[supercell-announce] Supercell1 RAID failure

Justin Dugger jldugger at osuosl.org
Mon Jun 9 18:17:22 UTC 2014


We regret to inform the Supercell community that supercell1 has
suffered a complete RAID failure. We're diligently working to restore
service to this machine, but it will naturally take some time to
rebuild. supercell's loss leaves us with roughly half the capacity of
VMs until repaired. Please bear with us!

I should remind everyone that Supercell is designed as a testing
platform; there are no backups, VMs are not DRBD replicated, and there
are no hot spare drives. As such only transient data should remain in
the cluster. We have had reports that VMs on supercell remain
functional; this is likely due to RAM caching. If there is data you
wish to retrieve from RAM, we'll leave the system on in degraded state
until later this afternoon (2-3 hours), but restoring continuous
integration testing is a higher priority than data recovery.

List of affected VMs:
drupaltestbot654-mysql.osuosl.test
drupaltestbot659.osuosl.test
drupaltestbot664.osuosl.test
drupaltestbot958-mysql.osuosl.test
drupaltestbot1823.osuosl.test
drupaltestbot1828.osuosl.test
drupaltestbot1833.osuosl.test
drupaltestbot1838.osuosl.test
drupaltestbot1843.osuosl.test
drupaltestbot1848.osuosl.test
drupaltestbot-staging-mysql.osuosl.test
drupalworkerdev1-mysql.osuosl.test
gnome-testing01.osuosl.test
gnome-testing02.osuosl.test
hive.osuosl.test
hphp-maverick-32.osuosl.test
hphp-maverick.osuosl.test
httpdtest
openmrs-devtest00.osuosl.test
windows-7

Justin Dugger
Senior Systems Administrator
OSU Open Source Lab


More information about the supercell-announce mailing list