Hello, I have an issue with moved ISPC to a new machine (DB only) and then db-dump-imported on newely intalled ISPC (2.2.33) that doesn't create folders. I looked into ispconfig.log, but after adding some site it gives out only: Code: 04.11.2009 - 13:20:19 => INFO - Signalfile Set: insert 04.11.2009 - 13:22:32 => INFO - Signalfile Set: update 04.11.2009 - 13:22:47 => INFO - Signalfile Set: update 04.11.2009 - 13:23:07 => INFO - Signalfile Set: insert 04.11.2009 - 13:24:02 => INFO - Signalfile Set: insert 04.11.2009 - 13:24:32 => INFO - Signalfile Set: insert 04.11.2009 - 14:57:02 => INFO - Signalfile Set: update 04.11.2009 - 15:04:28 => INFO - Signalfile Set: insert Any idea where to start the debugging this issue? Thanks!
Thanks Till for suggestion. After restarting daemon here are WARNs that I found in log: Code: 04.11.2009 - 16:40:18 => WARN - /root/ispconfig/scripts/lib/config.lib.php, Line 2667: WARNING: could not . /etc/apache2/envvars && httpd -t &> /dev/null 04.11.2009 - 16:40:18 => WARN - /root/ispconfig/scripts/lib/config.lib.php, Line 2681: httpd syntax seems to contain errors, reloading with old configuration 04.11.2009 - 16:40:21 => WARN - /root/ispconfig/scripts/lib/config.lib.php, Line 2166: WARNING: could not groupadd -g 12003 web3_anonftp &> /dev/null 04.11.2009 - 16:40:21 => WARN - /root/ispconfig/scripts/lib/config.lib.php, Line 2167: WARNING: could not useradd -d /var/www/web3/ftp -g web3_anonftp -m -s /bin/false -u 12003 web3_anonftp &> /dev/null what could first two rows mean? the second two - this user exists, how to fix this? Thanks!
I've just noticed that after experiencing that problem the DNS records are also not being added, as ISPC2 is skipping all other steps when the first isn't working... As a matter of first 2 errors is it possible that apache2.conf is bad or it has to do with users? Any suggestions?
Do you use the same distribution on the old and the new server? What's the output of Code: . /etc/apache2/envvars && httpd -t on the new server?
Yes, the same distribution (Debian), but newer version (etch > lenny).. Thanks for pointing me to this command! I've run it as normal user and it gave me error that root directory of one of vhosts isn't existing (/var/www/web21), but it was present, just it had 700 permission so it was inaccesable to apache user. After that I rerun the Code: . /etc/apache2/envvars && httpd -t as normal (or root) user and I become no error in shell. But, unfortunately, this error is still present in ispconfig.log when i try to update something in panel.
As of Bind issue: Code: # /etc/init.d/bind9 status bind9 is not running failed! but when I run: Code: ps aux|grep bind root 3676 0.0 0.0 3868 588 pts/1 S+ 16:25 0:00 grep bind bind 22120 0.0 0.9 282860 20844 ? Ssl 16:19 0:00 /usr/sbin/named -u bind -t /var/lib/named -S 1024 It seems to be running as a named, not bind or what?
I think could be a scenario: ISPConfig tries to create a new Vhosts_ispxx file it fails because generated one has some errors in it >> WARN#1 then it puts back old working Vhosts file >> WARN#2 Till, Falko, am I right? But now, how to debug it? How to see this non-working generated and then deleted file and try to find out what is wrong in it?? Thanks!
Yes. there is nothing deleted. The file is stored with the date and time appended to its name in the vhosts directory. For debugging, see here: http://www.howtoforge.com/forums/showpost.php?p=175961&postcount=6
Hoooray!! Thanks Till! I found it on my own hard way before reading your reply - I went to config.inc.php, disabled the httpd checking that left the unworking vhosts file and so I debuged it! As for the bugs in Vhost file, leftovers from old server: Code: @one host: php_admin_value open_basedir /var/www/web3:/tmp:/var/www/www.somedomain.net @other host: php_admin_value open_basedir /var/www/web10/:/tmp Can you help me out how to set open_basedir for specific sites that require it (using suPHP now)?