Scheduled downtime of NHR@FAU systems on Monday, February 26
There will be a scheduled downtime of all of NHR@FAUs HPC systems, including frontends and fileservers, on
Monday, February 26, starting at 08:30, and lasting the whole day.
Reason for the downtime is miscellaneous maintenance, including but not limited to some central fileservers, which is why you will be unable to even log in or access files for at least part of the maintenance period. As usual, jobs that would collide with the downtime will be postponed until the downtime is over.
We will update this post once batch processing resumes.
Updates:
2024-02-26 @ 20:50 – batch processing on Woody has been resumed
2024-02-26 @ 20:55 – batch processing on Alex has been resumed
2024-02-26 @ 21:20 – batch processing on TinyX has been resumed
2024-02-26 @ 21:30 – batch processing on Fritz has been resumed
2024-02-26 @ 21:40 – batch processing on Testcluster has been resumed
2024-02-27 @ 07:55 – batch processing on Meggie has been resumed
2024-02-27 @ 09:05 – $FASTTMP (/lustre) was unavailable on the Fritz frontends. This has been fixed. (Note that computenodes were not affected, just the frontends.)
2024-02-27 @ 11:20 – some users experienced bogus “over quota” messages on $HOME (/home/hpc) or $HPCVAULT (/home/vault) since the downtime; this should be fixed since around 10:15.
Things should be back to normal now. Please report remaining problems.
Unfortunately, we could not complete all maintenance tasks during this downtime. There will be another one in about two weeks.