While doing some Drupal installation stuff I was chown'ing a directory recursively. This was a big mistake as I didn't notice that I had changed back to the root directory. Now my ISPconfig, DNS (BIND9), MySQL, Password Keyring, and who knows what else doesn't work. Apache2 is still working but I can't log in to anything, Webmin is still working. Can someone tell me or point me to the proper file and directory owners for ISPConfig, BIND, MySQL and the Keyring? I'll give you my first born in return, or a beer.
You mean you changed everything recursively starting form / or starting from the website root /var/www ?
From / Not everything was changed but apparently enough to cause problems. Mainly with the servers. If anyone can email directory listings showing groups and owners from a Debian or similar system that would probably give the information I need. Just a thought. [email protected]
Doing the following fixed the DNS (taken from Falko's install instructions)... chown -R bind:bind /var/lib/named/var/* chown -R bind:bind /var/lib/named/etc/bind But the MySQL and ??? still won't start up. Trying to start MySQL puts the following in the daemon.log... Code: Apr 8 09:33:00 host mysqld_safe[27123]: started Apr 8 09:33:00 host mysqld[27126]: 090408 9:33:00 InnoDB: Started; log sequence number 0 17056275 Apr 8 09:33:00 host mysqld[27126]: 090408 9:33:00 [ERROR] Can't start server : Bind on unix socket: Permission denied Apr 8 09:33:00 host mysqld[27126]: 090408 9:33:00 [ERROR] Do you already have another mysqld server running on socket: /var/run/mysqld/mysqld.sock ? Apr 8 09:33:00 host mysqld[27126]: 090408 9:33:00 [ERROR] Aborting Apr 8 09:33:00 host mysqld[27126]: Apr 8 09:33:00 host mysqld[27126]: 090408 9:33:00 InnoDB: Starting shutdown... Apr 8 09:33:02 host mysqld[27126]: 090408 9:33:02 InnoDB: Shutdown completed; log sequence number 0 17056275 Apr 8 09:33:02 host mysqld[27126]: 090408 9:33:02 [Note] /usr/sbin/mysqld: Shutdown complete Apr 8 09:33:02 host mysqld[27126]: Apr 8 09:33:02 host mysqld_safe[27189]: ended Apr 8 09:33:14 host /etc/init.d/mysql[27391]: 0 processes alive and '/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping' resulted in Apr 8 09:33:14 host /etc/init.d/mysql[27391]: ^G/usr/bin/mysqladmin: connect to server at 'localhost' failed Apr 8 09:33:14 host /etc/init.d/mysql[27391]: error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)' Apr 8 09:33:14 host /etc/init.d/mysql[27391]: Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists! Apr 8 09:33:14 host /etc/init.d/mysql[27391]:
Progress... This got MySQL started Code: touch /var/run/mysqld/mysqld.sock chown mysql /var/run/mysqld/ chown mysql /var/run/mysqld/mysqld.sock /etc/init.d/mysql start Then chown recursively to mysql in /var/lib/mysql got most of the databases working. Still some odd errors though.
VMWare is hosed, synaptic doesn't work, can't get a root terminal (says wrong password), can do sudo though...
Most likely the esaiest way might be to do a complete reinstall. If thats not possible, you should consider to setup he same linux version on another PC or virtual machine and then compare the permissions of the folders. Personally I would do a reinstall as I expect that it will be shorter then trying to get the correct permissions again for thousands of files and directories.
Thanks. It took more than a month after the last time I did a complete reinstall to get all the applications and drivers installed correctly and everything tweaked. I'll just keep using the system in it's crippled form rather than do that again. Most things are back working. I'll reinstall the applications that aren't working and see if that helps.
