Our central email server has problems with its SCSI controller. Or possibly with one of its SCSI hard drives. Looks like I’ll be spending a couple of hours at work this evening trying to fix this and/or replace the faulty components. 🙁
Scuzzy SCSI
7 Responses to Scuzzy SCSI
Comments are closed.
More likely the disk rather than the controller – what sort of errors you get – timeouts?
PermalinkI suspect it’s the controller rather than the disk too, although the errors only appear when the disk is under heavy load, such as when running the AMANDA backups.
Just as well I managed to persuade our supplier to send me a spare disk and controller “for evaluation”, I guess.
Permalinkit is all over the disk, or certain sectors, ie same sectors again and again?
Certain sectors -> more likely disk
all over -> more likely controller or cable.
Of course it could be the PSU not handling the load…..
PermalinkOne more thought
I’ve got amanda 2.4.2p2 running on a Sunons 4.1.3 machine that seems to hang to thing when doing a full dump of all the filesystems more often than not.
But there are a couple of known dodgy disks on the system, although I see nothing on the log files to indicate SCSI errors.
The TCP/IP stack just locks solid….
PermalinkJust finishing up the job now – I’ve replaced the presumed-faulty disk with a new one. Fingers crossed …
Oh, and I’ve ordered a new dedicated backup server for the backups.
PermalinkNice big LTO tape drive and big holding area?
PermalinkModest tape drive, but holding area will be a 250GB disk … 🙂
Permalink