[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[HTCondor-users] Slot partitioning issue?

I was checking another issue with Condor when I discovered this:

# condor_q -run | grep $node
2431359.0   user1           7/16 19:14   0+20:09:47 slot1_2@xxxxxxxxxxx
2432569.0   user2           7/17 15:00   0+00:24:03 slot1@xxxxxxxxxxx

All claimed slots are supposed to be 1-core - and I'd expect *all*
Names to be of the form "slot1_i@xxxxxxxxxxx". What could produce
such a mix?

The original issue, for completeness:
There are some parallel universe jobs running, each requesting 8
machines/slots, but the total number of slots claimed by that user
(who limits himself to parallel universe) is not a multiple of 8.
What may have caused this?

$CondorVersion: 8.3.5 Apr 20 2015 BuildID: 315634 $
$CondorPlatform: x86_64_Debian7 $