During the Fleet Commander talk at DevConf'18 Adam Williansom approached us mentioning that Fleet Commander is definitely potential candidate for adding to their OpenQA testing.

https://openqa.fedoraproject.org/
https://fedoraproject.org/wiki/OpenQA

We (Fleet Commander Team + myself) are already in contact with Adam and I guess this is one of the best end-to-end way to have it tested.

I'm assigning this bug to myself just to keep updating the status of this effort and once it's done there I'd suggest to have this ticket closed.

An issue has been opened by @adamwill on fedora-qa: https://pagure.io/fedora-qa/os-autoinst-distri-fedora/issue/65

Note that at least current openQA is kinda focused on end-to-end testing, and we're mainly using it for very high-level "does everything basically work" testing. It is, for...various reasons I won't go into unless you want to know, not the ideal system for testing at any level below end-to-end, or for really deep dive white- or grey-box testing. What it's mainly good for is telling you 'does the entire thing-being-tested work with the whole combination of bits in this distribution compose or this distribution update, when tested very much from the outside, as a user would interact with it - absolutely no knowledge of the tested bits, mocking, etc.'.

So, we can definitely get you 'more than zero' testing, but possibly not everything you want :) For instance, testing on every upstream git commit, or something like that, would be quite a way out of the project's current scope, and there may be systems better suited to that.

So, we can definitely get you 'more than zero' testing, but possibly not everything you want :) For instance, testing on every upstream git commit, or something like that, would be quite a way out of the project's current scope, and there may be systems better suited to that.

I am all 5 for testing Fleet commander on Fedora. We cannot rely on downstream testing and you might find bugs even in GNOME. Therefore SSSD should be nice upstream and we should try to prevent creating regression in such complicated setup. And as it was mentioned in description of the ticket. The purpose of ticket is to test "interface" which we provides to Fleet Commander. Debugging in openQA of bugs which happened deep in stack is not simple.

Yup, that was basically my take on it: I don't think getting testing into openQA would really "resolve" the intent of this ticket.

I'm not sure how much is needed from the SSSD upstream here? The OpenQA work is tracked separately, so for the time being I'm moving this ticket to the CI milestone, but of course we can shuffle the ticket around..

SSSD is moving from Pagure to Github. This means that new issues and pull requests
will be accepted only in SSSD's github repository.

This issue has been cloned to Github and is available here:
- https://github.com/SSSD/sssd/issues/4643

If you want to receive further updates on the issue, please navigate to the github issue
and click on subscribe button.

Thank you for understanding. We apologize for all inconvenience.