The box was configured to do auto-updates, and it updated bind9 and overwrote the configuration files, and promptly failed to restart and has not restarted since. The 2 old config files bind.conf and bind.conf.local were backed up during the upgrade however even when restored it still fails to start. I suspect that more was overwritten that was not backed up somehow. The odd things is that it isn’t writting a failure notice in either syslog or messages. Any suggestions at this point would be most helpful. Thanks.
Is BIND seeing the zone files? I had 1 system (out of several) where BIND wouldn’t start after the update.
It was because the config file was looking in a different directory as were the zone files were.
It was looking in /var/named/chroot/var/named but the zone files were in /var/named
All i did was make a copy of the zone file and pasted it in the chrooted directory.
also under webmin/BIND module, there is a ‘check BIND config button’, did you try that already?
I did indeed ran check-config. Odd thing is that it started answering requests again after I turned off bind for all the sites, and then turned them on again. Weirdest thing I’ve seen in a while.
by switching bind on/off for the sites, it may have recreated the file-zones in the correct place.
You could check to see for double entries in the directories, though the path is different on Ubuntu, i think /var/bind instead of /var/named