[Linux-cluster] DLM Problem

[isplist at logicore.net]

>i mean, we had this clvmd issue quite a few times on the mailing list,
>and as far as i can remember there wasn't really a fix to this. (i
>solved it by rebooting the whole cluster back then... that was certainly
>bad).

Something went out of sync and it doesn't seem to make sense that it's gotten 
so complicated. I've not changed anything since everything was working just 
fine. 

When I built my first GFS cluster, there was a delay when first accessing the 
storage which eventually went away. When I fired up the (sama, unchanged) 
cluster after changing out some drives this time around, all hell broke loose.

If I did something wrong, I'd love to learn from the mistake but since I'm now 
knee deep in confusion, I can't seem to find the answer.

I'll try a full shutdown again and see if things get any better.

Mike