Frequent provisioning/capacity issues



What should our expectation be for execution time of the setup_remote_docker step be when working w/ the docker executor? I love all of the work around docker as a first class citizen in 2.0, but the remote docker env is proving to be fairly cumbersome. We’re seeing more frequent failures with errors along the lines of Got error while creating host: VM creation failed, and if it doesn’t fail, it always takes on the order of 1-2 minutes, when it is successful. Is that within the expected tolerance for execution time on that step? Is the circle team working on more reliable provisioning and capacity preparation to ensure both: (1) consistent and reliable timing and (2) less prone to failure?


There’s a bug right now causing it to be over a minute or timeout entirely. We’re deploying a fix today that will bring it back to 10-20 seconds.


Thank you, thank you, thank you, @rohara! I appreciate the quick response. If I could make one request, please consider broadcasting these types of issues, rather than waiting for users to report it.


All of our builds are back down to a nice, reliable 10 second remote docker setup time, so the fix is much appreciated.


This topic was automatically closed 41 days after the last reply. New replies are no longer allowed.