Unfortunately, we have run into the same issue... but we have found that
a small pause before shutting cman down works every time (script snippet
below -- sleep 1 was dicey, sleep 2 always worked, sleep3 just because):

service rgmanager stop
[ $? -eq 0 ] || exit

service gfs stop
[ $? -eq 0 ] || exit

service clvmd stop
[ $? -eq 0 ] || exit

service fenced stop
[ $? -eq 0 ] || exit

# invoking cman too early sometimes fails ... -t option too small?
sync
sleep 3

[ "$ACTION" = "leave" ] && service cman stop leave || service cman stop
[ $? -eq 0 ] || exit

service ccsd stop
[ $? -eq 0 ] || exit


On Mon, 2007-10-22 at 14:05 -0600, Josh Gray wrote:

> I see in the FAQ the proper way to leave a cluster is with 'cman_tool leave
> remove'  I am getting a  'device or resource busy' error when I do that even
> with GFS mounts removed and stopped.   What am I missing?  The only service
> I see up is a "fence" type called "default"  do I have to stop one more
> thing?
> 
> Basically doing this:
> 
> service nfs stop
> umount /export
> service gfs stop
> service rgmanager stop
> service clvmd stop
> cman_tool leave remove
> 
> Tells me this:
> cman_tool: Error leaving cluster: Device or resource busy
> 
> 
> Full snip from the server:
> 
> [EMAIL PROTECTED] ~]# cman_tool services
> type             level name         id       state
> fence            0     default      00010003 none
> [1 2 3]
> dlm              1     clvmd        00020003 none
> [1 2 3]
> dlm              1     rgmanager    00050003 none
> [1 2 3]
> dlm              1     nfs_data_vg  00040003 none
> [1 2 3]
> gfs              2     nfs_data_vg  00030003 none
> [1 2 3]
> [EMAIL PROTECTED] ~]# service nfs stop
> Shutting down NFS mountd:                                  [  OK  ]
> Shutting down NFS daemon:                                  [  OK  ]
> Shutting down NFS quotas:                                  [  OK  ]
> Shutting down NFS services:                                [  OK  ]
> [EMAIL PROTECTED] ~]# umount /export
> [EMAIL PROTECTED] ~]# service gfs stop
> [EMAIL PROTECTED] ~]# service rgmanager stop
> Shutting down Cluster Service Manager...
> Waiting for services to stop:                              [  OK  ]
> Cluster Service Manager is stopped.
> [EMAIL PROTECTED] ~]# service clvmd stop
> Deactivating VG nfs_data_vg:   0 logical volume(s) in volume group
> "nfs_data_vg" now active
>                                                            [  OK  ]
> Stopping clvm:                                             [  OK  ]
> [EMAIL PROTECTED] ~]# cman_tool leave remove
> cman_tool: Error leaving cluster: Device or resource busy
> [EMAIL PROTECTED] ~]# cman_tool services
> type             level name     id       state
> fence            0     default  00010003 none
> [1 2 3]
> [EMAIL PROTECTED] ~]# 
> 
> 
> 
> 


Robert Hurst, Sr. Caché Administrator
Beth Israel Deaconess Medical Center
1135 Tremont Street, REN-7
Boston, Massachusetts   02120-2140
617-754-8754 ∙ Fax: 617-754-8730 ∙ Cell: 401-787-3154
Any technology distinguishable from magic is insufficiently advanced.

Attachment: smime.p7s
Description: S/MIME cryptographic signature

--
Linux-cluster mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-cluster

Reply via email to