Alex Willmer

> [ansible/ansible#78111](https://github.com/ansible/ansible/pull/78111) replaces that PR in ansible/ansible, with a slightly different mechanism Setting a class attribute `pipelining = False` is what I had concluded.

Notes to myself, - OpenSolaris: initial Open Source release of Solaris by Sun, discontinued by Oracle - Solaris Express, Oracle Solaris: proprietary releases - Illumos: Foundation/umbrella continuing OpenIndiana (fka OpenSolaris),...

> not sure what is causing these CI failures, but they don't seem directly related to (or unique to) this patch. They're due to #1058, almost certainly unrelated to your...

> @moreati Any other requests for this one now that it's passing checks? Yes. Two suggested changes are still open (see previous comments), and I want to answer "what are...

> The reason I asked is that I looked through the comments here and _believe_ that I have answered the questions or have addressed them in code. If you can...

I should have said, the changelog entry can use the PR ID. A seperate issue isn't required. Feel free to use either though.

This still occurs intermittently on Azure Devops. While experimenting with Github Actions I'm seeing it reliably fail as follows, in py36-mode_mitogen-distro_centos7 ``` test_unknown (utils_test.CastTest) ... ok test_run_with_broker (utils_test.RunWithRouterTest) ... ok...

> The following changes to testlib.py and ssh_test.py in isolation each made it go away, I haven't tested them in combination Further testing revealed this is false. They might reduce...

I'm starting to think this is a combination of 1. Tests not closing their context with `context.shutdown()` 2. `cls.dockerized_ssh.check_processes()` in `teardownClass()` is in a race condition with `sshd` cleaning up...

Using Python 3.12 on the controller (macOS 14.6, venv) and a Debian 11 target (Proxmox LXC guest) I couldn't reproduce this. ``` ➜ tmp python3.12 -mvenv v312 ➜ tmp v312/bin/pip...