Should we keep Jenkins results around longer?
Ported from https://github.com/dotnet/roslyn/issues/6337
We have a few PRs on dotnet/roslyn that have been around long enough that the Jenkins results have been dropped. Is the dropping of build/test results intentional? Is it based on elapsed time or some fixed number of results per queue? Ideally we'd keep results around forever, right? Are we running into storage issues?
This is a storage issue right now. Hopefully will be resolved this weekend, when we move to a 7TB raid array. Right now we're on a 1TB disk for data. We may be able to keep results around a while (3-6 months?), but probably not artifacts.
I think we came to decision that we'll do this on per job/Repro basis. Here's the bug that's tracking it for Roslyn: https://github.com/dotnet/roslyn/issues/8553.