CI Error Tracker
This issue is for reporting problems with the GitLab CI testing that seem unrelated to a particular test but are caused by problems with the CI or other things in the repository. For example tests with memory corruption may usually pass the test and thus have gotten into master but occasionally cause "random" errors for pipelines. Other errors such as running out of swap space on the test machine or losing network connectivity etc. may occur. You can rerun particular runs when the error seems to be a system glitch to see if the problem persists (if that works also report it here).
These errors should be
- reported here so we can detect patterns and fix the bugs in the PETSc code or improve the test infrastructure to reduce system errors.
- listed in your MR/issue message with something like "This is related to CI (#951)", your MR/issue will then automatically appear in the #951 discussion here, which helps to track what's going on at the moment.
Please search for your given problem and add it to the thread if it has been previously reported or add a new thread otherwise.
Note that not all errors are a result of GitLab CI, most errors are in the branch; make sure your error is not directly caused by your branch before reporting it here.
This is a continuation of #360 (closed)