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

Re: [Condor-users] checkpoint / Restart



Vanilla means the checkpointing associated with the standard universe
is NOT possible.

That said you can achive a similar result if you take responsibility
for your own checkpointing and exit in time after the relevant vacate
signal is sent. there are a considerable number of config gotchas with
this set up (which aren't terribly well detailed in the manual)

The info is scattered across several different subjects but you need
to look at the KILL settings, transfer parameters and where you
persist state to...

Read posts passim on the subject if you want to try.

Matt

On Wed, 3 Nov 2004 15:09:47 +0100, Patrick Jaeger
<patrick.jaeger@xxxxxxxxxx> wrote:
> 
> 
> Hi every body
> 
> I would install CONDOR on Linux Cluster based on AMD opteron with RH AS 3
> kernel 64bits ,
> and Erik Paulson have indicated the following version :
> The IA32 RH9 dynamic binaries run just fine in vanilla mode.
> 
> Can you confirm if the  feature Checkpoint Restart is available with  this
> version of CONDOR
> 
> If it's OK have you some comparison with the MEIOSYS products ? .
> 
> thanks
> 
> Cordialement / Bests regards
> --------------------------------------------------------------------------------
> 
>   Patrick Jaëger
>       I/T Specialist    SSIS  =:)   IBM France
> tel : 01 49 05 52 74    gsm : 06 71 92 20 77
>    tel int : 335274     Email : Patrick.Jaeger@xxxxxxxxxx
> 
> _______________________________________________
> Condor-users mailing list
> Condor-users@xxxxxxxxxxx
> http://lists.cs.wisc.edu/mailman/listinfo/condor-users
>