Univention Bugzilla – Bug 53433
"connection to module process failed: setup" at startup and failure to launch
Last modified: 2022-08-02 02:10:09 CEST
Overview: When I try to start UCS after installation, I get the error message "connection to module process failed: setup", after which I get a loading screen that doesn't seem to go anywhere. Steps to reproduce: 1) Install UCS in a VM. I've had the same results in VirtualBox and virt-manager. Pick the default options. Alternatively, use a preinstalled VM-image. Usually I used the minimum requirements for the VM but I've also used up to 1.5 GB RAM and 20 GB HDD and encountered the same error. 2) Start the VM. After some initial loading, the error message appears. Actual results: After clicking the error message away, the system appears to be loading. However, even after waiting for several hours, nothing comes up. Expected results: Initial configuration should start. Build Date & Hardware: Build 2021-06-8 on Ubuntu 20.04.2 LTS x86_64
I forgot while I marked this for UCS 4.4, I encounter the same bug on 5.0.
the message "connection to module process failed: setup" means that the UMC server could not connect in 10 seconds to the system setup backend process. most often this happens because of not enough free RAM. more details might be available in /var/log/univention/management-console-module-setup.log and /var/log/univention/management-console-server.log.
Thank you for the help. If this is a known problem, maybe you should change the minimum requirements in the quick start guide and everywhere else where they are too low. The quick start guide states a minimum requirement of 1 GB memory but I had the error with any value I tried below 3 GB. Is there a public repo with the docs? Because if there was, I couldn't find it.
It should work flawless with 2 GB. https://docs.software-univention.de/quickstart-en-5.0.html says: "The system requirements vary considerably depending on the intended purpose and the number of users. The minimum requirements for the installation are 1 GB memory and 8 GB hard drive space."
Version: 4.4-8 errata958 (Blumenthal) Error: Connection to module process failed: setup Role: None
Created attachment 10902 [details] screenshot of error
I have just experienced this bug. I built four brand new virtual machines with the ISO for v5. The intent being to have one primary DC, two replicas, one backup. A pretty standard production setup. On boot, all four exhibited the error. I rebooted all four. Two of them worked post reboot. Two more do not. Googling led me to this bug.
root@unassigned-hostname:/var/log/univention# cat management-console-module-setup.log 13.01.22 20:43:37.875 DEBUG_INIT 13.01.22 20:48:36.189 DEBUG_INIT 13.01.22 21:01:32.842 DEBUG_INIT 13.01.22 21:07:34.688 DEBUG_INIT 13.01.22 21:15:34.537 DEBUG_INIT 13.01.22 21:31:11.727 DEBUG_INIT root@unassigned-hostname:/var/log/univention#
13.01.22 21:30:21.703 MAIN ( PROCESS ) : Server started 13.01.22 21:31:04.288 LOCALE ( WARN ) : Could not find translation file: 'umc-core.mo' 13.01.22 21:31:04.479 MODULE ( PROCESS ) : Setting auth type to None 13.01.22 21:31:04.507 LOCALE ( WARN ) : Could not find translation file: 'umc-core.mo' 13.01.22 21:31:06.150 MAIN ( PROCESS ) : running: /usr/bin/python2.7 /usr/sbin/univention-management-console-module -m setup -s /var/run/univention-management-console/824-1642131066149.socket -d 2 -l en_US.UTF-8 13.01.22 21:31:16.438 MAIN ( PROCESS ) : ModuleProcess: stopping 1086 ~
If I connect over the management interface from a remote browser, it works. It just doesn't work locally from the VM initial boot.
(In reply to Charles N Wyble from comment #7) > I have just experienced this bug. I built four brand new virtual machines > with the ISO for v5. The intent being to have one primary DC, two replicas, > one backup. A pretty standard production setup. > > On boot, all four exhibited the error. I rebooted all four. Two of them > worked post reboot. Two more do not. > > Googling led me to this bug. how much RAM does your VM has?
(In reply to Florian Best from comment #11) > (In reply to Charles N Wyble from comment #7) > > I have just experienced this bug. I built four brand new virtual machines > > with the ISO for v5. The intent being to have one primary DC, two replicas, > > one backup. A pretty standard production setup. > > > > On boot, all four exhibited the error. I rebooted all four. Two of them > > worked post reboot. Two more do not. > > > > Googling led me to this bug. > > how much RAM does your VM has? Hello, I can confirm I changed my VM ram to 2GBs. The .ova file defaulted to 1GB for some reason. Cheers,