Hi
I am trying to install/enable checkpoiting with slurm 2.3.5. blcr-0.8.4
is installed and the basic functionality is working. I can by hand
checkpoint a job and restart it on a different node. Then I had enable
checkpoiting/blcr in slurm. slurm writes all necessary information to
the configured files, when I request a checkpoint.
But I cannot restart the job with "scontrol checkpoint restart job_id".
In the jobs error_log there is the following output:
- open('/var/db/nscd/passwd', 0x0) failed: -13
- mmap failed: /var/db/nscd/passwd
- thaw_threads returned error, aborting. -13
- thaw_threads returned error, aborting. -13
Restart failed: Permission denied
What's wrong?
Best regards,
Tibor
