On 8 May 2019, at 10:24, Liang Qi <[email protected]> mentioned:
https://codereview.qt-project.org/260927
Shawn Rutledge (8 May 2019 15:43) replied
> I don’t understand why a change that is just sitting there on Gerrit
> can affect the merge. Nobody ever staged that one onto the “wrong”
> branch, right?
None the less, IIUC, apparently the automation involved did stage the
5.13.0 one when a merge to 5.13.0 (presumably the end of soft-branching)
brought in the change from 5.13 with the same ID, that had been merged
to 5.13 earlier. Quite why that would happen is unclear, and may be
considered a bug in whatever automation did it, but it seems this does
happen.
> And then the bot detected the duplication and auto-abandoned it.
No, the whole point of Liang's message is that he had to abort the build
(it wasn't auto-detected) and Frederik (an admin) had to manually
abandon it so that he could get on with his merge.
Eddy.
_______________________________________________
Development mailing list
[email protected]
https://lists.qt-project.org/listinfo/development
