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

Re: [Condor-users] Condor checkpointing




On Thu, Sep 24, 2009 at 8:33 PM, Mag Gam <magawake@xxxxxxxxx> wrote:
so, how does it work now? Because, I couldn't find any documentation
or example.
It would be nice if there was a simple HOWTO to create a checkpoint
enabled job. Or is there a best practices guide for it?

The Condor documentation covers checkpointing extensively:

http://www.cs.wisc.edu/condor/manual/v7.3/2_4Road_map_Running.html#SECTION00341100000000000000

http://www.cs.wisc.edu/condor/manual/v7.3/4_2Condor_s_Checkpoint.html 

http://www.cs.wisc.edu/condor/manual/v7.3/condor_compile.html

http://www.cs.wisc.edu/condor/manual/v6.0/3_9Setting_up.html#SECTION00493000000000000000

Finally, the condor_compile script is very straight-forward. I'd encourage you to read it, and the ld script in the lib/ directory of Condor. It should give you a good understanding of how exactly linking for checkpointing works in Condor.

-Erik

We lost several days worth of work due to a bad memory board on a
server. Don't want this to happen again...



On Fri, Aug 21, 2009 at 2:33 AM, Carsten Aulbert
<carsten.aulbert@xxxxxxxxxx> wrote:
> Hi
>
> On Friday 21 August 2009 05:25:37 Mag Gam wrote:
>> I am interesting in condor checkpointing because at my lab we have
>> models which run for days. But most of our code is done in Python,
>> Perl and awk. I was wondering if its possible to setup checkpointing
>> for something like this?
>
> Not right now. So far you have to condor_compile the stuff to get
> checkpointing.
>
> I think (Condor people should correct me here) that they plan of getting
> something more sophisticated out there which will allow exactly this. But I
> don't know if this already planned for the 7.6 release.
>
> Cheers
>
> Carsten
> _______________________________________________
> Condor-users mailing list
> To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx with a
> subject: Unsubscribe
> You can also unsubscribe by visiting
> https://lists.cs.wisc.edu/mailman/listinfo/condor-users
>
> The archives can be found at:
> https://lists.cs.wisc.edu/archive/condor-users/
>
_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/condor-users/