Executive summary: when BC deployed on OpenShift pod (re)Start, artifacts in the BC's .m2 are not reckognised, despite being shown in the UI

Details:
For context, I am using BC deployed on OpenShift v.311 from rhpds.
At a certain point, I had the aws machine stopped; I issued a rhpds start request, and I gained back again access to the openshift web console of my rhpds cluster.
Any other app deployed on my rhpds cluster was working just fine, but it was not possible to access BC via the usual http:// url as linked from the openshift web console which remained unvaried, with the webpage generated by openshift saying everything was running as normal but the pod was not accessible.
Back to the openshift web console, I scaled pod to 0 of the BC, and then back to 1 again.
Then the BC url was correctly reaching BC and I could log back in into BC.
I changed one of the bpmn asset in my project and I wanted to Deploy button to have it updated on the kieserver, but the Build failed on BC as from the messages it was not finding a required gav dependency, which was previously uploaded to the BC's .m2.
The only way I could have the project's dependency recognised, was to manually upload it again into BC, despite said dependency was correctly displayed already into the BC's UI for the .m2
Uploading the dependency again to the BC's .m2, going back to the project, clicking again the Deploy button, the Build this time succeeded.

I was able to capture on video as this occurred with the same behaviour while the BC's Project had actually 2 dependencies, to save bandwidth is a Red Hat GDrive link:
https://drive.google.com/open?id=1BtqInOc6ddBWZ6KCcn7p5NwDhJyhrtw4