home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- From: Ken.Wallace@f111.n440.z2.fidonet.org (Ken Wallace)
- Path: sparky!uunet!pipex!demon!purplet!f111.n440.z2.fidonet.org!Ken.Wallace
- Subject: Dying IDE disk drive.
- Organization: FidoNet node 2:440/111 - A6 BBS, Harpenden Herts UK
- Date: Sun, 13 Dec 1992 02:29:00 +0000
- Message-ID: <4.2B2BBAC1@purplet.demon.co.uk>
- Sender: usenet@demon.co.uk
- Lines: 34
-
- Hello Eric!
-
- Saturday December 12 1992 23:36, Eric Youngdale wrote to All:
-
- EY> It seems that I am usually answering questions, but now the tables
- EY> are turned. This morning my IDE disk died. Not completely, but I get all
- EY> kinds of HD errors, and an occasional HD reset, followed by a kernel panic
- EY> because it is unable to read the inode table. All of this came about
- EY> unexpectedly with no warning while I was in the middle of using the
- EY> system. The thing that is really queer is that two of the partitions on
- EY> the disk are just fine. Unfortunately these are my DOS and my swap
- EY> partitions - my root and home partitions are the two that seem to be
- EY> afflicted. I can do an od on both the dos and swap partitions and I get
- EY> no errors. The HD error message says that the error code is 0x10, the
- EY> status is 0x59. The file hdreg.h says that 0x10 means ID_ERR, whatever
- EY> that is.
-
- Sounds like a problem I've had a couple of times resulting in the partition
- table looking something like:
-
- hda1 1 1 250 ...
- hda2 1276321924649%90
- hda3 1920809809 903 90
- hda4 500 500 600
-
- deleting the last three partitions and creating them with the original
- figures again fixes it, then fsck (with -av because you get screens and
- screens of 'block xxxxxx marked busy is not in any file') brings the
- filesystems back. Like you, I could see no reason for it and everything
- seemed to work ok afterwards - hasn't happened to me for a while now so
- I assumed it had gone away (I still make a paper copy of the partition
- table though!)
-
- Ken
-