Date: Wed, 10 Dec 1997 20:57:53 -0600 (CST) From: Kevin Day <toasty@home.dragondata.com> To: jkh@time.cdrom.com (Jordan K. Hubbard) Cc: freebsd-current@FreeBSD.ORG Subject: Re: 971208 and the nfs rollback didn't fix it... Message-ID: <199712110257.UAA22490@home.dragondata.com> In-Reply-To: <24108.881808195@time.cdrom.com> from "Jordan K. Hubbard" at "Dec 10, 97 06:43:15 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
> > Does anyone really feel the need to investigate this? I realize userland > > programs should never crash the kernel, but... is this a big enough deal for > > someone to want to investigate? > > Yes, I'd say it is. Is this 2.2.5 or -current, however? > > Jordan > -current. I have two machines running the exact same setup. (I installed one, then duped the hard drive, so they are identical, software wise). Totally different hardware though... Both machines can reproduce the crash. It doesn't work on a 2.2.1 machine I have though. Essentially, here's the best I can pin it down. A user has eggdrop1.2.0 semi-installed on here. Somewhere along the way he lost his eggdrop binary. He sets up botchk to run in his crontab anyway. Botchk executes his conf file, which should bomb immediately... It executes /home/chronos/egg/eggdrop1.2.0/Pipeline, and nstead it either randomly reboots the machine, or garbles the first 5 letters in the file. bash$cat /home/chronos/egg/eggdrop1.2.0/Pipeline #!/home/chronos/egg/eggdrop1.2.0/ # ^- set that to the directory eggdrop is in ie "#!/users/lamest/egg/eggdrop" # # This is a sample configuration file for your bot. You will definitely # want to edit this, to set up your bot. Right now it creates a bot called # "LamestBot" which sits on channel #lamest. turns in to: Az^MDme/chronos/egg/eggdrop1.2.0/ # ^- set that to the directory eggdrop is in ie "#!/users/lamest/egg/eggdrop" # # This is a sample configuration file for your bot. You will definitely # want to edit this, to set up your bot. Right now it creates a bot called # "LamestBot" which sits on channel #lamest. If it doesn't scramble the file, it jumps straight to the ram test. No panic message, nothing. If it *does* scramble the file, that user can't do anything from that point on. All he see will see is 'fork: resource temporarily unavailable' until I reboot the machine. Does this make sense to anyone? Kevin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199712110257.UAA22490>