I'd rather not loose my job...

Sage

13 short of a dozen
Regular
Man this sucks. Today I come into work and one of the drives in one of the IBM servers RAID arrays is down. so, i go to replace it according to the documentation. little problem- i dont know which drive is which, and none of them have little red lights lit up. Okay, i THINK i know which is the bad one... so i pull it out and it tells me now there's another defunkt drive... OOPSIE! Okay I put it back in... but i put it in th wrong slot (there's two open slots). Uh oh, now its got a little red light. yeah that's bad. ok i pull it and put it in the other slot. sht. still a red light. okay well ill fin which drive does need replacing.. pull the next drive and it appears to be the one i need to yank. ok i put the new drive in. damn. red light. ok hmm. pull another drive, and it tells me that another drive is defunkt. ok return to last situation. well, only 4 and 5 are defunct. thats good- 5 was the original one and 4 is the one i buggared. sht. why isnt it rebuilding? ok restart... oh good now the red light on 5 is blinking so that means it's rebuilding. but the red light on the other one is not... so its still defunct, thats bad. and the computer wont start up either. it says there's a disk failure... yup, prolly cause it's rebuilding the array.. or something... so now our scheduling / billing apps is down. our DNS server is down, and people cant get into some of their documents. the server was backed up friday to tape... HOPEFULLY. Well, the guy that I'm replacing that they want to get rid of is comming now and he's gonna charge a whole lot of money. he will hopefully be here in the next hour or so.


sht i hope i dont loose my job.


oh yeah and my cousin WAS dating the boss. thats how i got it. she's not anymore. that doesnt help.
 
xServers

edit: err you mean the drive.. duh... i should be more specific....

they say xSeries on them. they are SCSI U160.
 
All relatively recent deskstar drives are fine from a reliability standpoint (most GXP60, all GXP120 and up). Besides, IBM's SCSI drives are the ultrastar series and they didn't have these issues.
 
Guden Oden said:
most GXP60, all GXP120 and up
The GXP60 line had a hellacious failure rate too, but not quite as bad as the GXP75 (I think it was 75 :? )

I had a 60, it failed; I was sad. :(
 
The 60 series had a firmware flaw that caused the failures, it may have been the same with the 75 series too, but those had seriously bad bearings as well. IBM released updated firmware for the 60 drives long ago that prevented the failure from happening.

Apparantly the bad firmware hovered the r/w heads constantly over a particular section of the disk when idle which either caused heat buildup to demagnetize the tracks, or it caused buildup of "pixie dust" on the heads that prevented them from functioning properly (also heat related as I recall). I don't remember exactly what the full story was on that one.
 
Back
Top