Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 6 Mar 2001 15:44:06 -0500 (EST)
From:      Andrew Gallatin <gallatin@cs.duke.edu>
To:        Christian Weisgerber <naddy@mips.inka.de>
Cc:        freebsd-alpha@FreeBSD.ORG, phk@FreeBSD.ORG
Subject:   Re: Dump segv w/devfs
Message-ID:  <15013.19478.880920.81093@grasshopper.cs.duke.edu>
In-Reply-To: <20010306210654.A25183@kemoauc.mips.inka.de>
References:  <9806fj$1si$1@kemoauc.mips.inka.de> <Pine.LNX.4.21.0103050923450.16541-100000@zeppo.feral.com> <15011.59405.333814.964428@grasshopper.cs.duke.edu> <98328r$f38$1@kemoauc.mips.inka.de> <15013.11144.757425.205222@grasshopper.cs.duke.edu> <20010306210654.A25183@kemoauc.mips.inka.de>

next in thread | previous in thread | raw e-mail | index | archive | help

Christian Weisgerber writes:
 > Andrew Gallatin:
 > 
 > > OK, it seems to be related to devfs.  Dumping with /dev/null as a file
 > > crashes.  Dumping to a normal file elsewhere works.
 > 
 > No, it doesn't. 
 > I originally noticed the problem when my backup script, which runs
 > 
 >     dump 0af /holding/disk $filesystem
 > 
 > failed.  And this problem still persists.
 > (/holding/disk is just a normal file on an FFS.  Different disk,
 > same SCSI adapter.)
 > 
 > Dumping to /dev/null is merely another convenient way to reproduce
 > this.
 > 
 > > Dumping through a redirect to /dev/null (tcsh) seems to work as
 > > well.
 > 
 > Fails as well.

I've only seen it once & cannot make it happen again.

I've had dump in a while 1 loop for 2 hours.  I can't get it to crash.

Since nobody else can reliably reproduce this but you, can you please
build dump with symbols and get a coredump, or run it under gdb?

Drew

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15013.19478.880920.81093>