Can I use yggdrasil during baobab maintenance?

HPC Technical
1

I try to run sbatch jobs on yggdrasil, but:

  • saact throws errors:
sacct: error: slurm_persist_conn_open_without_init: failed to open persistent connection to host:slurm1:6819: Connection refused
sacct: error: Sending PersistInit msg: Connection refused
sacct: error: Problem talking to the database: Connection refused
  • when I try to read data in my scratch folder I get disk errors: disk I/O error
  • slurm_load_jobs throws: slurm_load_jobs error: Invalid job id specified while the job id is correct
  • jobs fail after 1s without a reason

Thank you!

2

Hi @maciej.falkiewicz

Yes you can use Yggdrasil during Baobab maintenance. At least this is how it is supposed to work:)

Unfortunately, we had a storage issue on Yggdrasil Current issues on Baobab and Yggdrasil - #69 by Yann.Sagon during Baobab maintenance.

About the error with sacct. Indeed, this command uses the slurm database that is hosted on Baobab and thus may be unavailable. With sbatch you should’ve no problem at least as long as you don’t include the line #SBATCH --cluster in your script.

1 Like