Reuti wrote:
This is also some kind of personal taste. Some prefer classic spooling as you
can check all the information of a job as they are just stored as text files.
And it even handles a large number of nodes before it gets performance
problems. Maybe Chris can make a statement about it as he runs large clusters
with classic spooling IIRC.
Hmm, but it has obviously nothing to do with my spooling problem I
guess. I asked another university. They use the same Berkeley DB as I do
and they have not that problem. You have much experience with the SGE.
Do you think I should compile and install another version?
You tried a file in / and it's working then, e.g. copy "/bin/true" to "/true"?
Yes, with /true it works then. But not with "startmpi.sh" --> /stamp.
There I get "got wrong number of arguments". Probably because it does
see only the first 6 characters and not "-catch_rsh $pe_hostfile".
Ursula
_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users
