Date: Sun, 26 Aug 2001 16:45:26 -0600 (MDT) From: Alex Varju <varju@antiflux.org> To: freebsd-stable@freebsd.org Subject: 4.4-rc instability Message-ID: <Pine.LNX.4.21.0108261603170.9348-100000@okcomputer.antiflux.org>
next in thread | raw e-mail | index | archive | help
Hi there, I have been running FreeBSD-stable for several years now, and things have been relatively reliable. A few days ago, both of my main machines started crashing intermittently. I had done a buildworld for each of them within the last two weeks, although they are not running from the same code snapshot. For the first machine, the errors first started as signal 11 core dumps for random processes. Things were bad enough that I decided a reboot might be in order. The next morning when I returned to the machine, it had panicked overnight, and the filesystem was damaged enough that fsck had to be run by hand. (I have soft-updates enabled). At this point I figured that I'd been unfortunate enough to grab a bad snapshot of the code, and went to update my kernel. A cvsup of the repository succeeded, but then my cvs update for the RELENG_4 branch caused another panic partway through. Needless to say, this left the filesystem in a less than happy state. Having been burned more than I was prepared to handle, I decided to pop my trusty old 4.3-release CD in and do an upgrade overtop of the existing installation. Partway through the upgrade, I got another panic. While I'm not sure about all of them, the most recent panics have all been in the filesystem code. The second machine started to go haywire about a day later. I was working in X at the time, and the entire system froze. Being unable to find any way to do a clean reboot, I was forced to reset the machine. When it came up again, it was unstable. It will run for a while, and then after a number of processes have been executed, things will start to randomly core dump. The machine does not appear to panic as often, though. When it dies, it gives errors like "page fault while running in supervisor mode". It should be mentioned that I have added new ram to each of these machines within the last month. Thinking about it now, it seems to me that the problems on the second machine could be attributed to this. It could be that it runs fine for a while, and then when the memory starts to fill up, it goes bad. I have just taken the extra chip out of that machine, and am going to try another installworld. I could be wrong, but I don't feel like the problems on the first machine are related to the memory. It seems almost like something has been corrupted in the filesystem. If this is the case, is there any solution other than formatting the drive? Thanks in advance, Alex. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.21.0108261603170.9348-100000>