Re: [HTCondor-users] Question about condor_history, Run_Time, and RemoteWallClockTime.

[Stub <spamrefuse@xxxxxxxxx>]

On Tuesday, October 28, 2014 5:57 PM, Stub wrote:

Hi,

When I use the condor_history to check the "RUN_TIME" of a job, I am puzzled about the given time.

I'd like to use an example:

* A job runs 1 hr on a pool PC and gets evicted.
* The job then spends 2hrs hanging around in the queue.
* The job is restarted and completes after 3 hrs on another pool PC.

What time does condor_history report as "RUN_TIME"? "4hrs" or "6hrs" ?

I did understand that the RUN_TIME also includes possible time on the PC as suspended. There's no mentioning of the intermediate time in the queue, so I guess the above example should render 4 hrs RUN_TIME. I just want to be sure.

Thank you!
R.L.


------------------------------

As follow up on my own email:

Our HTCondor network of pool PCs consists of Win7 public library PCs.
Users can switch the PCs on and off ad libitum.

So occasionally HTCondor is running a job, when the PC is suddenly switched off, without giving HTCondor the time to gracefully handle the situation.

It takes the HTCondor master a while of waiting time to realize that it is wiser to give up on that dangling job and restart it elsewhere. In this case the RUN_TIME parameter is muddled up, for which I guess HTCondor has no blame.....but it also means that in this setup the RUN_TIME parameter should not be used for accounting and/or billing users.... 

Did/Do other people encounter similar issues?

R.L.