stream of puppet errors: missing fix for AI-4868 (new selinux_puppet) which went into production last week
replicating the (enormomous) INDICO volumes on the new critical VMs caused the project to run out of quota (reported at C5). Increased.
during same time the "ARC" server ran out of memory (stuck command pileup), might have affected INDICO ability to monitor their space use
now running "accessTime" also from outside of the fileservers. Funny side effects (much increased latency), caused some wild volume re-replication.
part of getting rid of "monitor", and of running AFS-less fileservers
Discussion: "don't touch", i.e continue to run from inside fileservers, from local cron job (not "monitor")
one ticket for a "computing group" (RQF0788182) was apparently more a request for group space (not some new UNIX group that needed HEPIX scripts).
Discussion: will block the creation of new /afs/cern.ch/group areas from AIM. This will cause errors (but is a rare operation) - need to get in touch with requester to figure out what they want.
There are minutes attached to this event.
Show them.