Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - segelfreak

Pages: [1]

I'm looking at Jabber module and was playing around a bit with it over the past days.

After having solved this issue (,34817.0.html) with the kind help of doncamilo, it still appears things are not as they should be with this module.
I basically unchecked all setting in the module admin page, with the exception of multi-user chat. Thus, no shared roster or vcard info has been enabled.
Now, when I add the an account to Trillian, I (still!) see a complete contact list, including users with deactivated jabber accounts.
Obviously the admin options are not taking effect and also the generated list is just picking all domain users.

How to patch this best way? Any ideas?

Hi All,

We have organized users in dedicated organizational units (i.e. containers), but unfortunately, jabber account settings are only available for "default container" users.
I really do not want to setup all users again (within the default container) in order to get them a jabber account on the server and apart from this I could only imagine now to setup duplicated accounts for jabber only.
None of them are really getting me excited... :-\

Any idea how to enable jabber accounts also for non-default container users?

Other modules / sa-learn / Spamassassin error
« on: December 14, 2019, 09:02:20 pm »

Spamassassin is giving me a hard time.
I'm trying to run sa-learn, but no matter what I try, I'm always getting an error. debug log shows me that there's a problem with the DB connection:

Code: [Select]
dbg: bayes: unable to connect to database: Access denied for user 'amavis'@'localhost' (using password: YES)
This also keeps the system from learning by feeding in some Bayes box files.

Annone having an idea how to get it working?

Other modules / struggeling to get virtual machine module working
« on: April 12, 2019, 09:05:07 am »
Dear All,

First of all: I'm a noob when it comes to KVM/qemu and I'm struggeling hard to get the virtual machine module working.
To be more precise:

  • I do get a new virtual machine running (starts successfully) with basic setup, NAT and just an live image of ubuntu server activated to keep things light and easy
  • when I open the console, it will give me just a very small window where I can just see the error "Server disconnected (code: 1006)"
  • when I try to connect via VNC, some unencrypeted connection will be established and I'm prompted for password. However, none seems to work, neither blank, or any of my login credentials.
  • I tried to play around a bit with the password in the mas files, but no success, either

The bottom line is that I can't see anything what's happening, subsequently can't install anything... any ideas?


I have a small problem, getting zentral server running in a virtual environment. Maybe someone can help me out...?

I'm going to use non-routable ip addresses in this "example", just to "mask" the real ip's, so you can ignore the obvious consequences form this, of course.

The virtual machine got an ip like, but the router is only I've tried to set the static ip with net mask, but then zentral will not accept any other gateway at all.
The only way I got it accepted was to assign it with net mask, in order to keep the gateway within range.
So far, so good. somehow awkward, but it works and the remote web admin is accessible.
Now, there's a mirror for updates at, which is in fact non-reachable, due to the fact that apt will try to reach the mirror without using the gateway. I've now tried to set a static route to the mirror via the gateway, but zentral complains that the default gateway is already set.
I can of course set a manual route via console command, but without changing the stubs it will be not permanent.
So, what I wonder now, am I thinking in a wrong way, or is there an issue with the network interfaces setting "rule checks", which will prevent me from setting up the network correctly?

Any ideas on how to overcome this in the best convenient way??


After having switched to 5.0, I can't see any shared contacts or shared addressbooks anymore. As a consequence, it seems I can also not add any users to the permissions, i.e. cannot share my calendars anymore.
Any solution at hand?

Installation and Upgrades / LOG for DHCP not working under 5.0 ?
« on: October 06, 2017, 05:27:59 pm »

Seems to me that DHCP Log is not collecting any data.
Can someone check and confirm?

Installation and Upgrades / grant AD users access to the web admin GUI
« on: September 09, 2017, 08:39:03 pm »

Having set up a new server with 5.0, I recognize that it seems impossible to grant users from the AD Administrator Access to the Web GUI.
In 4.2 one could simply add them under "System/General/Administrator Accounts" and then they were listed along with the local admin accounts.

Now in 5.0, I can still add them and they will be still listed, however, I notice that they are now listed with the domain part added (i.e. domain\username), but login will result in a "long password" failure.
I need to give the "domain admins" access to the Dashboard, so they can add users via the gui.

Anyone having a clue what needs to be done?

(PAM is enabled under LDAP Settings)

Dear All,

We use a 4.23 installation in a non-profit organization, unfortunately with some more people having access as administrators then I would have preferred. Now it appears that one of the admins did run the upgrade from 4.2 to 5.0 by "mistake" and it ended in a totally destroyed system. Maybe one shouldn't be too much surprised that it didn't work as straight as it should, especially since the upgrade was offered at all times in such a prominent way. But I'd never expected to have the system left in such a bad state after all.

Without having had the chance to have a deep look into the whole mess, it seems like even the distro upgrade from 14.04 to 16.04 did not complete well and effectively most of zentyals packages, incl. core, did not correctly upgrade, either.

Now its stuck in an endless circle of uncompleted configuration attempts (for zentyal-core), incomplete php5 replacements and whatever else.
Big mess. Non of the "usual" ways worked so far, although I tried not only apt-get -f install's, etc. to forced install or removal attempts. It's always trying to configure the core, returns an error and will break the upgrade at that point.

Now, unless someone has a better approach, I seem to have two options:
Install 4.2 again from scratch and use one of the backups we made from time to time.
Alternatively, try to bring the upgrade to an end, but I don't think it makes much sense to try to repair this destroyed linux deployment by hand, as there seem to be too many libraries effected. Instead I could save all important data, make a fresh 5.0 installation and restore as much as possible from any manually saved data.

Since I can't use zentyals backup module anymore, is there any source, guide, white paper or similar which I could us to manually backup the latest states, i.e. the AD "deployment" in any way?

As you can imagine, I'm pretty open to any suggestions which will help me to save as much as possible from the installation.
Easy to do for all home directories, etc.. I also don't care too much about firewalls, etc. but setting up the AD again from nothing would be painful. It's about 50 accounts...


Dear All,

I was wondering if someone could help me to understand the feature of activating openchange account for distribution/security groups?

What is the difference between a distribution group with or without having an open change account and how can I us the open change account when activated?

Thanks in advance!

Email and Groupware / fetchmail in a multidrop environment
« on: March 02, 2016, 11:23:31 pm »

due to some restrictive domain setting, all mail for a domain ist sent to only one single mailbox (catch all).

I would now like to retrieve the email and forward them to each dedicated user on the server.
The problem is that the emails do not have any x-envelope header, so this make things even worse as it would drop all email sent to more than one recipient in the domain.

Now I saw this example in the manual from fetch mail:

     user maildrop with pass secret1 to golux ’hurkle’=’happy’ snark here

This means, within one poll, I could catch all email and still send them to dedicated recipients.

Looking at the fetch mail.rc.mas file, I see that it's basically using a foreach instruction. so I thought of using this to make one poll with the "to" instruction and then it should add all local accounts, so it can parse each email in the mailbox.

Would this be possible and if yes, how would the .mas file would look like?

Or am I missing something?

Email and Groupware / write access to shared address book
« on: March 02, 2016, 10:46:07 pm »
Hi All,

I'm struggling to get the write access done right for the shared address book.
I simply want to complete the address info of the registered users, such as phone numbers, etc.

I added modifiers = (username); to the sogo.conf.mas and restarted openchange form the dashboard.
After this, the "save" button appears in the webmail fronted, when I try to change an existing entry for the shared address book. However, when I press save, it will close the windows but not update the info.
I also tried the superuser option and also got the save button, but still no success. Changes are simply not written back.

who can help?

Pages: [1]