Hi,

I'm performing my MAS 8 installation only for the sake of knowing how to do it and provide a learning environment to users and developers used to Maximo 7.X

I managed to install OpenShift in Single-Node and install the code portion of MAS using the Maximo Application Suite Automation operator in OpenShift

According to these steps:

https://github.com/cloud-native-toolkit/operator-masauto/blob/main/docs/MAS-Operator-Deployment.pdf

I must now run the initialSetup and upload the license I had me generated.

1) Navigate to https://admin.inst1.apps.mas8.domain/initialsetup

2) Download and add the certificate to the browser Trust Store, when I go back to that URL or the api.inst1.mas8.domain, the URLs show as safe and trusted

3) I enter username and password found from the inst1-credentials-superuser secret in OpenShift, seems to be working

4) I get a blue spinning circle forever

My browser console is full of these errors :

initialsetup:1 Access to XMLHttpRequest at 'https://api.inst1.apps.mas8.domain/schemas/workspaces' from origin 'https://admin.inst1.apps.mas8.domain' has been blocked by CORS policy: No 'Access-Control-Allow-Origin' header is present on the requested resource.

---

One last thing I need to mention is to get anything openShift related accessible from my Windows laptop (the OpenShift is running inside a VMWare workstation) I edited my Windows hosts file with a bunch of entries ; don't know if anything is missing to get the initialsetup up and running

192.168.2.100    api.mas8.domain
192.168.2.100    oauth-openshift.apps.mas8.domain
192.168.2.100    console-openshift-console.apps.mas8.domain
192.168.2.100    grafana-openshift-monitoring.apps.mas8.domain
192.168.2.100    thanos-querier-openshift-monitoring.apps.mas8.domain
192.168.2.100    prometheus-k8s-openshift-monitoring.apps.mas8.domain
192.168.2.100    alertmanager-main-openshift-monitoring.apps.mas8.domain
192.168.2.100    admin.inst1.apps.mas8.domain
192.168.2.100    api.inst1.apps.mas8.domain
192.168.2.100    auth.inst1.apps.mas8.domain
192.168.2.100    auth.inst1.apps.mas8.domain/login
192.168.2.100    home.inst1.apps.mas8.domain
192.168.2.100    masdev.home.inst1.apps.mas8.domain
192.168.2.100    auth.inst1.apps.mas8.domain/initialSetup

----------

Any clues would be appreciated!

Thanks!

Francois

Francois,

I ran into this same issue and after 30 minutes to an hour it just resolved itself.  I am not sure what process hadn't completed, but apparently there must be some registration process that occurs to ensure the proper CORS headers.

I would give it an hour and come back to see if it hasn't done the same for you.  Maybe Steven Shull will have a more technical explanation as to what the internal process is.

Jason

Jason,

Thanks for the reply.

Actually I finished "core" installation yesterday morning, and it's been like that ever since.

The only culprit I can see with my limited knowledge is that 2 pods or not working, they are crashing and restarting

The 2 "ins1-coreapi" are the only 2 pods that are failed or running with 0 out of X container, all other pods seem fine

I attached the log from the first of the failing pods

Thanks again for your help, much appreciated

Attachment(s)

pod log.txt   5 KB 1 version

Have you tried killing those pods and seeing if they get recreated?  I realize this is not exactly helpful information, but I have found for transient errors this is sometimes the answer, perhaps someone with more detailed knowledge of the install process can provide a better response.

Yes I did. If I keep them alive (like the have been for a while) they auto-restart by themselves and fail again, and we can see the count of restart going up.

The 3 pods are out of a ReplicaSet

And they fail over and over. That being said I don't even know if this is related to my main issue which is the setup page will not load and the CORS error I am seeing, but I will try to the pods issue as maybe the root cause for now

This is not a bug or any issue. You just have to accept the "api" URL's certificate to move forward from the spinning circle. I have it documented it here (see Accept Certification section) https://mam-hol.eu-gb.mybluemix.net/ocp_8.9/post_deployment/

Note that you won't encounter this situation if you use public certificates like Let's Encrypt. 

I did. Now both the api and admin URL once I hit them they show as secured because i imported the certificates in the trust store.

I don't know which one I installed though ; are the certificates different if I exported when I was on the admin URL compared to the api URL?

I might just remove all certificates I imported and try again by going to initial setup first, replacing the admin for api, and getting that certificate and import it