Displaying report 1-1 of 1.
Reports until 17:35, Tuesday 20 October 2015
H1 AOS (DCS)
dan.moraru@LIGO.ORG - posted 17:35, Tuesday 20 October 2015 (22701)
summary of DCS maintenance
DCS has fully recovered from this morning's power outage.  Affected by the power outage were all compute nodes, a couple of switches, the ACSLS server, and a disk expansion chassis.  The switches and ACSLS server have single power supplies and so could not be moved onto UPSes previously without disruption.  They are now UPS-protected, as is the E18X expansion chassis that had mistakenly been left off UPS.  We took advantage of the downtime to patch and reboot all Solaris servers.  The LDAS gateway failed to come back up and required intervention, but that issue has been resolved.  The Condor central manager is now running on new hardware.  The gstlal-calibration packages were updated cluster-wide, and llldd partitions are now locked into memory on all servers other than dmt-er.
Displaying report 1-1 of 1.