- Sent from my Android device with K-9 Mail. Please excuse my brevity. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150327/ae60c707/attachment.html>
Kal Black
2015-Mar-27 18:56 UTC
[Gluster-users] Can not mount volumes after upgrade to Gluster 3.6.6
Hello Joe, Your solution fixed the issue. However the .vol files were not missing but instead have been regenerated (or modified) after running the glusterd --xlator-option "*.upgrade=on" -N Unfortunately, I didn't do a backup of the vol files before running the command so I can't see how those files actually changed, but I did a "ls" on the /var/lib/glusterd/vols/dbs before and after the execution and can see that the .vol files changed time stamp and size. I would of like to know what exactly happened and if running this command is necessary after an upgrade, probably it is a good idea to let know glusterfs guys to included it in the "GlusterFS upgrade from 3.5.x to 3.6.X" documentation. Thanks a lot! # ls -l /var/lib/glusterd/vols/dbs/ total 44 drwxr-xr-x 2 root root 4096 Oct 27 15:10 bricks -rw------- 1 root root 1796 Oct 27 15:10 dbs.10.0.1.103.data-exports-dbs_content.vol -rw------- 1 root root 1796 Oct 27 15:10 dbs.10.0.1.104.data-exports-dbs_content.vol -rw------- 1 root root 1298 Oct 27 15:10 dbs-fuse.vol -rw------- 1 root root 16 Mar 27 14:25 cksum -rw------- 1 root root 356 Oct 27 15:10 info -rw------- 1 root root 84 Oct 27 15:10 node_state.info -rw------- 1 root root 18 Mar 27 14:25 quota.cksum -rw------- 1 root root 0 Mar 26 16:27 quota.conf -rw------- 1 root root 12 Oct 27 15:10 rbstate drwxr-xr-x 2 root root 4096 Mar 27 14:27 run -rw------- 1 root root 1526 Oct 27 15:10 trusted-dbs-fuse.vol # glusterd --xlator-option "*.upgrade=on" -N librdmacm: couldn't read ABI version. librdmacm: assuming: 4 CMA: unable to get RDMA device list # ls -l /var/lib/glusterd/vols/dbs/ total 56 drwxr-xr-x 2 root root 4096 Oct 27 15:10 bricks -rw------- 1 root root 1979 Mar 27 14:27 dbs.10.0.1.103.data-exports-dbs_content.vol -rw------- 1 root root 1979 Mar 27 14:27 dbs.10.0.1.104.data-exports-dbs_content.vol -rw------- 1 root root 1298 Oct 27 15:10 dbs-fuse.vol -rw------- 1 root root 1352 Mar 27 14:27 dbs-rebalance.vol -rw------- 1 root root 1352 Mar 27 14:27 dbs.tcp-fuse.vol -rw------- 1 root root 16 Mar 27 14:27 cksum -rw------- 1 root root 356 Oct 27 15:10 info -rw------- 1 root root 84 Oct 27 15:10 node_state.info -rw------- 1 root root 18 Mar 27 14:27 quota.cksum -rw------- 1 root root 0 Mar 26 16:27 quota.conf -rw------- 1 root root 12 Oct 27 15:10 rbstate drwxr-xr-x 2 root root 4096 Mar 27 14:27 run -rw------- 1 root root 1526 Oct 27 15:10 trusted-dbs-fuse.vol -rw------- 1 root root 1580 Mar 27 14:27 trusted-dbs.tcp-fuse.vol On Fri, Mar 27, 2015 at 10:51 AM, Joe Julian <joe at julianfamily.org> wrote:> We hit this in IRC recently. > > Somewhere during the upgrade process your .vol files were deleted (or > renamed). Recreate them by stopping glusterd then running: > > glusterd --xlator-option "*.upgrade=on" -N > > > On March 27, 2015 6:02:47 AM PDT, Kal Black <kaloblak at gmail.com> wrote: > >> Hello, >> After upgrading from Glusterfs 3.5 to Glusterfs 3.6.2, on Ubuntu 12.04, I >> am unable to mount gluster volumes any more. Found the same issue reported >> already and the answer suggested that a path to the volume is being used, >> during the mount attempt, instead of the volume name. This however seems >> not to be the case according to what I am seeing. Any ideas? >> >> # gluster volume info >> Volume Name: dbs >> Type: Replicate >> Volume ID: 84984920-2dde-4b40-8041-5d6d514adf37 >> Status: Started >> Number of Bricks: 1 x 2 = 2 >> Transport-type: tcp >> Bricks: >> Brick1: 10.0.1.103:/data/exports/dbs_content >> Brick2: 10.0.1.104:/data/exports/dbs_content >> >> When I try to mount I got the following: >> >> # mount -t glusterfs 10.0.1.103:/dbs /dbs_content >> WARNING: getfattr not found, certain checks will be skipped.. >> Mount failed. Please check the log file for more details. >> >> In the log I am getting: >> >> [2015-03-26 20:34:31.582757] I [MSGID: 100030] [glusterfsd.c:2018:main] >> 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.6.2 >> (args: /usr/sbin/glusterfs --volfile-server=10.0.1.103 --volfile-id=/dbs >> /dbs_content) >> [2015-03-26 20:34:31.587971] E [glusterfsd-mgmt.c:1494:mgmt_getspec_cbk] >> 0-glusterfs: failed to get the 'volume file' from server >> [2015-03-26 20:34:31.588053] E [glusterfsd-mgmt.c:1596:mgmt_getspec_cbk] >> 0-mgmt: failed to fetch volume file (key:/dbs) >> [2015-03-26 20:34:31.588330] W [glusterfsd.c:1194:cleanup_and_exit] (--> >> 0-: received signum (0), shutting down >> [2015-03-26 20:34:31.588373] I [fuse-bridge.c:5599:fini] 0-fuse: >> Unmounting '/dbs_content'. >> >> >> Best, >> >> ------------------------------ >> >> Gluster-users mailing list >> Gluster-users at gluster.org >> http://www.gluster.org/mailman/listinfo/gluster-users >> >> > -- > Sent from my Android device with K-9 Mail. Please excuse my brevity. >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150327/77404294/attachment.html>