Server crash - is recovery possible?
I'm running a RH 7.0 box at home that does (a) NAT for the local network (b) mail server for multiple domains (c) web server for multiple domains and (d) DNS for multiple domains Yesterday I attempted to ssh login from a workstation and got a message "..exited with Signal 11". I attempted to log on from the console and could not - got screen full of messages and numbers. At this point at least the web server was still running. I hit reset to reboot, got all the standard messages until "Checking root filesystem...." About 5-10% in, then: "Warning ..fsck.ext2 for device /dev/hda2 exited with signal 11 [FAILED] " Looked in my configuration book this morning ... hda2 is "/" . The only fortunate thing is that /home and /var are separate partitions on a second hard drive. Is there any way to recover from this? Barring that - is there any way to recover any config files from various places in /etc on this partition. If not I guess its a complete reload - mainly a problem with mail and DNS config files. I am only mildly competent on this stuff. If you need more info, I'll get it, although I don't know how to capture that screen full of info at the abort point. Dick Richard Goodman (508) 757-3452 r.goodman@11harvard.com 11 Harvard St, Worcester MA 01609
On Mon, Aug 26, 2002 at 09:02:20AM -0400, Richard Goodman wrote:
Yesterday I attempted to ssh login from a workstation and got a message "..exited with Signal 11". I attempted to log on from the console and could not - got screen full of messages and numbers.
About 5-10% in, then: "Warning ..fsck.ext2 for device /dev/hda2 exited with signal 11 [FAILED] "
Is there any way to recover from this? Barring that - is there any way to recover any config files from various places in /etc on this partition. If not I guess its a complete reload - mainly a problem with mail and DNS config files.
The hard drive is probably ok, but I would look at my hardware first. Lots of signal 11 (segmentation fault) errors usually indicates bad memory. See: http://www.bitwizard.nl/sig11/ -- Randomly Generated Tagline: "As a competitor, we might be better off if they shipped it." - Sun Microsystems's James Gosling on Windows NT 5.0
Thanks, Theo I read the link, and went home and swapped out a bad SIMM. Everything seems to be up and running. The only anomoly is that after a shutdown completes, I seem to get a similar memory segment error (although not Signal 11). However its been so long since I've done a shutdown that I don't remember if this is new - I can live with it. Dick At 09:40 AM 8/26/2002 -0400, you wrote:
Yesterday I attempted to ssh login from a workstation and got a message "..exited with Signal 11". I attempted to log on from the console and could not - got screen full of messages and numbers.
About 5-10% in, then: "Warning ..fsck.ext2 for device /dev/hda2 exited with signal 11 [FAILED] "
Is there any way to recover from this? Barring that - is there any way to recover any config files from various places in /etc on this
On Mon, Aug 26, 2002 at 09:02:20AM -0400, Richard Goodman wrote: partition. If
not I guess its a complete reload - mainly a problem with mail and DNS config files.
The hard drive is probably ok, but I would look at my hardware first. Lots of signal 11 (segmentation fault) errors usually indicates bad memory.
From: Richard Goodman <r.goodman@11harvard.com>
I read the link, and went home and swapped out a bad SIMM. Everything seems to be up and running. The only anomoly is that after a shutdown completes, I seem to get a similar memory segment error (although not Signal 11). However its been so long since I've done a shutdown that I don't remember if this is new - I can live with it.
I have a RedHat 7.0 machine that always spits out an error message just after printing "Power down" and before turning its power off. It goes by quickly and I have never really tried to fix it because at my school there was no class on how to debug software running on a machine with no power and I have never learned the skill. It seems to be the last words of a process that is not properly killed by shutdown but dies when it finds the Operating System missing. It may be related to the RAID disk. It seems harmless. -- Keith
participants (3)
-
Keith Wright -
Richard Goodman -
Theo Van Dinter