Note of the week:
1) certificates on submit06 are expiring, in 6 weeks,
planning to do end af august so that we have still fews weeks
2) submit00 is always overloaded
action (Maria): remove the 00 from the ssh rebalancing so that one log in intentionally i.e. to run mathematics ...
(check if is still on slurm)
Ongoing:
- matt:
- importing jacob's slurm job efficiency monitoring
- David:
- linux security upgrade ongoing https://cleo.mit.edu/issues/6178, (see status from archi)
- ceph 20.2.23 upgrade, one of the old failure mode bug fixes in
long term:
review the security mitigations of the cluster.