I *think* this is a bug in NATTkA? I recently created a new 'stablereq' bug for net-dns/nsd-4.8.0-r1, Bug #930033, and then ran this command: > nattka sanity-check 930033 NATTkA output the following text: > WARNING:nattka:Running in pretend mode. > WARNING:nattka:(pass --update-bugs to enable bug updates) > INFO:nattka:NATTkA starting at 2024-04-14 23:04:07.026521 > INFO:nattka:Found 1 bugs > INFO:nattka:Bug 930033 (STABLEREQ) > gentoo -- updating profiles cache: x86 > INFO:nattka:All good > INFO:nattka:CC arches: amd64@gentoo.org x86@gentoo.org > INFO:nattka:Expanding package list > INFO:nattka:New package list: =net-dns/nsd-4.8.0-r1 amd64 x86 > > INFO:nattka:New comment: None > INFO:nattka:NATTkA exiting at 2024-04-14 23:04:22.982823 > INFO:nattka:Total time elapsed: 0:00:15.956302 The first two lines suggest that no changes are made to the bug, but about ~4mins after running the command, changes appeared on the bug. I can't tell if they are from some server-side component of NATTkA or just Bugzilla being sluggish. I would assume that if NATTkA is saying it's running in "pretend" mode by default, that no changes should happen to the bug at all, but it looks like NATTkA made changes anyways.
That update was done with the "NATTkA bot" bugzilla account (which runs automatically), unless you have + using the api key for it you can't do that yourself. If used your own api key and without pretend mode, it'd show your own account. Or am I misunderstanding something here?
(In reply to Ionen Wolkens from comment #1) > That update was done with the "NATTkA bot" bugzilla account (which runs > automatically), unless you have + using the api key for it you can't do that > yourself. > > If used your own api key and without pretend mode, it'd show your own > account. > > Or am I misunderstanding something here? So the "NATTkA bot" is a completely separate thing from app-portage/nattka? I thought they were related. Does this bot account do stablereq processing entirely on its own shortly after such a bug is created?
The nattka bot *is* using app-portage/nattka, just without pretend running on our infra. So I wouldn't call them unrelated -- but you can use it locally yourself (useful to confirm and the like), it just won't be using the same account if you were to actually update bugs. ...which is not something you normally need to do given infra's nattka checks all open keywording/stablereq bugs for updates. Usually pretty fast (few short minutes at most) to pickup changes. Like if added CC-ARCHES keyword, then it'll go ahead and cc amd64@g.o and friends for you after a short wait. Anyhow, the command you ran locally did nothing to the bug.
(In reply to Ionen Wolkens from comment #3) > The nattka bot *is* using app-portage/nattka, just without pretend running > on our infra. So I wouldn't call them unrelated -- but you can use it > locally yourself (useful to confirm and the like), it just won't be using > the same account if you were to actually update bugs. > > ...which is not something you normally need to do given infra's nattka > checks all open keywording/stablereq bugs for updates. Usually pretty fast > (few short minutes at most) to pickup changes. Like if added CC-ARCHES > keyword, then it'll go ahead and cc amd64@g.o and friends for you after a > short wait. > > Anyhow, the command you ran locally did nothing to the bug. Yup, so this boils down to a misunderstanding on my part, then. I thought the tool being used locally triggered the bot to do something to a bug. Would explain why there seemed to be a time lag between when I ran the command locally and then the bug got modified by the bot -- automated action on the server. Okay, TIL!
