fas username: @konradm package: fftw issues to fix: https://bugzilla.redhat.com/show_bug.cgi?id=1729652 nonresponsive maintainer bug: https://bugzilla.redhat.com/show_bug.cgi?id=1729652 fedora-devel mail: https://bugzilla.redhat.com/show_bug.cgi?id=1952959 comainainers: ?? I am interested in comaintaining the package: no additional notes:
Updated (For Brevity):
The bug in question was reported two years ago, and reconfirmed for each release from 30-34, by 3 different people. I've personally put in the time to find the root cause, then suggest possible fixes, then finally to implement a fix myself and send a patch.
Conrad responded promptly after I send up a flair a few months ago to confirm he is still active. However, despite my efforts to reduce the burden on him, the issue is still stalled with no progress in sight. Again, he isn't responding to status queries, and ignored a request for a timeline estimate. It's now been two months since I posted a patch.
I'm asking FESCo to intervene in some way to help resolve this.
This is a ticket for the Fedora's Policy for nonresponsive package maintainers.
Followup: https://bugzilla.redhat.com/show_bug.cgi?id=1729652#c22
Metadata Update from @churchyard: - Issue tagged with: nonresponsive maintainer
Response to the discussion from the minute notes:
There's already a stalled upstream issue, which I mentioned in the BZ in april.
I appreciate that the maintainer is a volunteer. But he also serves a gatekeeper function. If he uses his discretion to avoid any action among reasonable alternatives, there's nothing I can do. I've put in more and more effort to help resolve this, which has yielded exactly nothing so far. It is what it is, but it also sucks very much.
Given how much time has passed, and my interactions with him, I feel the maintainer feels uncomfortable taking any action on this, for whatever reason. And two years on after the problem was first reported, that's not likely to spontaneously change.
My patch isn't small, but some significant changes were necessary to enable cmake support for several coexisting package variants (single,double,quad,etc') . Again, if the maintainer isn't inclined towards inaction, the size of the patch is not what's stalling progress.
In fact, I suggested a trivial fix early on, which is to simply drop the partial cmake support files from the package and let cmake users use cmake's support for pkg-config as a backend. That should have been doable with minimal effort. The maintainer even indicated this seems a reasonable fix (2021-04-24 comment). But there's been no progress since. That was 2 months ago.
Metadata Update from @churchyard: - Issue tagged with: meeting
This ticket will be discussed during today's meeting: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/P5KOID677ZHCIW4ULTJJIE3OUWZ6IJ4L/
@larryanon Do you consider https://bugzilla.redhat.com/show_bug.cgi?id=1729652#c28 good enough to close this ticket?
Metadata Update from @decathorpe: - Issue untagged with: meeting
dcantrell is co-maintainer now, which is great, but there's been no progress on the bug since two that happened two weeks ago or so - I understand he's on vacation?
As long as it doesn't slip his mind when he gets back, I don't think any further action from fesco is/will be needed. I'm just waiting for the BZ to be resolved.
Thank you.
I'll close this and just comment here anyway in case @dcantrell doesn't help you after he's back from vacation, so we can poke him if needed :)
Metadata Update from @churchyard: - Issue close_status updated to: Invalid - Issue status updated to: Closed (was: Open)
@churchyard, I asked for fesco's help because I've spent months trying to get the maintainer to address a packaging issue which has a trivial fix.
Right now, It looks like fesco immediately snapped into action by appointing a verified packager to resolve the situation, who then promptly and without a word of warning went on vacation, going on 3 weeks now.
Am I the problem??? Am I simply naive to a fault? Should I just accept that there is absolutely, positively, nothing at all I could ever do which would move the needle even a single millimeter? Do I need to accept, into the core of my being, that engaging with the fedora packaging community is an utter waste of time and effort?
And just to be clear, I fixed this problem for myself 3 months ago. It took an unpleasant afternoon of digging. Since then, I've been trying to fix the problem for other people. I'm only trying to help make fedora better by making it as easy as possible for the packagers to fix this for everyone. But, I've been trying now for 3 months and it's become crystal clear that I am a complete ass for trying to actually get something fixed in Fedora.
This is so frustrating! Something this easy should not require this much effort.
The only possible conclusion I can draw from this is that there is something insanely broken either in the Fedora model or in it's community.
I'm sure fesco only wants to help. But this has been a fucking awful experience and I will never again in my life try to improve things in Fedora. I'm going to write this important lesson on a little post-it note, and I'm going to superglue to it my screen, just so my natural inclination to help doesn't lead me into this kind of misery again in the future. And if that doesn't work, I'll get a fucking tattoo of it on my cornea.
Fuck.
I understand that it is frustrating, belive me, I've been there as well. But please do note that Fedora is driven by volunteers. No need to use expressive language to vent out your frustrations.
David is back from his vacation today, but I imagine he has a pile of emails to go trough any many other stuff to deal with which might be more important or urgent to him then this thing where he volunteered to help. Please, don't attack the people who try to help you.
Login to comment on this ticket.