This afternoon the airconditioning units of the NMV room at Versatel Nossegem (where I have my kit colocated) all failed together (yes, clustering is great, so you can all go down at once!). This made the temperature rise from a nice and fresh 22°C to 65° in some places of the room.
In the meanwhile everything’s restored, but I just had a server act up a little in such a way that I had to powercycle it to get it back into gear. After this I could clearly see there were some disk problems causing above acting up. Apparently the heat had given my second disk (in software RAID1) some delusions of grandeur:
hda: Maxtor 6Y080L0, ATA DISK drive
hdc: ÿ, ATA DISK drive
hda: max request size: 128KiB hda: 156250000 sectors (80000 MB) w/2048KiB Cache, CHS=65535/16/63, (U)DMA
hda: cache flushes supported
hda: hda1 hda2 hda3 hda4
hdc: max request size: 1024KiB
hdc: 18374966859414961920 sectors (143554428589 MB) w/32640KiB Cache, CHS=65535/255/63
143554TB of IDE disk, with 32 megs of cache! Aircon failures clearly improve your harddisks, I can highly recommend them! 🙁
Incoming Links (via Technorati):
Nothing Reported
December 29th, 2007 at 02:57
and i thought that the google report showed that temperature variances did not necessairely have a big influence on hdd failure rate – so in essence that not cooling a disk does not mean it will fail faster (but ofcourse 65 degrees room temp is perhaps already above the specs inside your case …)
December 29th, 2007 at 09:40
Temperature variation probably doesn’t help – but a harddisk is already warmer than the temp of your chassis, which is already warmer than room temperature.. so I dunno. Maybe it’s even a coincidence… 😉