[6253] in linux-scsi channel archive
Temperature Sensitive Drive.
daemon@ATHENA.MIT.EDU (Mark Garlanger)
Sat Apr 3 00:15:19 1999
Date: Sat, 03 Apr 1999 05:08:47 +0000
From: Mark Garlanger <garlangr@cyberramp.net>
To: linux-scsi@vger.rutgers.edu
Hi everyone,
I have a problem that hopefully someone will have some experience
with...
Short story:
----------
Drive only seems to work after it exceeds some minimum
temperature (which happens to be above room temperature (~72 degrees
F.)) Any ideas?
Long story:
----------
I currently have a drive that is acting up. I think I may have cause
the problem... some background. It's a Micropolis 8.7G 7200RPM. It had
been working good in my old system, but
I did not have very good airflow, If I left the system on for a long
time (a couple of weeks) and then attempted to do something that was
disk intensive the system would sometimes lock-up, but it would ALWAYS
work fine once I let it cool down and restarted ( This probably happened
a total of 3 times.) In my new system, I got a BayCooler for it, but
unfortunely, the mounting screws on the BayCooler did not match up with
the drive's. Also my other 3 1/2" to 5 1/4" bracket would not fit with
the fans mounting of the BayCooler... So I ended up mounting the drive
the only way possible - with only 2 screws(the front right and the rear
left). I didn't like to do this but I thought that keeping the drive
cool would be more important that this mounting, (now I think this may
have cause the problem??? Maybe some type of warping of the case???)
But after about 3 weeks with it on, the drive locked-up. The
temperature of the drive did not seem very warm due to the fans. I then
cycled the power, which did not work, I then let it cool to room
temp(about 1 hour) and turn it back on, but it would not boot and the
bios would either say 'Drive not ready' error or it would take a very
long time to post the drives (controller: AIC-7895). I thought the
drive was dead, it would not boot no matter what I did, but then I
discovered that if I used the Adaptec BIOS to verify the disk surface,
it would eventually come back to life!!! Initially it said every sector
was bad: 0, 1, 2, 3, 4, 5, 6 I was thinking that it may be time to
play with a few drive platters BUT then I noticed it skipped 7 and said
8 was bad! I continued to let it scan the disk, as it warmed up, I was
getting less and less bad sectors. I restarted the scan at the
beginning of the drive and sectors that initially displayed an error was
now working although some were still showing up as bad... As I
continued, the errors became less (about 1 in a few thousand), and
less. Eventually it did not give any errors for the last ~40% of the
drive. I then reset the system and was able to get the system to boot
WITHOUT any problem. I was able to copy as much as I wanted, and the
system stayed up for a few hours before I decided to turn it off to see
if it would still have a problem. The problem remained once the drive
cooled down...
It seems like the drive now has a lower temperature limit that has to
be exceeded before the drive will work!! Since the temperature is
higher than room temperature it is a real pain, plus I don't know how
much I can trust this disk now.
So, do you think only using two screws caused this problem, is there
any ideas how to reverse this?? Reformat when disk is at room
temperature? mount with only the other two screws for a few weeks???
Anything???
Thanks in advance,
Mark
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
