Built on openSUSE testing channel live (early-adopters/developers only)

@kupan787 Re:

Definitely, and already has with the presumaly upstream pending ssh change you’ve reported already.

That’s the worry really. Plus, since we arn’t releasing the rpms for Tumblweed, you have to build from source to test the result.

But on the other hand, you could try removing the old Tumbleweed repo and adding the Leap 15.2 repo, I can let you know here once we have a Leap 15.3 repo. That way you at least are testing our latest code as 4.0.1 was actually release on the 2nd August 2020:

So that would definitely make your reports more relevant. It doesn’t adhere to the ideal of running rpms built on the actual OS they target, which all our release rpms are, i.e. all Leap 15.2 rpms are build on 15.2 itself. But it does mean you are running current code which will be better all around, plus you are using Leap 15.2/15.3 packages that you shoe horned in anyway. If you go this way you should end up with 4.0.5 as of now and it will be interesting to hear if this works. Plus if/when we fix anything you report you can see if it’s successfully instantiated in a ‘real’ rpm. Otherwise you are stuck with source builds which don’t update themselves so no end-to-end testing of what we will actually release.

Thanks a tone for your continued involved in this way. It’s an invaluable and central part of how we got to where we are now and much appreciated.

See how you get on with running the Leap 15.2 repo instead. Though you would be advised to uninstall the existing rpm, then remove the old Tumbleweed repo and the /opt/rockstor dir then add the new repo and take it from there. If all goes well you should get 4.0.5. Apologies for not thinking of this before but given your approach in shoe horning the 15.2 lib into Tumbleweed anyway it may make for an interesting experiment. We just have to keep an eye on the ‘setup’ not being standard as it were. And hopefully we can tell, such as @Flox has done already, the actual cause anyway. There are also likely kernel issues, but again we can hopefully narrow down situations there also. These will mainly be down to how the btrfs commands change in their output. But again it could be interesting to see these coming down the line.

Thanks again and do keep us posted on this experiment. And if all falls to pieces then you can always stand up a Leap 15.2 base instead. Or better still build the actual installer as the resulting system is likely very much smaller than what you likely have there anyway which is always nice.

2 Likes