Release Regressions for 8.8

[jbclements]

Here are the current results for checking release regressions:

• crypto-test (test-fail) : https://plt.cs.northwestern.edu/release-pkg-build/server/built/test-fail/crypto-test.txt
• typed-compose (build-fail) : https://plt.cs.northwestern.edu/release-pkg-build/server/built/fail/typed-compose.txt

[sorawee]

crypto-test is due to #4532. We agree to revert the PR from both release and master.

[jbclements]

Oh, master too? Sounds good to me. In that case I can revert on master and cherry-pick it to the release, which feels nicer to me than reverting it on the release only and probably missing it in the next release.

[sorawee]

Yep, master too.

[samth]

Yes, I think we're going to abandon the hash-table changes entirely and make something new.

[jbclements]

@scolobb It looks like typed-compose is failing on the release build with an out-of-memory error during a package build. It also looks like the package is not failing on the master branch. Do you have any idea what's going on here?

[samth]

This is very unlikely to be @scolobb's fault. The release-pkg-build often has more timeouts than the regular pkg-build, since it's running more things in parallel and is on a different machine.

[jbclements]

I see. In that case, I'm going to check that one off as being complete. Let me know if you think that's not the right thing.

[jbclements]

Separately, I'm thinking we should wait until the release-pkg-build finishes next to make sure that crypto-test is now passing.

[scolobb]

I hope @samth is right, because I don't have any idea how to go about debugging the kind of issue you are pointing out @jbclements 😄

[jbclements]

Yep, latest release-pkg-build shows crypto-test as no longer failing.

[jbclements]

Closing this as complete.