[Gluster-users] Getting glusterd to shutdown cleanly

[Kaleb S. KEITHLEY]

On 12/02/2013 11:16 AM, Ellison, Bob wrote:
> I guess there’s not much interest in shutting down glusterfs cleanly L
>
> As I can get the stock Redhat6 nfs to shutdown cleanly, is there any
> compelling reason to use gluster nfs3 as opposed to the stock redhat6
> version of nfs?

The stock redhat6 kernel nfs server (knfsd) isn't gluster aware.

The correct way to stop a gluster volume, including its associated nfs 
server, is with the gluster cli, e.g. `gluster volume stop $volname`.

The glusterfsd init.d script is only intended to be used for system 
shutdown, e.g. `init 0` or `init 6`.


>
> Thanks,
>
> Bob
>
> *From:*gluster-users-bounces at gluster.org
> [mailto:gluster-users-bounces at gluster.org] *On Behalf Of *Ellison, Bob
> *Sent:* Wednesday, November 27, 2013 3:40 PM
> *To:* gluster-users at gluster.org
> *Subject:* [Gluster-users] Getting glusterd to shutdown cleanly
>
> Hello,
>
> I’m trying to get glusterd to shutdown cleanly on a server. I’m running
> 3.4.1-3.
>
> I fixed an issue in /etc/init.d/glusterd with RETVAL in the stop()
> function (apparently also fixed upstream).
>
> I now have one remaining issue: the nfs server does not shut down:
>
> # Running system
>
> [root at ads2 ~]# ps -e | grep gluster
>
> 5994 ?        00:00:00 glusterd
>
> 6008 ?        00:00:00 glusterfsd
>
> 6012 ?        00:00:00 glusterfsd
>
> 6016 ?        00:00:00 glusterfsd
>
> 6644 ?        00:00:00 glusterfs
>
> 17710 ?        00:00:09 glusterfs
>
> # umount the lone gluster fuse mount point
>
> [root at ads2 ~]# umount /content
>
> [root at ads2 ~]# ps -e | grep gluster
>
> 5994 ?        00:00:00 glusterd
>
> 6008 ?        00:00:00 glusterfsd
>
> 6012 ?        00:00:00 glusterfsd
>
> 6016 ?        00:00:00 glusterfsd
>
> 6644 ?        00:00:00 glusterfs
>
> [root at ads2 ~]# service glusterfsd stop
>
> Stopping glusterfsd: [  OK  ]
>
> [root at ads2 ~]# ps -e | grep gluster
>
> 5994 ?        00:00:00 glusterd
>
> 6644 ?        00:00:00 glusterfs
>
> [root at ads2 ~]# service glusterd stop
>
> [root at ads2 ~]# rd:[  OK  ]
>
> [root at ads2 ~]# ps -e | grep gluster
>
> 6644 ?        00:00:00 glusterfs
>
> [root at ads2 ~]# ps -ef | grep gluster
>
> root      6644     1  0 14:51 ?        00:00:00 /usr/sbin/glusterfs -s
> localhost --volfile-id gluster/nfs -p /var/lib/glusterd/nfs/run/nfs.pid
> -l /var/log/glusterfs/nfs.log -S
> /var/run/7c1d38f7dcadc6442a09210bf552c5dc.socket
>
> Is this by design?  Any suggestions?
>
> Thanks,
>
> Bob
>
>
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://supercolony.gluster.org/mailman/listinfo/gluster-users
>

-- 

Kaleb