Mon, 09 Jan 2006
On Wednesday, the HDD in my workstation crashed - firstly it looked like a flaky cable (I rebooted the system), but after few minutes (while doing an emergency backup) the drive disconnected from the bus and never came back. I tried to do a classical refridgerator resuscitation, but I was only able to make the drive visible on the bus, but it did not spin up.
Fortunately, I had a few weeks old backup on my secondary HDD (did you know that cp -lR and a local rsync can make a decent HDD-based backup system?). O&G managed to deliver a new HDD the next day, so I am moving my system back to the usable state.
It is interesting how a not so big company like O&G can often provide better support/replacement parts than big players like HP, IBM, or even SGI, and for much less money.
In the meantime, the same drive in Odysseus failed the S.M.A.R.T. self-test. I have removed it from the RAID-5 array and inserted it back, and it synchronized itself correctly. A subsequent self-test was OK as well (and no sectors were relocated), so I hope the drive will work correctly. I already have a set of new HDDs and a 3ware SATA controller for Odysseus, but I simply do not have time to do the upgrade.