dld@ihlpf.ATT.COM (Dennis L DeBruler) (03/18/88)
One Monday morning I noticed that my disk drive sounded different. I checked /usr/adm/unix.log and found some entries of the following form: HDERR ST:51 EF:10 CL:FF9B CH:FF02 SN:FF0E SC:FF02 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:9500 Mon Feb 22 12:44:26 1988 drv:0 part:2 blk:37687 rpts:1 Mon Feb 22 12:44:29 1988 HDERR ST:51 EF:10 CL:FF7E CH:FF02 SN:FF04 SC:FF02 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:DF00 Mon Feb 22 12:52:25 1988 drv:0 part:2 blk:35826 rpts:1 Mon Feb 22 12:52:26 1988 HDERR ST:51 EF:10 CL:FFBF CH:FF02 SN:FF09 SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:DB00 Mon Feb 22 12:52:52 1988 HDERR ST:51 EF:10 CL:FFBF CH:FF02 SN:FF09 SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:D900 Mon Feb 22 12:52:56 1988 HDERR ST:51 EF:10 CL:FFBF CH:FF02 SN:FF09 SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:D900 Mon Feb 22 12:52:57 1988 HDERR ST:51 EF:10 CL:FFBF CH:FF02 SN:FF09 SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:D900 Mon Feb 22 12:52:58 1988 HDERR ST:51 EF:10 CL:FFDB CH:FF02 SN:FF0B SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:D500 Mon Feb 22 12:53:44 1988 drv:0 part:2 blk:41781 rpts:1 Mon Feb 22 12:53:50 1988 HDERR ST:51 EF:10 CL:FFBF CH:FF02 SN:FF09 SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:9100 Mon Feb 22 12:57:10 1988 drv:0 part:2 blk:39988 rpts:1 Mon Feb 22 12:57:10 1988 HDERR ST:51 EF:10 CL:FFD3 CH:FF02 SN:FF07 SC:FF01 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:9100 Mon Feb 22 13:18:14 1988 drv:0 part:2 blk:41267 rpts:1 Mon Feb 22 13:18:14 1988 HDERR ST:51 EF:10 CL:FFD5 CH:FF02 SN:FF08 SC:FF02 SDH:FF27 DMACNT:FFFF DCRREG:97 MCRREG:9100 Mon Feb 22 13:21:51 1988 drv:0 part:2 blk:41396 rpts:1 Mon Feb 22 13:21:51 1988 These entries are from a day a few weeks after they started. I also learned on that fateful Monday morning that our building had lost air conditioning during the weekend. Back then, our guards were closing the doors to the rooms and my 3B1 shares the office with a LaserWriter. Since the block numbers varied so much, I had assumed that my problem was an intermittent electronic failure due to overheating. So I called the Hot Line, expecting them to interpret the log reports and conclude that the mother board needed replacement, thus saving the contents of my 67MB disk. Unfortunately, he didn't know how to interpret the log entries, his "fix" was to try reinitialization of the disk. This is easy for him to say, but hard for me to do. There is an option in the diagnostics for recording a bad block in the bad block table that I would have tried if it was complaining about just one bad block. However, since it was complaining about many blocks and since the machine seemed to recover from the problems OK, I lived with it while I stewed about what to do. In the mean time, the battery failed. (Makes you wonder how long this machine, which I had had for less than 90 days, had set in a warehouse.) Thus I got a motherboard replacement anyhow, but the problem did not go away. Thus I broke down and again backed up all of my disk by datalinking the files at 9600 baud to my friendly, neighborhood comp. center machine, reinitialized the disk, and restored the files. The reinitialization did not report any bad blocks. So I did it twice to make sure. To my pleasant surprise, when I restored all of the files, I haven't had a disk log entry since. -- Dennis DeBruler (312)416-5182 (ihnp4!)ihdld!dld AT&T-BL IHP 1F-114
davek@heurikon.UUCP (Dave Klann) (03/21/88)
In article <4043@ihlpf.ATT.COM> dld@ihlpf.ATT.COM (Dennis L DeBruler) writes:
[Discussion about hard disk errors, and recovery...]
I had this same EXACT problem on my 20x1 7300 a few months ago. I DID try
to enter some of the "bad" blocks into the spares table (with the diagnostics
disk). The system just kept reporting more "bad" blocks. I finally solved
the problem the same way Dennis did (fortunately 20MB isn't as big a pain
to back up as 67MB :-) .
My questions: Why does this happen? Will it happen again? Can it be
avoided? What IS the true meaning of Life?
I'm not really expecting any answers, just trying to start a new topic for
discussion...
David Klann