1.0.7 to 1.1.0 beta upgrade issues

[pgentoo]

Hey guys,

My server was running on 1.0.7, or a slight patched version of that, with dovecot support and roundcube.

I attempted an update to 1.1.0 using the nightly update script from trunk, which had some problems, so I went ahead and did the manual update procedure using the 1.1.0 beta zip from sourceforge.

This last attempt had many issues as well, as after installation all of my apache virtualhosts and postfix aliases/mailboxes/domains/etc were all blank.

I got everything manually up and running again, but would like to get ispcp fixed and back into a working state. The problem i think starts that when setting all the domains/aliases/etc to "change" in the database, they had errors during processing which I believe cascaded into the rest of these issues.

If i login as admin to the control panel, and view the debugger, it shows that i have 747 requests waiting and 47 database errors. It lists out one error from each category, allowing me to set it to change. All of the erros are like:

otice: Undefined variable: TR_DOMAIN_MESSAGE in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291
Notice: Undefined variable: TR_ALIAS_NAME in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291 - Set status to 'change'
Notice: Undefined variable: TR_ALIAS_ERROR in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291
Notice: Undefined variable: TR_SUBDOMAIN_ALIAS_NAME in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291 - Set status to 'change'
Notice: Undefined variable: TR_SUBDOMAIN_ALIAS_ERROR in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291
...etc.

It seems like there are a lot of these template values which aren't populated, but i'm not even sure where they should be coming from.

Can someone please help me out and point me to the place where i need to configure these, and maybe some list of the variables i need to load and sample values?

I'd really like to get this back up and running ASAP, so your prompt reply is appreciated.

Thanks!

---

Bit more detail. If i look in the database at a domain record, the status shows this:

get_tag() | [ERROR] '; dmn [ispcp.mydomainhere.com] timestamp entry BEGIN.
' ne '; dmn [ispcp.mydomainhere.com] timestamp entry END.
', '; dmn [ispcp.mydomainhere.com] timestamp entry BEGIN.
' or '; dmn [ispcp.mydomainhere.com] timestamp entry END.
' missing in src !

Hopefully that helps narrow down the issues.

[pgentoo]

So i should have mentioned. My server died, and i have restored it from backup. It was on a copy of 1.0.7 with a few patches, but after the restore i didn't get ispcp working properly. I just replaced all the config files, and got the database back online with the old data, and figured i'd run an upgrade to 1.1.0 instead of trying to get 1.0.7 working.

It looks like the issue i encountered was that the bind conf files weren't matching the expected values in the database. Is there a way i can make ispcp rebuild these? If I manually go through and create a zone file (mydomain.com.db) in /etc/bind/ispcp/, and then copy it to the /etc/ispcp/bind/working/ directory, reset the domain to "change" and then rerun the request manager, it works. Just a tedious process of manually fixing each of these entries.

Seems like there is probably a way to say "please rebuild ALL my configs from the database, because I know they are not good".

I'll get this going manually for now, as there are only a few errors so far, but please chime in if there is a way to force regeneration of all the configs.

Thanks!

---

Ok, i made it a ways through this, but i'm stuck now. I set all my tables to "change", like done in the nightly update script, but when i run the request manager afterward, all my domains process, however all of my other tables (subdomain, domain aliases, mailboxes, etc...) are all still in the "change" state. I don't see any errors in the logs that would indicate that something went wrong.

Any advice is appreciated...

[pgentoo]

Finally following back up on this as I need to get this server back up and running 100%. I think the problem is that the bind zones are not 100%, and its breaking the update process. Is there any way (script, etc) that is available to rebuild all these files so that I can let the request manager process all the changes?

Any advice is appreciated...

---

When i run the ispcp-rqst-manager, it processes one item, but if i look at the debugger i see stuff like this for each type of item:

Notice: Undefined variable: TR_ALIAS_NAME in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291 - Set status to 'change'
Notice: Undefined variable: TR_SUBDOMAIN_NAME in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291
Notice: Undefined variable: TR_MAIL_NAME in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291 - Set status to 'change'
Notice: Undefined variable: TR_HTACCESS_NAME in /var/www/ispcp/gui/include/Smarty/sysplugins/smarty_internal_data.php on line 291 - Set status to 'change'

---

looks like this bug: http://isp-control.net/ispcp/ticket/1963

Anyone have a root cause?

[ephigenie]

These error messages are due to the translation system.