Date: Sun, 23 Aug 1998 06:03:21 +0900 From: Tetsuro FURUYA <ht5t-fry@asahi-net.or.jp> To: jdunn@aecinfo.com Cc: Tetsuro FURUYA <tfu@ff.iij4u.or.jp> Subject: Re: wd0 problems Message-ID: <199808222103.GAA01699@galois.tf.or.jp> In-Reply-To: Your message of "Wed, 19 Aug 1998 16:30:00 -0400 (EDT)" References: <Pine.BSF.3.96.980819162642.1054A-100000@unix.aecinfo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi,
You seems to have trouble in your EIDE drive.
There are so many mails in here and there.
But, I will restated some prescription here.
(I am sorry being so late to respond you , I am not so good at english,
and cannot read and write ml everyday.)
In Message-ID: <Pine.BSF.3.96.980819162642.1054A-100000@unix.aecinfo.com>
"Julian C. Dunn" <jdunn@aecinfo.com> wrote:
> This morning I came into the office to find that one of the FreeBSD
> boxes had this line scrolling across its screen:
>
>
> swap_pager: indefinite wait buffer: device 131073, blkno: 688, size: 4096
This statement was generated in /usr/src/sys/vm/swap_pager.c.
One process was accessing disk, and at that time, disk I/O error
occurred. After that, swap_pager want to access disk, but cannot
perform disk access, and doing nothing but waiting,
and the system hangged up.
> I restarted it, and it now gets to the following point:
>
>
> Automatic reboot in progress...
> /dev/rwd0s1a: CLEAN FLAG NOT SET IN SUPERBLOCK (FIXED)
> /dev/rwd0s1a: 23066 files, 215974 used, 240409 free (38738 frags, 25209
> blocks, 8.5% fragmentation)
> wd0: interrupt timeout:
> wd0: status 58<rdy,seekdone,drq> error 0
> wd0: interrupt timeout:
> wd0: status 58<rdy,seekdone,drq> error 1<no_dam>
>
>
> and then the system hangs, with the hard disk making seek noises over and
> over. Is this an indication of a hard drive failure?
When EIDE disk sectors were broken, these error messages often appear.
In this case, your disk sectors must be broken, and your wd driver
is mal-functioning.
The prescription is already known.
In order to handle bad secters, and as you cannot pass fsck,
once you should start system in single-user mode (boot(8)) and reboot.
Reboot will set clean flag in superblock, so you can mount file system again.
Next boot should start at single-user mode,
and mount / /usr /var read-only.
you should use
bad144(8) (scan mode is recomended),
or badsect(8) and fsck.
At this time, you had better use /usr/sbin/bad144 for scan mode.
And reboot the system again.
Of course you should read on-line manual very carefully.
Now the kernel source code.
After rebooting, You have to search /usr/src/sys/i386/isa/wd.c
for lines such as,
du->dk_timeout = 1 + 3;
and increase the value 1 + 3.
And compile and install kernel again.
In doing so, it would be helpful to install kernel debugger ddb(4) to
your kernel. When system hangs while fsck or bad144,
you can invoke kernel from system console by typing Cntrl-Alt-Esc.
And by this operation, you can avoid further bad blocks.
Please see to it that X ignores this keys.
And after the seeking sound of disk stops, type "continue", this
will make timeouts function forcibly, and several attempts will
recover system.
If system fails to recover, type "panic" to ddb prompt.
This will reboot system rather safely.
* Auto reboot for a panic can be installed with ddb.
There are several fine idea that the failure of the EIDE disk
is caused by the heat or the unstable power supply.
Tetsuro, Furuya. ht5t-fry@asahi-net.or.jp
========================================================================
TEL: 048-852-3520 FAX: 048-858-1597 ||
E-Mail: 8==------
ht5t-fry@asahi-net.or.jp , tfu@ff.iij4u.or.jp * ||
pgp-fingerprint: \|/
pub Tetsuro FURUYA <ht5t-fry@asahi-net.or.jp>
Key fingerprint = F1 BA 5F C1 C2 48 1D C7 AE 5F 16 ED 12 17 75 38
=========================================================================
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808222103.GAA01699>