Date: Mon, 18 Dec 1995 14:59:05 -0700 From: Nate Williams <nate@rocky.sri.MT.net> To: Terry Lambert <terry@lambert.org> Cc: hlew@genome.Stanford.EDU (Howard Lew), questions@FreeBSD.org Subject: Re: undump program Message-ID: <199512182159.OAA24671@rocky.sri.MT.net> In-Reply-To: <199512182105.OAA12378@phaeton.artisoft.com> References: <Pine.SUN.3.91.951218122743.25311A-100000@vegemite.Stanford.EDU> <199512182105.OAA12378@phaeton.artisoft.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > Actually, I think a better question would be.... Is there an undump > > program to take a core dump and run it? > > Core dumps wouldn't be core dumps if they were runnable. Unless you specifically designed your program to dump core and be ran afterward. ;) > The big problem with a core dump is that the condition that caused the > dump to occur exists in the state of the dump as an imminent problem > after an undump. > > The next big problem is that you can't reopen the fd's associated with > the fd's the process had open, since they file names aren't part of > the state information. Don't do anything in your program which has to be re-run everytime the program is re-started. FWIW, the original 386bsd->FreeBSD 1.0 upgrade script consisted of two un-dumped perl scripts. It was easier to provide a binary which did the upgrade rather than provide perl and the script both, so I had perl dump core and then found a version of undump on the net and compiled it for FreeBSD. However, that was 2 years ago, so I doubt I'd be able to find it. However, you might try archie or lycos and see if you can find it. Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199512182159.OAA24671>