Intermittent terraform-provider-google-beta build failures (exit code 143)

[roaks3]

Background

The terraform-provider-google-beta "GoTest" GitHub Action frequently fails during the "Test" step with what looks like a SIGTERM (example https://github.com/hashicorp/terraform-provider-google-beta/actions/runs/5193230431). Note that this failure is intermittent:

Run make docscheck
  make docscheck
  make test
  shell: /usr/bin/bash -e {0}
==> Checking that code complies with gofmt requirements...
go vet
go test  -timeout=30s $(go list ./... | grep -v github.com/hashicorp/terraform-provider-google-beta/scripts)
?   	github.com/hashicorp/terraform-provider-google-beta	[no test files]
make: *** [GNUmakefile:1[2](https://github.com/hashicorp/terraform-provider-google-beta/actions/runs/5193230431/jobs/9363491219#step:6:2): test] Terminated
Error: Process completed with exit code 14[3](https://github.com/hashicorp/terraform-provider-google-beta/actions/runs/5193230431/jobs/9363491219#step:6:3).

There appears to be a known issue for some users of ubuntu-latest that looks similar to ours, where a resource limit is being hit, and the job is canceled:

• ubuntu-latest: jobs fail with error code 143 actions/runner-images#6680
• The runner has received a shutdown signal. actions/runner-images#7188

I was not able to find other reports of go test consuming an unusual amount of resources, but due to the large number of tests/files we use in this project, it seems possible that we are hitting a memory, file descriptor, or CPU limit.

Since this job actually just needs to run a small number of unit tests, one resolution might be to further limit the tests/files that need to be scanned, similar to hashicorp/terraform-provider-google-beta#3687.

What kind of contribution is this issue about?

• provider build

[rileykarson]

These appear to have been resolved by hashicorp/terraform-provider-google-beta#5733

[github-actions]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.
If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.