MSP-Greg

> Since there's a successful test run with Ruby 2.7 via GitHub Actions Do you have any links for that? I'd like to see what test commands it was running....

@ashmaroli > Regarding the failures (runs with ruby-head) you have highlighted, all of those are timezone related mismatch between (DST ??) Windows and posix and may need some fudging of...

Seems reasonable. Currently, code is: ```ruby workers = ENV['WEB_CONCURRENCY'] ``` Is the following what you'd like it updated to? ```ruby workers = if (t = ENV['WEB_CONCURRENCY']) == 'auto' && defined?(::Concurrent)...

In this case, the 'code' changes are simple. I haven't written any tests yet. Ideally, we load `concurrent-ruby`, but I don't really want to load it in the main test...

@tannakartikey Thank you for your work on this. I think this should be merged for v7. Question for you and everyone else. When any repo has a major release, people...

Thanks. I think I reviewed the code a while ago, today I just double checked the 'conversation'. Or, I missed that. Sorry. I'll mark as reviewed and wait a few...

@tannakartikey 1. Please see https://github.com/MSP-Greg/puma/commit/d1a88c5c5ea9dc96decfba560b1be7d46c499654. It adds a method to standardize things. Any thoughts? 2. I had errors in `test_config.rb`, lines 488 & 489 need `on_` changed to `before_`. Confused...

This is the systemd notify socket, so it probably isn't the same issue as libev failures. The teardown code might be amiss, as the server might still be shutting down...

Re the branch, see https://github.com/MSP-Greg/puma/actions/workflows/tests.yml?query=branch%3A00-aa-test-1 The only failure was a TruffleRuby job that passed, but just missed the 8 minute timeout... **EDIT:** I'd like to see the retries count lower,...

> the systemd integration is trying to send stuff on the socket when the server is shutting down Not totally sure, but the test seems stable after I removed the...