From: "Martin J. Bligh" <mbligh@aracnet.com>
To: linux-kernel <linux-kernel@vger.kernel.org>
Subject: [Bug 439] New: 2.5.63-bk5-7 IDE disk hangs with multiple disks
Date: Tue, 04 Mar 2003 21:32:45 -0800 [thread overview]
Message-ID: <140840000.1046842365@[10.10.2.4]> (raw)
http://bugme.osdl.org/show_bug.cgi?id=439
Summary: 2.5.63-bk5-7 IDE disk hangs with multiple disks
Kernel Version: 2.5.63-bk7
Status: NEW
Severity: normal
Owner: alan@lxorguk.ukuu.org.uk
Submitter: adam@yggdrasil.com
Hardware Environment: At least two IDE hard disks
Software Environment: linux-2.5.63-bk7
Problem Description:
If I do "dd if=/dev/zero of=/dev/discs/discN/disc bs=65536"
under linux-2.5.63-bk5 or bk7, dd seems to write to the disk for
a couple of seconds and then hangs. Oddly, sync will return, but
the process remains stuck in the "D" state and an attempt to strace
the process will hang strace (the strace process can then be killed,
but not the original dd process).
I have also observed this with other block sizes. However,
on one of the computers that I tried, a dd with a transfer size of
512 bytes seemed to go on indefinitely (actually, I only let it run
for about a minute, and the computer had 2GB of RAM), but the disk
activity light turned off after about a second, and would not even
flash on when I did a "sync", at least as far as I could tell.
Also on this system, if I pull a big rsync, the rsync will
hang in a "D" state after transfer about a megabyte if and only if
there is a second IDE disk. The second disk in this case is on a
different IDE cable, but the same controller. I have observed this
with three different models of Maxtor IDE hard drives on two different
P4 motherboards from different manufacturers, but both with Via
chipsets (sorry for so much similarity between the two cases).
I am using IDE in default pio mode because of problems with
IDE modules that I experienced about a month ago (loading the
appropriate DMA-based IDE module would immediately result in a kernel
crash of some kind, but I haven't tried this recently).
I will investigate this further after I've upgraded to 2.5.64,
but that will probably be tomorrow, so I thought I ought to report
this in the meantime.
Steps to reproduce:
Let's assume that you have a disk that you want to erase at
/dev/discs/disc1/disc (you must have at least two disks in the system).
You would do:
dd if=/dev/zero of=/dev/discs/discs1/disc bs=65536
reply other threads:[~2003-03-05 5:22 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='140840000.1046842365@[10.10.2.4]' \
--to=mbligh@aracnet.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®