Add Openshift provider: Credential validation was not successful: 503 "Service Unavailable"

Support
#1

Hi,

I tried the demo version of ManageIQ with the Docker image and everything was ok.

I deployed the VMware appliance and perform same operations. Adding a git repo and credentials for Ansible was ok. But adding OCP provider return Credential validation was not successful: 503 “Service Unavailable”.

In the logs

[root@localhost ~]# grep -e "ocp" /var/www/miq/vmdb/log/evm.log
[----] I, [2020-11-13T03:21:02.448713 #2502:2abf98e192f4]  INFO -- : MIQ(ManageIQ::Providers::Openshift::ContainerManager#with_provider_connection) Connecting through ManageIQ::Providers::Openshift::ContainerManager: [ocp4exp.testing.local]
[----] W, [2020-11-13T03:21:02.777157 #2502:2abf98e192f4]  WARN -- : MIQ(ManageIQ::Providers::Openshift::ContainerManager#authentication_check_no_validation) type: [:bearer] for [] [ocp4exp.testing.local] Validation failed: error, 503 "Service Unavailable"

No more info about what could be the error.

Please help me Obi Wan, you’re my only hope.

#2

Is the hostname & port correct? Is the hostname reachable from the VMWare that runs ManageIQ?

Can you oc login into the cluster using this hostname & port?

Guessing here, but 503 might be a response from the openshift router, it’s what it says when there is no Route matching the given domain name/path. If so, it means you reached the cluster with correct IP address, but not correct HTTP request for the apiserver.

Which openshift version is this?

#3

Yes hostname and port ok.
And yes it’s reachable. I tested to login directly via oc command from the appliance.
It’s ocp 4.6.

No prob with the same config in the docker miq image and in the openshift based installation.

#4

I fixed it by doing a reset of the database with the appliance_console.
It works now with the same configuration.