V4.5+ Testing Channel Changelog


Released in Testing Channel updates on 9th November 2022.

Our first v4.5 Testing Channel release.

We are, in this release, starting out a little differently than our prior releases. This new approach was covered in our last stable channel announcement here:

this is said aforementioned thread.


As per our usual practice, I’d like to thank all those ‘out here’ and behind the scenes that have made this somewhat significant release possible. Most specifically our very own venerable @Flox. A ‘co-maintainer of sorts’ who has been instrumental in guiding and supporting this and all releases for a number of years now. And the author of a number of pivotal fixes and developments.

We are also grateful (and dependant upon) all update channel subscribers. My appologies for dragging my feed on re-establishing the stable subscription shop setup. All in good time and again, thanks to all those who so persistently help to keep us a-float. I’ll get the new shop sorted directly after a number of pressing matters here in development and in my own personal/business emigration elements.

I would like also to again thank all those participating in our forum and GitHub reporting and bug investigation. Unfortunately this first v4.1.0-0 to v4.5.0-0 development run has been a particularly tricky one with regard to community involvement. Due mainly to the massive changes required rendering the testing branch code essentially unusable for months at a time. On reflection, I think we should maybe have simply releases what we had on the planned (regular/frequent) release cadence. But our testing channel releases have maintained such a long history of being basically usable that it felt like going against the grain. But as a result many fewer folks were able to report issues, but on the other hand, how many “it doesn’t start” issues do we actually need? But to be fair we have always introduced the testing releases as experimental and known broken, as they have yet to be proven to work across many setups ‘in-the-field’, thus are considered broken by default. @Flox has begun some significant work on some far more advanced automated end-to-end testing that should improve our confidence prior to each testing release. But alas all things take time, and this particular release has taken far more than it’s fair share. So lets get right to the changelog for 4.5.0-0 (from v4.1.0-0):

Changelog summary:

Please note our new contributor @parth for their email fix on ‘local’ domain configs. My apologies for this fix taking so long to be release. We also have the super helpful @Hooverdan’s fix for the Scrub UI failing on large pools, a nice fix that has also taken a long time to see the light of day.

A gentle request

As as been our way, or mostly intended way, we have always encouraged, and been encouraged by, user feedback. The testing channel is king in this regard, especially given the massive number and far reaching nature of this release. We very much need this release tested to exhaustion. Please report here on the forum, in discrete threads, issue that your find that are not already identified in our GitHub repos. The intention, now we are back to hopefully more frequent releases, is to push things fast and often to the new ‘testing ground’ of the testing branch. We need to, as quickly as possible, get our entire project rid of it’s technical dept. And the changes in this release should help get us started this next journey.

As is, this release looks indistinguishable from v4.1.0-0, but rest assured, there are many changes, mainly in our dependencies, and many more to come, that all add up to our way forward. But we had to start somewhere and that where is here.

As we move and settle our dependencies, we can begin approaching a few more feature additions, especially given our much improved automated unit testing in this release and the QA testing to come.

Test away and report rapidly as we work against the forces of kipple.

Enjoy and report.