[GE users] qmaster hang

murple andreas.kuntzagk at mdc-berlin.de
Tue Dec 1 14:31:43 GMT 2009


I just had the case that a user tried to delete 1000 jobs using qmon. 
This caused qmaster to consume all memory and start swapping. The node 
had a load of upto 20 and was quite unresponsive.
I use 6.2 with classic spooling with NFS3. Is this a known bug? A reason 
to migrate to bdb spooling? Or to update to newer version?
In the recent days I already had some instabilities with this head node. 
But today was the first time I was watching it while it "crashed".

regards, Andreas


To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].

More information about the gridengine-users mailing list