Christoph Thiede

Only one failure left which seems unrelated to my changes. Apart from that, the `MCMcmUpdater>>#doUpdate:upToUpdate:` proposal is pending discussion on squeak-dev: [The Inbox: MonticelloConfigurations-ct.166](http://forum.world.st/The-Inbox-MonticelloConfigurations-ct-166-mcz-td5128353.html)

Afaik downgrading is not possible. My current approach always uses the latest image that was released before the requested version number, see `squeak::latest_release_before()` (https://github.com/hpi-swa/smalltalkCI/pull/506/files#diff-f15a119f2269fdd82da32c9e83444165027e5d69e93ecc94bed8da5f86761ade) :-)

Just a wild guess, but I have experienced similar errors in my image a few times before I uploaded System-ct.1136 into the Squeak Inbox. Could this be an explanation for...

My usage scenario would be something like this: ``` Package MyProject-Core: MyClass #foo #bar Package MyProject-Tests: MyUnitTest MyAcceptanceTest ``` Both tests access `MyClass`, but only `MyAcceptanceTest` accesses `MyClass >> #bar`....

This _could_ be related to my observation that it is not possible to snapshot & reopen some images for a few days: https://lists.squeakfoundation.org/pipermail/squeak-dev/2022-March/219479.html

I could reproduce this issue with multiple Squeak repositories for each of which the Squeak Test Runner displays the coverage results correctly:   The following kinds of methods are...

A simple utility would be this: ```sh smalltalkci --vm https://github.com/OpenSmalltalk/opensmalltalk-vm/releases/download/202112201228/squeak.cog.spur_linux64x64.tar.gz ``` For now, I will implement the download and unpacking myself. :-)

Would a wrapper around the Transcript be so hard? :-)

Then there would be `#become:`, at least in Squeak. :D But I see your concerns, hm ... 🤔

Sounds convincing, but I am currently not in the position to overview the entire scope of such a decision. :-) By the way, I would also discuss this change with...