Date: Sun, 23 Feb 1997 12:50:44 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: current@freebsd.org Subject: Interesting CVSUP failure Message-ID: <199702231950.MAA23263@phaeton.artisoft.com>
next in thread | raw e-mail | index | archive | help
I just resolved an interesting cvsup failure, which I think is indicative of a "nice" boundry failure in the FFS code. During an earlier cvsup, the disk where my tree is filled up during the transfer, and an out of range block was allocated to a file. I cleaned up the disk to get (a lot) more free space, but subsequent cvsup's all failed at the point of the allocation (obviously) and cvsup kept reporting a failure (the wrong one). I finally sat down to spend some time fixing the error, and turned on my console monitor, and noticed that the disk error was occuring. Using some brute-force FS debugging tools, I located the offending file, deleted it, and the cvsup ran to completion. Is there any chance that cvsup could report a more meaningful message for SIGABRT? In any case, there's also the boudry allocation problem in FFS on a hard disk-full (as opposed to a soft one from hitting the reserve) that should be looked into. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199702231950.MAA23263>