Skip to content

Investigate Resource Issues in CI Tests #161

Description

@wulfraem

Description

When running CI tests with cargo nextest default build job count, the build process may or may not run into issues, that point towards consuming too many resources during build and fail.

Errors

In cases, where the CI tests are canceled as described above, they usually do so with the following error message in the Cargo nexttest step:

Logs in step:

Error: Process completed with exit code 143.

Also, this message has been observed in a case, where the test was not canceled, but ran for almost an hour:

The hosted runner lost communication with the server. Anything in your workflow that terminates the runner process, starves it for CPU/Memory, or blocks its network access can cause this error.

Current Solution

To mitigate the resource consumption, the build jobs have been limited to 2 by default, which seems to allow the tests to complete successfully.

Investigation

We should investigate, why we are actually running into these resource issues, and if we can mitigate them without limiting the build process' performance. As these issues seem to have started appearing with the bump of the IOTA dependencies from v1.15.0 to v1.20.1, it might be a good idea, if streamlining the features of our dependencies in that direction would be a solution and/or if we could improve in other aspects of our CI tests.

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't workingchorebumping dependencies, updating/adding CI steps, maintenance, etc.

    Type

    Projects

    Status
    Product Backlog

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions