expected failures on EOS storage get out of date

[phil-davis]

We did not manage to get the core API tests running in CI with EOS storage. That means that we do not get automated feedback abut which things are passing and failing with EOS storage. So the contents of https://github.com/owncloud/ocis/blob/master/tests/acceptance/expected-failures-API-on-EOS-storage.md get out-of-date.

What should we do about that?

• try harder to get the tests running in CI with EOS storage
• get someone to locally run the tests "every so often" and adjust the expected-failures file
• something else?

[phil-davis]

@dpakach can you add some comments about what was difficult when trying to run EOS in CI.

I added this to the QA/CI/TestAutomation project so that we notice it to discuss. After discussion then we can see what effort is needed...

[butonic]

There is also work on a Cephfs storage driver. One by @thmour one by @aduffeck Something the keep in mind...

[individual-it]

issues in CI see #586 & owncloud/product#156
but currently we are even failing to run EOS locally or on hetzner servers see #541 (comment)

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 10 days if no further activity occurs. Thank you for your contributions.

[refs]

How important is this? Bringing @micbar to the party. It sounds relevant enough to pull it into the backlog.

[micbar]

@refs @phil-davis at the moment, this is not possible.

[phil-davis]

When someone wants to know how well EOS storage is working, we could get another someone to run the core API tests locally against an EOS storage setup. And they could update the expected-failures. It just depends on how often to put in the effort of this.

[butonic]

eos tests are no longer covered by ocis ci. it should run in reva or elsewhere.