[3722] in Release_7.7_team

home help back first fref pref prev next nref lref last post

Re: netscape coredump can cause IRIX panic

daemon@ATHENA.MIT.EDU (John Hawkinson)
Thu Jan 30 03:34:53 2003

Date: Thu, 30 Jan 2003 03:34:47 -0500 (EST)
Message-Id: <200301300834.DAA19366@multics.mit.edu>
To: "t. belton" <tbelton@MIT.EDU>
CC: Jonathon Weiss <jweiss@MIT.EDU>, bug-infoagents@MIT.EDU,
        release-team@MIT.EDU
In-reply-to: "[17641] in Athena Bugs"
From: John Hawkinson <jhawk@MIT.EDU>

Now that limitting the coredumpsize to 0 is part of the default
dotfiles, can we please remove it from the mozilla and netscape
scripts? Some of us would like to get core dumps and it's annoying
that "unlimit coredumpsize" is insufficient, instead one must use a
custom mozilla wrapper.

--jhawk
p.s.: "mozilla 1.2.1 seems to core dump for me occasionally,
 but of course I can't tell you how or why or when."


| I thought there was already a limit coredumpsize 0 in the wrapper script,
| all platforms. I'll check and see; if not I'll add it.
| 
| -Todd
| 
| On Wed, 8 Mar 2000, Jonathon Weiss wrote:
| 
| > 
| > Hi Todd,
| > 
| > We've discovered that IRIX has a bug that can cause a panic when
| > programs run out of AFS try to coredump onto local disk.  SGI is aware
| > of the problem, but it will probably be a long time before they have a
| > fix available.  Since netscape is more prone to coredumping than a lot
| > of things we'd appreciate it if you could add somethign to the
| > netscape wrapper to limit coredumpsize to zero.  Arguably this could
| > be done for all platforms, since it's not like people around here are
| > debugging the coredumps, but setting it up for IRIX should reduce the
| > number of people getting hurt.
| > 
| > 	Thanks,
| > 	Jonathon
| > 
| > 
| 
| --[17641]-- (pref = [17640], nref = [17654])

home help back first fref pref prev next nref lref last post