samirmajen
samirmajen
Yep, I have the exact same problem. I'm using the latest spark-operator (v1beta2-1.2.0 with Spark 3.0.0) and trying to run the py example - executor is defined but does not...
looks like this change https://github.com/pixie-io/pixie/pull/1601 fixes the artifact tracker issue, cluster is still showing the "updating" state and the vizier operator now coredumps: goroutine 1 [running]: k8s.io/client-go/discovery.convertAPIResource(...) external/io_k8s_client_go/discovery/aggregated_discovery.go:88 k8s.io/client-go/discovery.convertAPIGroup({{{0x0, 0x0},...
i get the same issue - any solution to this ?
Thanks @olivierpilotte I've also been battling with bit docker for a while, unfortunately i don't think this project is fit for purpose.
I managed to get it stable in docker (so should work for you in k8s) - i used your changes above and also commented out anything that needs an outbound...
@olivierpilotte this might solve your issue: https://github.com/teambit/bit-docker/issues/7
Try adding this to your deployment spec: ``` spec: containers: - name: test stdin: true tty: true ```