[Gluster-users] [Gluster-devel] Gluster 3.6.4 release testing on IB

[Raghavendra Talur]

On Aug 20, 2015 07:27, "Raghavendra Talur" <rtalur at redhat.com> wrote:
>
>
> On Aug 20, 2015 06:54, "satish kondapalli" <nitw.satish at gmail.com> wrote:
> >
> > Please ignore the previous log. I am seeing that error because i am
trying to mount tcp volumes on RDMA.
> >
> > Here is the error message i am seeing while mounting the volume over
RDMA:
> >
> > [2015-08-20 01:19:48.032815] I [MSGID: 100030] [glusterfsd.c:2018:main]
0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.6.4
(args: /usr/sbin/glusterfs --volfile-server=hp1.info
--volfile-server-transport=rdma --volfile-id=/test-volume2.rdma /mnt_rdma)
> >
> > [2015-08-20 01:19:48.324420] I [dht-shared.c:337:dht_init_regex]
0-test-volume2-dht: using regex rsync-hash-regex = ^\.(.+)\.[^.]+$
> >
> > [2015-08-20 01:19:48.332499] I [client.c:2280:notify]
0-test-volume2-client-0: parent translators are ready, attempting connect
on transport
> >
> > [2015-08-20 01:19:48.341445] I [client.c:2280:notify]
0-test-volume2-client-1: parent translators are ready, attempting connect
on transport
> >
> > [2015-08-20 01:19:48.342494] W [rdma.c:1221:gf_rdma_cm_event_handler]
0-test-volume2-client-0: cma event RDMA_CM_EVENT_ADDR_ERROR, error -19 (me:
peer:)
> >
> >
> > Can some one please take a look.  is my infiniband not setup properly?
> >
> >
> > Sateesh
> >
> >
> > On Wed, Aug 19, 2015 at 6:05 PM, satish kondapalli <
nitw.satish at gmail.com> wrote:
> >>
> >> Hi,
> >>
> >> I am seeing below error when i tried to mount a volume over RDMA. Can
some one explain what the issue is by looking the log?
> >>
> >> [2015-08-20 00:56:38.666653] I [MSGID: 100030]
[glusterfsd.c:2018:main] 0-/usr/sbin/glusterfs: Started running
/usr/sbin/glusterfs version 3.6.4 (args: /usr/sbin/glusterfs
--volfile-server=hp1.info --volfile-server-transport=rdma
--volfile-id=/test-volume1.rdma /mnt_rdma)
> >>
> >> [2015-08-20 00:56:38.689086] E
[glusterfsd-mgmt.c:1494:mgmt_getspec_cbk] 0-glusterfs: failed to get the
'volume file' from server
> >>
> >> [2015-08-20 00:56:38.689188] E
[glusterfsd-mgmt.c:1596:mgmt_getspec_cbk] 0-mgmt: failed to fetch volume
file (key:/test-volume1.rdma)
> >>
> >> [2015-08-20 00:56:38.689487] W [glusterfsd.c:1194:cleanup_and_exit]
(--> 0-: received signum (0), shutting down
> >>
> >> [2015-08-20 00:56:38.689531] I [fuse-bridge.c:5605:fini] 0-fuse:
Unmounting '/mnt_rdma'.
> >>
> >> [2015-08-20 00:56:38.704250] W [glusterfsd.c:1194:cleanup_and_exit]
(--> 0-: received signum (15), shutting down
> >>
> >> [2015-08-20 00:56:38.704250] W [glusterfsd.c:1194:cleanup_and_exit]
(--> 0-: received signum (15), shutting down
>
> 1. How are you performing the mount?
> Using mount command or by passing all parameters to glusterfs?
>
> 2. It is necessary that you configure your infiniband devices to have
IPoIB. If you perform a nslookup on hp1.info from client machine the IP you
get,  is it same as the one you created volume with?

Here is the admin doc for gluster rdma
http://gluster.readthedocs.org/en/latest/Administrator%20Guide/RDMA%20Transport/

>
> >>
> >>
> >> Sateesh
> >
> >
> >
> > _______________________________________________
> > Gluster-devel mailing list
> > Gluster-devel at gluster.org
> > http://www.gluster.org/mailman/listinfo/gluster-devel
> >
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150820/5d014567/attachment.html>