Page 1 of 1

Restart the run when the checkpoint folder is missing

Posted: Tue Dec 13, 2022 4:25 pm
by sofenkumarjena
Hi user,
Last night there was uneven breakdown in Curta cluster due to cooling failure. Few of our cases terminated after almost 45 hours of run. The checkpoint file is not saved for these cases, however a temporary restart file is available in the folder. Is there any way to recover the results for further run.

It will be great help for me


With Thanks and Regards

Sofen

Re: Restart the run when the checkpoint folder is missing

Posted: Tue Dec 13, 2022 9:56 pm
by Yvan Fournier
Hello,

Depending on whether this us complete it might work or not, but simply try moving the files in rhe temporary checkpoint to the checkpoint directory, directly under "checkpoint".

Best regards

Yvan