Since the commenters made a very good case that bad RAM byte 64h
should not affect Linux, I retested the RAM a 35C (maximum operating
temperature for the Dell Inspiron 2650 laptop) to see if the rest of
the RAM had weak bytes. Only 64h tested bad.
Since the chamber was going, I removed the hard drive to retest it at
the higher temperature. Rather than just a basic e2fsck test (which
had been done several times), I added the -c option (badblocks). Oh
Oh. The drive hung up about 20 minutes through the test with its
"access light" on steady but giving no response.
The drive tested the same at 20C with either the -c or -c -c options.
However, it always passed e2fsck OK without the -c option set. Another
similar drive passed all of these tests OK, so the setup is valid.
Bad drive: Hitachi Travelstar 2.5" 40GB 5400rpm Feb-03
Passed: e2fsck -f -n -t -v /dev/sdb2
Failed: e2fsck -c -c -f -n -t -v /dev/sdb2
Failed: e2fsck -c -f -n -t -v /dev/sdb2
Passed: badblocks -b 4096 -v /dev/sdb2 9504871 (read only)
Failed: badblocks -b 4096 -n -v /dev/sdb2 9504871 (non-destructive r/w)
Passed: e2fsck -f -n -t -v /dev/sdb2 (no bad blocks although badblocks
just failed)
So it appeared to be a hangup in the drive's internal controller
when it has a long series of read/writes to do. The "locate" database
search may have triggered it in situ. I doubt I will ever know and no
longer care.
Cost of saving Dell Inspiron 2650 (original cost ~$800)
"Technician" @ $40/hr 57 hr ...... $2280.00
Book "Understanding the Linux Kernel" 49.95
Ice cream to sooth nerves (6 times) 33.37
Replacement hard drive (160GB) 93.75
Rebate from wife for saving environment -22.78
Total 2434.29
Jim Kuzdrall
_______________________________________________
gnhlug-discuss mailing list
gnhlug-discuss@mail.gnhlug.org
http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
