[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] lock_dlm to lock_nolock



Hi,

On Tue, 2008-07-08 at 13:20 -0400, Ozgur Akan wrote:
> Hi,
> 
> Can I mount a gfs filesystem formatted with lock_dlmlock and use it
> without  a problem in the cluster if I have proper fencing and that fs
> is mounted to only one node at a time?
> 
Single node DLM is quite possible, and I use it for testing from time to
time. Below though you appear to be using lock_nolock which is also ok
provided you only use it on one node at a time,

Steve.

> mount -o
> lockproto=lock_nolock /dev/mapper/cluster_vg-test2_lv /gfstwo/
> 
> 
> [root rhtest01 ~]# ./ping -rw /gfstwo/test 1
> data increment = 1
>   140012 locks/sec
> [root rhtest01 ~]# gfs2_tool df /gfstwo/
> /gfstwo:
>   SB lock proto = "lock_dlm"
>   SB lock table = "testcluster:gfstwo"
>   SB ondisk format = 1801
>   SB multihost format = 1900
>   Block size = 4096
>   Journals = 3
>   Resource Groups = 60
>   Mounted lock proto = "lock_nolock"
>   Mounted lock table = "testcluster:gfstwo"
>   Mounted host data = ""
>   Journal number = 0
>   Lock module flags = 1
>   Local flocks = TRUE
> 
> 
> thanks,
> Ozgur Akan
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]