ruoxuan95 and djoahi have reported that in their runs of E2E tests for the CRW 2.2 workspaces (cpp, perl, python, etc.) the tests often fail to deploy CRW or fail to start a workspace due to image pull timeouts.
So... I'm wondering if we can use the image puller (or image puller operator - see CRW-875) to precache all the CRW 2.x images in the target cluster, THEN run the E2E tests.
Maybe it would make sense to run the puller once a day to ensure that you have fresh(ish) CI images to test?
WDYT dnochevn rhopp@redhat.com ?
Latest CRW 21.4 image puller:
https://quay.io/repository/crw/imagepuller-rhel8?tab=settings