An API request to a specific build_id for our project can take anywhere from 1-10 seconds (seems to be pretty uniformly distributed). I’ve noticed a huge circle_yml field in the response, which is several 10s of KBs. I’m only interested in the output logs for each step, is there any better way to fetch these with the API?
Hello,
It’s unsupported and may change in the future, but you can use the endpoint we use to display the logs to the frontend.
https://circleci.com/api/v1.1/project/:vcs-type/:org/:project/:build_id/output/:step_number/:container_number?circle-token=<user_token>
The step numbers start at 100 for 2.0 jobs
1 Like
Is there anything like this that doesn’t use the container number? The context is I’m trying to get information about all the build steps and the corresponding S3 log URLs, but the only info available for each job is the vcs, org, project, build number.
Ah, sorry to be confusing. The container number is 0 in all cases unless you are using parallelism, in which case it starts at 0 and goes up to how ever many containers you are using for that job.
Does that make it easier?
1 Like
This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.