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

RE: [Condor-users] Why could this match not be made?



> And so on. But there are machines that meet the requirements! 
> I'm looking at them, sitting Unclaimed+Idle. Very frustrating.

This has something to do with stard communication to the collector. I
restarted condor on one of the machines that was showing in the
condor_status output has being Unclaimed+Idle for a long time (>20
minutes). I talk directly to the machine now I see:

[0] > condor_status -direct ttc-bs3066-196.altera.com
 
Name          OpSys       Arch   State      Activity   LoadAv Mem
ActvtyTime
 
vm1@TTC-BS306 WINNT51     INTEL  Unclaimed  Idle       0.760  2015
0+00:06:46
vm2@TTC-BS306 WINNT51     INTEL  Unclaimed  Idle       0.000  2015
0+00:06:46
 
                     Machines Owner Claimed Unclaimed Matched Preempting
 
       INTEL/WINNT51        2     0       0         2       0          0
 
               Total        2     0       0         2       0          0

But if I ask condor_status for information about this machine it's very,
very old:

[0] > condor_status -const 'machine=="ttc-bs3066-196.altera.com"'
 
Name          OpSys       Arch   State      Activity   LoadAv Mem
ActvtyTime
 
vm1@TTC-BS306 WINNT51     INTEL  Unclaimed  Idle       0.760  2015
0+00:41:48
vm2@TTC-BS306 WINNT51     INTEL  Unclaimed  Idle       0.000
2015[?????]
 
                     Machines Owner Claimed Unclaimed Matched Preempting
 
       INTEL/WINNT51        2     0       0         2       0          0
 
               Total        2     0       0         2       0          0

It appears that there is a considerably lag to update machine status
information in the collector. The UPDATE_INTERVAL is 150 seconds on this
machine -- but it appears to be taking longer than 150 seconds for the
information to be refreashed at the collector. Does anyone know why? And
how I might remedy this?

- Ian