[Linux-cluster] Problem in clvmd/dlm_recoverd

[Tom Lanyon]

On 19/11/2008, at 3:00 PM, Tom Lanyon wrote:

> We haven't been able to catch it quick enough to determine which  
> process is using all CPU.
>
> The other option is that we're just seeing a huge amount of glocks  
> created on the node running backups and all others (webservers) are  
> just hanging whilst trying to access files. I've just done some  
> fairly aggressive tuning of the GFS mounts on all nodes; hopefully  
> this fixes it!
>
> Regards,
> Tom

After tuning some GFS parameters yesterday, last night's backup ran  
without a hitch! :)

Tom