Installing aborts with error message

[dirkschoening]

Hello everyone,

I'm fairly new to the whole cluster management game, so please excuse me if the questions seem a bit basic.

I'm currently setting up a small test cluster with TrinityX, 1 controller and 5 nodes. The controller is running Rocky 9.5, the whole thing is set up as described in the documentation. During the installation, while Ansible Playbook is running, the following error message appears and the installation aborts:

TASK [trinity/init : Verify internal controller interface is configured] *************************************************************************************
fatal: [controller1]: FAILED! => {"changed": false, "msg": "Interface with config set in 'trix_ctrl1_ip' (10.141.255.254) or 'trix_ctrlX_ip' (, ...) not found. Please make sure the interface is configured and up"}

Have any of you ever had this error? Does anyone have any ideas as to what could be causing this error?

[bartlamboo]

Hello @dirkschoening

This would mean that you haven't manually configured your network interfaces yet.
The trinityx installer doesn't do this for you, but it does check if the addresses you've configured in the all.yaml have been assigned to an interface.

Once you add the configured ip addresses to the correct interface, all should be fine when you re-run the playbook.

[aphmschonewille]

If the host/controller name matches the config for ctrlX in the all.yml file and the trusted interface has been correctly set as well (also all.yml file), the needed interface will be configured free of charge. There is a limitation: the playbook can then only handle this fallback when one trusted interface has been set. The logic is: set when you can based on irrefutable evidence.
But as already mentioned, it's better to have all involved interfaces configured before running the playbook.

-A

[dirkschoening]

Hello @dirkschoening

This would mean that you haven't manually configured your network interfaces yet. The trinityx installer doesn't do this for you, but it does check if the addresses you've configured in the all.yaml have been assigned to an interface.

Once you add the configured ip addresses to the correct interface, all should be fine when you re-run the playbook.

Hello @bartlamboo

thank you very much, that helped :)