Date: Sat, 12 Feb 2005 13:55:34 -0700 (MST) From: "Ryan Sommers" <ryans@gamersimpact.com> To: "Ashwin Chandra" <ashcs@ucla.edu> Cc: freebsd-hackers@freebsd.org Subject: Re: VMCORE NOT FOUND Message-ID: <2777.24.107.98.253.1108241734.squirrel@24.107.98.253> In-Reply-To: <003201c5113e$c56321e0$58e243a4@ash> References: <001901c510e1$aecc7c90$58e243a4@ash> <200502122218.42458.doconnor@gsoft.com.au> <003201c5113e$c56321e0$58e243a4@ash>
next in thread | previous in thread | raw e-mail | index | archive | help
Ashwin Chandra said: > I actually changed dumpdir to /usr/crash since /var did not have enough > space to accomodate 512 MB. But /usr has 60gigs. When the kernel panics > it > just crashes, I never get a message saying "dumping...". Savecore upon You dump to a device, not a path. Because the kernel is crashing file-system consistency can't be assumed. Therefore you can't dump to a file-system. Instead you dump to a device, typically the swap device. Then after you reboot and ensure ensure file-system integrity you can save the dump onto the file-system for analysis. I believe you are getting a little mixed up in all the terms: dump device: the device the kernel will save a dump to. (This is a device, not a file-system path) dump dir: the file-system path a core from the dump device will be saved to by savecore. The kernel does not dump to the dump dir. The kernel dumps to the dump device, the savecore utility retrieves a previous dump from the dump device and saves it in the dump dir. Like most things, this is outlined very well in the handbook: http://www.freebsd.org/doc/en_US.ISO8859-1/books/developers-handbook/kerneldebug.html#KERNELDEBUG-OBTAIN -- Ryan Sommers ryans@gamersimpact.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2777.24.107.98.253.1108241734.squirrel>