Domain Error... How to fix

[calguy1000]

Here's the story.

I have a VPS running ubuntu 8.0.4 LTS (of which I have root access). Originally this server was running ISPCP 1.0.0 but we recently upgraded it to ISPCP 1.0.7 due to load problems and the desire to migrate domains off of this server onto another ISPCP 1.0.7 server.

I am managing about 30 domains on this server.

The Server:
An Ubuntu 8.04 LTS install slightly modified:
- eAccelerator installed
- upgraded php files from dotdeb

The Upgrade Process
To say the least, it was painful. Even though the instructions I found on the ispcp wiki for upgrading an ubuntu server seem to be largely correct.

Because Ubuntu 8.0.4 comes with Perl 5.8 instead of Perl 5.10 I installed ActiveState perl 5.12 and a whole bunch of packages from cpan. including the berkely db, which I had to find an appropriate version of, and compile from scratch. Also the upgrade process removed our wildcard dns entries, and a few other customizations to the various configs that were easily restored. I think it was finally working properly.

Now to begin the migration process I used the backup-restore tools that are found in the contrib folder of the source package to test the migration of one domain, and the backup seemed to work.... HOWEVER the domain that I 'backed up' now has an 'error' on this server that I cannot resolve.

[u]The symptoms:[/u]
- The domain is no longer present in /etc/bind/named.conf though the file is present in /etc/bind/<domain>.db
- The web files are still present in /var/www/virtual/<domain>
- The domain is no longer present in /etc/postfix/ispcp/domains
- I receive an error when viewing this domain in the ISPCP web gui:
sys_command() | [ERROR] External command '/usr/sbin/postmap /etc/postfix/ispcp/domains' exited with value -1 ! (this line is present in the status field of the domains table of the ispcp database)
- The status of the domain is error, and clicking on it only displays the details page.
- I do not see any valid information in the ispcp log files (most of the files are empty infact)

The Question is:
- How do I restore this domain to working order (as I have not yet migrated it to another server).

Any help would be very appreciated.

Thanks in advance.

[joximu]

Hi

execute the command manualy:

/usr/sbin/postmap /etc/postfix/ispcp/domains

and see what's the problem, solve the problem, set the status flag of the domain to change (or toadd) and rerun the request manager...

/J

[calguy1000]

Once I manually add the domain back in to that file it succeeds. but does not resolve the error in the database or restore the dns files.
Is there a way to do that?

(07-09-2011 01:31 AM)joximu Wrote:  Hi

execute the command manualy:

/usr/sbin/postmap /etc/postfix/ispcp/domains

and see what's the problem, solve the problem, set the status flag of the domain to change (or toadd) and rerun the request manager...

/J

[joximu]

either goto the debugger page in adminpanel (as admin) and reset the status field - or do this manually (status -> "toadd" or "change" instead of the error message)...
Then rerun the request manager.

/J

[calguy1000]

The debugger page in admin panel seems to have done it.

Thanks very much.

(07-09-2011 07:50 AM)joximu Wrote:  either goto the debugger page in adminpanel (as admin) and reset the status field - or do this manually (status -> "toadd" or "change" instead of the error message)...
Then rerun the request manager.

/J