[Gluster-users] can't mount 3.7.9 with 3.6.0 fuse

[Dmitry Melekhov]

29.03.2016 17:48, Atin Mukherjee пишет:

Hello!

Thank you!

[root at test ~]# LC_NUMERIC="en_US.UTF-8"  mount.glusterfs cl1:/rep-volume 
replica/
[root at test ~]#

works for me.

It's strange that RH still didn't update glusterfs version in RHEL 6 yet 
though ... :-(


>
> On 03/29/2016 07:07 PM, Atin Mukherjee wrote:
>>
>> On 03/29/2016 06:24 PM, Dmitry Melekhov wrote:
>>> 29.03.2016 16:46, Atin Mukherjee пишет:
>>>
>>> Hello!
>>>
>>>
>>>> On 03/29/2016 05:18 PM, Dmitry Melekhov wrote:
>>>>>
>>>>> Hello!
>>>>>
>>>>> I installed 3.7.9 on two centos 7 hosts, and I can mount it with
>>>>> mount.glusterfs ,
>>>>> I can do this on RHEL 6.7 host too, this host has old version:
>>>>>
>>>>> [root at roxar ~]#  mount.glusterfs cl2:/rep-volume replica/
>>>>> [root at roxar ~]#
>>>>> [root at roxar ~]# which mount.glusterfs
>>>>> /sbin/mount.glusterfs
>>>>> [root at roxar ~]# rpm -qf /sbin/mount.glusterfs
>>>>> glusterfs-fuse-3.5.2-1.el6.x86_64
>>>>>
>>>>> because it was installed from some 3-rd party repository long time ago
>>>>> and never used in production.
>>>>>
>>>>> But I can't mount using version supplied with Centos 6.7:
>>>>>
>>>>> [root at test ~]# mount.glusterfs cl2:/rep-volume replica/
>>>>> Mount failed. Please check the log file for more details.
>>>>> [root at test ~]#
>>>>>
>>>>>
>>>>> [root at test ~]# rpm -qf /sbin/mount.glusterfs
>>>>> glusterfs-fuse-3.6.0.55-1.el6.x86_64
>>>>>
>>>>> Could you tell me what can cause this?
>>>> Could you attach the glusterd & mount log?
>>> This is from one host with centos 6.7:
>>> [root at makosh ~]# cat /var/log/glusterfs/mnt-replica.log
>>> [2016-03-29 10:44:54.096391] I [MSGID: 100030] [glusterfsd.c:2019:main]
>>> 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version
>>> 3.6.0.55 (args: /usr/sbin/glusterfs --volfile-server=cl2
>>> --volfile-id=/rep-volume /mnt/replica)
>>> [2016-03-29 10:44:54.096967] I
>>> [options.c:1201:xlator_option_init_double] 0-fuse: option
>>> attribute-timeout convertion failed value 1.0
>>> [2016-03-29 10:44:54.096990] E [xlator.c:406:xlator_init] 0-fuse:
>>> Initialization of volume 'fuse' failed, review your volfile again
>>  From the logs its clear that attribute-timeout validation failed as it
>> expects an integer instead of a float.
>>
>> Raghavendra may shed some more light into it.
> I just realized that there is a old thread [1] which talks about a
> similar issue and the work around is also available.
>
> [1]
> https://www.gluster.org/pipermail/gluster-users/2014-December/019911.html
>>> Log from another host is attached.
>>>
>>> There is nothing in glusterd log at this time, may be I need to increase
>>> logging output, but I don't know how :-(
>>> Thank you!
>>>
>>>
>>>
>>>>> Thank you!
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Gluster-users mailing list
>>>>> Gluster-users at gluster.org
>>>>> http://www.gluster.org/mailman/listinfo/gluster-users
>>>>>
>> _______________________________________________
>> Gluster-users mailing list
>> Gluster-users at gluster.org
>> http://www.gluster.org/mailman/listinfo/gluster-users
>>