Hi Dave,

Thanks for your answer.

El 13/06/11 16:31, Dave Love escribió:
Javier Lopez Cacheiro<[email protected]>  writes:

I don't know if anyone else has come over this same problem with 6.2u5
I've seen consumables go negative, but when things were too frantic to 
investigate.

and if there is a workaround for it.
Not that I know.  It might be related to
https://arc.liv.ac.uk/trac/SGE/ticket/793, which I can't remember seeing
recently and never tried to debug.

Interestingly enough this issue is also related to parallel jobs using a PE. This was similar to the previous issue with 6.2u1 where using a PE with the fillup allocation policy lead to unexpected results with consumables.

In this case the important distinction is that the last job to enter in execution was a non parallel job.

By the way, we have done some regression testing and verified that the old bug with PE+fillup and consumables is now solved, so this seems something different.

Cheers,
Javier

<<attachment: jlopez.vcf>>

_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to