Closed gcheon closed 4 years ago
This has most commonly been observed to be caused by the user command running out of memory. There is a bug opened for the Pipelines API team to improve the error message a bit.
Suggest to try using a VM with more memory.
Makes sense, thank you for the fast reply!
Some of my jobs fail with the status message:
Execution failed: action 4: waiting for container: starting container: running ["docker" "start" "-ai" "9c5ba104cf1fe58325c9905a11de15cd2d03dc6dee636e51b48f536865a07765"]: exit status 255
When I check the logs from these jobs, it seems like the program I was running on docker was running fine and printing outputs until the job died, and the logs just stop without any error messages.
Do you have any clue what might be causing this? Also - is there any way to keep the docker container alive for debugging after the dsub job dies?