Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 20 Aug 1999 23:32:35 +0100
From:      David Malone <dwmalone@maths.tcd.ie>
To:        freebsd-stable@freebsd.org
Subject:   Re: vm_fault: pager read error on NFS filesystems.
Message-ID:  <19990820233235.A21737@walton.maths.tcd.ie>
In-Reply-To: <19990820201827.A22891@lanczos.maths.tcd.ie>; from David Malone on Fri, Aug 20, 1999 at 08:18:27PM %2B0100
References:  <199908171256.aa44534@salmon.maths.tcd.ie> <19990820201827.A22891@lanczos.maths.tcd.ie>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Aug 20, 1999 at 08:18:27PM +0100, David Malone wrote:

Some experimentation produced the following results. I created an
executable and then either truncated or removed it on the NFS server
while is was running. If you catch the signal you are sent you get
a different effect (equivelent of a double fault I guess).

3.2-STABLE, as of about 3 weeks ago.
====================================
truncate		Panic (it was trying to dump a core for the process).
remove			vm_pager error and sigfault.
catch sig + remove	vm_pager error loop - may be killed by kill -9.

4.0-CURRENT, also about 3 weeks ago.
====================================
truncate		process stuck in vompar while handling the pagefault.
remove			vm_pager error and sigfault.
catch sig + remove	vm_pager error loop - process can't be killed.

I tried quickly on a Redhat 6 box, catching the signal and changeing the
execuatble results in the process spinning and also being unkillable.

I still haven't managed to get init stuck in vmopar - it isn't entirely
clear to me why it would be interested in the processes text at all.

	David.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990820233235.A21737>