[torqueusers] Checkpointing and restart with torque 2.4 with BLCR

Rajiv Rajaian rajiv.care at gmail.com
Tue Mar 30 04:55:38 MDT 2010


Hi Alexander
I ve replaced those scripts now I'm getting some error as unsupported
version. Im using Fedora core 3 .. Whether it ll be a problem
May I  know in which OS you have tested the checkpoint feature

Heres the error im getting in /var/log/messages

*Mar 30 16:23:23 gcluster checkpoint_script: Invoked:
/var/spool/torque/mom_priv/checkpoint_script 15313 10.gcluster.grid guser02
guser02 /var/spool/torque/checkpoint/10.gcluster.grid.CKckpt.10.gcluster.grid.1269946403
15 -
Mar 30 16:23:23 gcluster kernel: blcr: request from pid 15324 for
unsupported version 0.10
Mar 30 16:23:23 gcluster checkpoint_script: Subcommand (cr_checkpoint
--signal 15 --tree 15313 --file ckpt.10.gcluster.grid.1269946403) failed
with rc=53: Failed cr_init(): Requested kernel interface version is not
supported
Mar 30 16:23:23 gcluster pbs_mom: LOG_ERROR::blcr_checkpoint_job, checkpoint
script returned value 53

*
On Tue, Mar 30, 2010 at 3:40 PM, Alexander Oltu <Alexander.Oltu at uni.no>wrote:

> On Tue, 30 Mar 2010 15:20:56 +0500
> Rajiv Rajaian wrote:
>
> > Hi Alexandar
> > Im here with attaching the blcr_checkpoint & blcr_restart scripts
> > which Im using here. pls refer that
> > If u have that working  scripts will u please share those scripts.. im
> > struggling a lot with this ..
>
> Just remembering that I used scripts from torque 2.4.5 source
> directory in contrib/blcr
>
> they looks differently and requires 8 arguments instead of 7 as yours
> does.
>
> Regards,
> Alex.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20100330/d47eefdf/attachment.html 


More information about the torqueusers mailing list