Fresh install master-slave setup problem

[sotero]

Hi,

I'm setting up a fresh installation of a master-slave setup, on my masyer I have only network and usersandgroups modules, on one of my slaves which will be mail server I have only network, usersandgroups, and mail, but as soon as I activate mail module, If I reboot my machine it stops responding. Tryed this several times, and it always stops responding after I activate mail module, anyone with this problem?

Is the problem caused because mail server is on a slave? I have two other slaves, a file server and a web server, all working fine.

[alvinquah]

Hi,

I'm setting up a fresh installation of a master-slave setup, on my masyer I have only network and usersandgroups modules, on one of my slaves which will be mail server I have only network, usersandgroups, and mail, but as soon as I activate mail module, If I reboot my machine it stops responding. Tryed this several times, and it always stops responding after I activate mail module, anyone with this problem?

Is the problem caused because mail server is on a slave? I have two other slaves, a file server and a web server, all working fine.

I have the same problem as well. It is actually not responding but instead in the background it is running S20ebox script which is consuming 100% of your resource. There is a also bug that disable slapd daemon which is required to communicate with the master server. Hope they have solve it by now. Which userandgroup version you are using? Zentyal has released a new version couple of days ago, 2.02.

[haliparotin]

hi! what can i do to update my Zentayal 2.0 to 2.0.2? thank you...

[alvinquah]

hi! what can i do to update my Zentayal 2.0 to 2.0.2? thank you...

sudo apt-get update, i think what will do...

[haliparotin]

i try that but they have a problem.. the error is NO_PUBLIC KEY... they have something wrong to my server?

[alvinquah]

try this before sudo apt get upadte.

sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 10E239FF

[haliparotin]

i try that but this the error again.. HTTP fetch error 7:couldnt connect: connection time out.. but my server is have an internet...

[alvinquah]

i guess they are having issue with the server now...

[haliparotin]

ok.. but theres have any way to solve this?

[alvinquah]

Tried the latest userandgroup modules. As promised, now users/groups addition and deletion works like a charm! nice....

But i am still facing what the thread starter has encountered. Server stuck at splash screen.... Any advice. Thanks

[sotero]

I updated to 2.0.2 aswell, but the mail server keeps getting stuck at splash screen.

[exekias]

Hi alvinquah and sotre:

Did you activate PAM option in LDAP settings? More info would be nice in order to fix that issue

Thanks for your feedback

[sotero]

Yes, I have PAM enabled, default shell, Bash.

Thank you.


EDIT: Just disabled pam, and configured mail server again, but, again, when I activated mail module, on the slave, the machine hangs.

[UdoB]

Did you activate PAM option in LDAP settings? More info would be nice in order to fix that issue

(Joining the thread here)

I have a clean test environment with a Master and a Slave. I did enable Groupware/Zarafa. Booting the Slave does not finish.

This behavior is the same independent of the presence of the master. Even when the Master is powered down I get something like this:

Code: [Select]

 CPU[||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||100.0%]     Tasks: 48 total, 3 running
  Mem[|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||273/433MB]     Load average: 2.65 1.90 0.80
  Swp[                                                               0/311MB]     Uptime: 00:02:58

  PID USER     PRI  NI  VIRT   RES   SHR S CPU% MEM%   TIME+  Command
 1091 ebox      20   0 83336 64540  3984 R 37.0 14.5  0:46.30 /usr/bin/perl /usr/share/ebox-usersandgroups/slave-sync
 1057 ebox      20   0 47352 40212  4048 R 37.0  9.1  0:10.77 /usr/bin/perl /etc/rc2.d/S20ebox start

The memory usage increments by 1 MB per second (!) until all memory (including swap) is used...

I will run "apt-get --purge remove ebox-zarafa && apt-get autoremove" now.
 
15 minutes later: after doing so I see again "slapd" running on the Slave. In the above situation there was no slapd.



Next round: I did install Zentyal Components "Web Server" and "Email Service" (Approx. 44 packages).

Code: [Select]

  CPU[||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||100.0%]     Tasks: 32 total, 2 running
  Mem[||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||             228/433MB]     Load average: 2.09 1.68 0.71
  Swp[                                                                                 0/311MB]     Uptime: 00:02:39

  PID USER     PRI  NI  VIRT   RES   SHR S CPU% MEM%   TIME+  Command
  898 ebox      20   0 92236 70648  4036 R 73.0 15.9  0:58.27 /usr/bin/perl /etc/rc2.d/S20ebox start
So the problem is not inside "ebox-zarafa" but within "ebox-mail".


Best regards

[alvinquah]

Hi alvinquah and sotre:

Did you activate PAM option in LDAP settings? More info would be nice in order to fix that issue

Thanks for your feedback

Hi cperez,

PAM in my case is not enabled at all in all my troubleshooting. So i dun think it is the issue with PAM. Thanks