Howto reset initial configuration?

[marOlon]

I have installed Zentyal without a problem.
But with the initial configuration wizard something got wrong. After selecting 'Gateway', 'UTM' and 'Unified Communications' and click on Install I've got the message "Unable to fetch some archives, maybe run apt-get-update or try with --fix-missing".

At that moment the install-cd was not in the drive. After putting in the cd I did a browser-back and tried again, and after 4 actions I got the message that the configuration was saved. But the only thing I get in the dashboard is Apache, nothing else.

After that I also tried to install the profiles again under 'Zentyal Components', but unfortunately with the same result.

How can I reset the configuration and run the wizard again?
I have searched in the documentation, the forum and on internet, but couldn't find any answer to that.

Please, can someone help me with this problem?
Thank you in advance.

[Sam Graf]

Hi,

To my knowledge, the only way to reset the configuration is to purge installed modules. But it sounds like only Zentyal's core was installed.

I do have a question: Did your machine have a working Internet connection during installation?

[ian]

Hello,

In terminal mode on the server use following command to reconfigre Zentyal:

sudo dpkg-reconfigure ebox

Best regards

[marOlon]

@Sam Graf
Thank you for your reply.

"I do have a question: Did your machine have a working Internet connection during installation?"
The machine was connected directly to my modem/router which has a working Internet connection.

However, I don't know if the initial configuration module could have made a network connection then, because at the moment there is even no network interface installed, only apache.

[Sam Graf]

During the early phase of the installation, was the Ubuntu/Zentyal installer able to get an address via DHCP?

The reason I ask is that it sounds to me like Zentyal is not able to fetch its modules from the Internet during install, leaving you only with the core installed from disc. You should be able to add or remove modules via Software Management even with just the core installed (assuming Ubuntu's initial configuration of the network hardware went well), but I think you need a working Internet connection. And it seems as if you don't.

So perhaps there is a hardware or driver issue preventing Ubuntu from setting up a working connection. That's why I wondered if it was successfully getting an address via DHCP.

[marOlon]

Hello,

In terminal mode on the server use following command to reconfigre Zentyal:

sudo dpkg-reconfigure ebox

Best regards

Thank you Ian.

I tried your suggestion and this is what happened:
- I was asked which port to use for https (443) which is already in use.
I accepted this and after that Apache was stopped and restarted, thats all.

[marOlon]

During the early phase of the installation, was the Ubuntu/Zentyal installer able to get an address via DHCP?

The reason I ask is that it sounds to me like Zentyal is not able to fetch its modules from the Internet during install, leaving you only with the core installed from disc. You should be able to add or remove modules via Software Management even with just the core installed (assuming Ubuntu's initial configuration of the network hardware went well), but I think you need a working Internet connection. And it seems as if you don't.

So perhaps there is a hardware or driver issue preventing Ubuntu from setting up a working connection. That's why I wondered if it was successfully getting an address via DHCP.

Thank you Sam.

My box gets an IP from the modem/router (I can ping the router and also another computer in the network).

I also tried the suggestion of Ian (see reply to him).
After that I tried again to install 'Gateway', 'UTM' and 'Unified Communications' (dashboard option Zentyal Components).
Result:
"The package installation has not finished correctly. More information on the logs." Same error as in my first post.

But I don't know in which log and what I must look for.

Do you have any further idea?

Thank you for your help.

[SamK]

Hi marOlon,
I cannot offer a direct solution but would like to ask which version have you installed?

The reason for asking is that I experienced similar symptoms when recently creating a test-rig using the 2.1-1 Beta version of the installer.  The cause could not be established and as there were no reports of it happening to others, it was put down as a one-off glitch.

I was fortunate that the problem occurred in a test set-up.  My answer was to completely wipe the disk (including the MBR) and conduct a fresh install.  This installation worked without errors.
 

[marOlon]

@SamK, Thanks for your post.
I have installed version 2.0 (core version 2.0.16).

In another post (not to similar) there was speaking of kind of clean-up, wipe-out, delete the Regis db.
Maybe that could be a solution for my problem?

However, I don't know how or where I could do that.

Greetings.

[SamK]

In another post (not to similar) there was speaking of kind of clean-up, wipe-out, delete...

However, I don't know how or where I could do that.

If you are asking how to wipe your disk, using a distro such as Parted Magic is one of the easiest ways.  You can get it from http://partedmagic.com.  It contains various disk wiping tools (and many other useful tools also).

Just be aware if you decide to conduct a disk wipe that it will remove everything on the disk so a re-install is required from scratch.

Also if the problem is not a corrupt installation but is a hardware or driver problem it may occur again in a post wipe installation.
   
   

[marOlon]

Hello all,

I have looked in some logs (from ebox-configuration-report).

Some notifications in ebox.log are:
2011/05/21 18:25:01 INFO> Progress.pm:128 EBox::CGI::Progress::loadAds - Try load ads from: 0
2011/05/21 18:25:01 INFO> Progress.pm:131 EBox::CGI::Progress::loadAds - Don't exist file /usr/share/ebox-software/ads/ads_C, load deafult ads
2011/05/21 18:25:03 INFO> ebox-update-packages:56 main::__ANON__ - Starting package installation process
2011/05/21 18:25:08 INFO> ebox-update-packages:116 main::__ANON__ - Package installation process finished
2011/05/21 18:25:25 INFO> Global.pm:535 EBox::Global::saveAllModules - Saving config and restarting services:
2011/05/21 18:25:29 INFO> Base.pm:151 EBox::Module::Base::save - Restarting service for module: events
2011/05/21 18:25:30 INFO> Base.pm:151 EBox::Module::Base::save - Restarting service for module: logs
2011/05/21 18:25:30 INFO> Base.pm:151 EBox::Module::Base::save - Restarting service for module: apache
2011/05/21 18:25:35 INFO> EventDaemon.pm:301 EBox::EventDaemon::_loadModules - EBox::Event::Dispatcher::Log loaded from registeredDispatchers
2011/05/21 18:25:35 INFO> ebox-apache-restart:52 main:: - Waiting for apache to shutdown, attempt 0
2011/05/21 18:29:16 DEBUG> ProgressIndicator.pm:102 EBox::ProgressIndicator::retrieve - Progress indicator with id 8842 not found
2011/05/21 18:29:16 ERROR> Base.pm:353 EBox::CGI::Base::__ANON__ - Exception: Progress indicator with id 8842 not found
2011/05/21 18:29:18 DEBUG> ProgressIndicator.pm:102 EBox::ProgressIndicator::retrieve - Progress indicator with id 8842 not found
2011/05/21 18:29:18 ERROR> Base.pm:353 EBox::CGI::Base::__ANON__ - Exception: Progress indicator with id 8842 not found
2011/05/21 18:29:20 DEBUG> ProgressIndicator.pm:102 EBox::ProgressIndicator::retrieve - Progress indicator with id 8842 not found
2011/05/21 18:29:20 ERROR> Base.pm:353 EBox::CGI::Base::__ANON__ - Exception: Progress indicator with id 8842 not found
2011/05/21 18:29:22 DEBUG> ProgressIndicator.pm:102 EBox::ProgressIndicator::retrieve - Progress indicator with id 8842 not found
2011/05/21 18:29:22 ERROR> Base.pm:353 EBox::CGI::Base::__ANON__ - Exception: Progress indicator with id 8842 not found
2011/05/21 18:29:24 DEBUG> ProgressIndicator.pm:102 EBox::ProgressIndicator::retrieve - Progress indicator with id 8842 not found
2011/05/21 18:29:24 ERROR> Base.pm:353 EBox::CGI::Base::__ANON__ - Exception: Progress indicator with id 8842 not found
2011/05/21 18:30:22 INFO> ebox-update-packages:56 main::__ANON__ - Starting package installation process
2011/05/21 18:30:25 INFO> ebox-update-packages:116 main::__ANON__ - Package installation process finished
2011/05/21 18:30:34 DEBUG> ProgressIndicator.pm:40 EBox::ProgressIndicator::create - Invalid value for Total ticks: 0.
It must be a non-zero positive number
2011/05/21 18:30:57 ERROR> Base.pm:346 EBox::CGI::Base::__ANON__ - Software caused connection abort at /usr/share/perl5/HTML/Mason/Request.pm line 126.

Stack:
  [/usr/share/perl5/HTML/Mason/Request.pm:126]
  [/usr/share/perl5/HTML/Mason/Request.pm:514]
2011/05/21 18:31:10 ERROR> Base.pm:346 EBox::CGI::Base::__ANON__ - Software caused connection abort at /usr/share/perl5/HTML/Mason/Request.pm line 126.

Stack:
  [/usr/share/perl5/HTML/Mason/Request.pm:126]
  [/usr/share/perl5/HTML/Mason/Request.pm:514]
2011/05/21 18:46:51 INFO> ebox-update-packages:56 main::__ANON__ - Starting package installation process
2011/05/21 18:46:55 INFO> ebox-update-packages:116 main::__ANON__ - Package installation process finished
2011/05/21 18:47:08 DEBUG> ProgressIndicator.pm:40 EBox::ProgressIndicator::create - Invalid value for Total ticks: 0.
It must be a non-zero positive number

Some notifications in ebox.log.d/error.log are:
[Sat May 21 17:21:13 2011] [notice] Apache/2.2.14 (Ubuntu) mod_ssl/2.2.14 OpenSSL/0.9.8k mod_perl/2.0.4 Perl/v5.10.1 configured -- resuming normal operations
Use of uninitialized value $reason in length at (eval 102) line 74.
[Sat May 21 17:21:22 2011] [error] server reached MaxClients setting, consider raising the MaxClients setting
Reconnecting to redis server (1 try)... at /usr/share/perl5/EBox/Config/Redis.pm line 567, <GEN3> line 1.
Reconnecting to redis server (1 try)... at /usr/share/perl5/EBox/Config/Redis.pm line 567, <GEN5> line 1.
Argument "C" isn't numeric in addition (+) at /usr/share/perl5/EBox/CGI/Progress.pm line 127, <GEN5> line 38.
Argument "/usr/share/ebox-software/ads/ads_" isn't numeric in addition (+) at /usr/share/perl5/EBox/CGI/Progress.pm line 127, <GEN5> line 38.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN5> line 48.
Reconnecting to redis server (1 try)... at /usr/share/perl5/EBox/Config/Redis.pm line 567, <GEN8> line 1.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN5> line 60.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN5> line 72.
Reconnecting to redis server (1 try)... at /usr/share/perl5/EBox/Config/Redis.pm line 567, <GEN10> line 1.
Reconnecting to redis server (1 try)... at /usr/share/perl5/EBox/Config/Redis.pm line 567, <GEN12> line 1.
Argument "C" isn't numeric in addition (+) at /usr/share/perl5/EBox/CGI/Progress.pm line 127, <GEN12> line 39.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN12> line 49.
Reconnecting to redis server (1 try)... at /usr/share/perl5/EBox/Config/Redis.pm line 567, <GEN15> line 1.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN12> line 61.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN12> line 75.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN12> line 127.
Use of uninitialized value in subroutine entry at /usr/share/perl5/EBox/Config/Redis.pm line 539, <GEN12> line 139.
[Sat May 21 18:25:35 2011] [notice] caught SIGTERM, shutting down
[Sat May 21 18:25:41 2011] [warn] Init: Session Cache is not configured [hint: SSLSessionCache]

... and more from error.log and errors in software.log (see attachements).

It has become a long list, sorry for that.
Maybe the logs can help to solve my configuration problem.

Thank you in advace.

[marOlon]

Hello Zentyal fans,

According to the errors in de logs (see attachements in my previous post -may 23-, specific software log) it seems that the initial configure part of core version 2.0.16 was somewhat corrupt. To get a clean system maybe I have to start overnew and get me core version 2.0.20.

Do you agree with that?

Thank you for your advice.

[SamK]

I am unable to comment on the contents of the logs but make more generalized observations.

If the original attempted installation failed due to hardware or driver based issues, (as mentioned in posts #4 and #9) it may occur again at the next attempt.

If the problem is some form of corruption, it may be easier and more reliable to start afresh rather than attempting to repair a severely broken installation.

Hope that helps.