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

Re: [HTCondor-users] Accessing historical condor usage statistics

Another possibility is to make a big ASCII tuple using condor_history, then run some code to pull out whatever you want:

/usr/bin/condor_history \
   -format "%d." ClusterId -format "%d " ProcId -format "%s " Owner \
   -format "%d " IsAnalyJob -format "%s " LocalQue \
   -format "%s " LastRemoteHost -format "%d " JobCurrentStartDate \
   -format "%d " CompletionDate -format "%f " RemoteSysCpu \
   -format "%f " RemoteUserCpu -format "%f " RemoteWallClockTime \
   -format "%d " ExitStatus -format "%d " ImageSize \
   -format "\n" In > $tempScanFile


On 1/4/2016 4:03 PM, Michael V Pelletier wrote:
From: Todd Tannenbaum <tannenba@xxxxxxxxxxx>
Date: 01/04/2016 02:32 PM
> 2. I have not personally used the condor_stats command for years; these
> days here at UW we instead use the condor_gangliad to send usage info to
> Ganglia and also use the condor_pool_job_report script to generate daily
> reports.  See

I can put in a plug for Ganglia - I just set it up at the end of December and was pleasantly surprised at how easy it was using the EPEL repository packages for CentOS/Red Hat. I had it up and running on the central manager with just a couple of lines of config tweaks, including starting the condor_gangliad. The fact that it aggregates data from clusters of machines right out of the box was a big benefit for what I'm working on now - I needed something that could give me a comprehensive overview of the pool quickly without a lot of tinkering, and that's just what I got.

   -Michael Pelletier.

HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting

The archives can be found at: