PostgreSQL is an open source database. I guess that they like databases and understand when they can be useful.
But PostgreSQL has no bug tracker. Patches must be sent to mailing lists. Bugs are sent to a mailing list, apparently via a web-form, and some bugs are listed on a page, apparently written by a person.
Want to see what bugs are open and what’s happening to them? Can’t. Want to see what’s the current status of your patch? Can’t.
16 thoughts on “PostgreSQL has no Bugzilla”
I agree with you that any project should have a bugzilla. However, in the PostgreSQL case in particular, it is curious how they have one of the best managed OSS apps out there, with much fewer bugs than any random OSS project (with bugzilla). So, maybe there’s something in there, in the fact that they control the project like this. Who knows.
Welcome to how linux works…at least WRT patch tracking. And IMO most of people likes the email system, just because it saves time.
foo: uh? a large percentage of the top Linux hackers use the kernel bugzilla these days: http://bugzilla.kernel.org/ Even Linus comments on BZ posts from time to time.
Perhaps they should move into Launchpad! It has better usability (especially towards people that are reporting bugs) than Bugzilla, and it’s readily available and hosted. It is also pretty able in features.
Postgres will never uses a proprietary service controlled by a private company. No chance of that happening
They don’t have bug tracking infrastructure maybe because the team is not that large and nobody took care of setting up anything. Postgresql team is not that big and maybe they prefer dedicating this resources to the proper development. Not saying that this is good or bad, but that it just happened. I might be wrng too. :-)
Personally I dislike that I have to go to some website to comment on a bug that get notified about over e-mail. Why is it that bugzilla don’t support comments via e-mail. For GNU Octave (which I’m involved with) this is a deail-breaker, so we discuss bugs via e-mail like the postgress people do. It’s not a great solution, but it’s better than bugzilla :-(
Bugtrackers don’t fix bugs, People fix bugs.
The project has recently been adding some infrastructure to address this area. While patches are still mainly handled over e-mail, you can check the status of submitted patches at http://wiki.postgresql.org/wiki/Development_information As for the bug list, you seem to feel that one generated by a person, like the PostgreSQL TODO, is somehow inferior to one generated by software. That’s a valid opinion, but given the low number of bugs in PostgreSQL they’re doing something right. A high percentage of submissions via the web form are in fact not bugs, just people confused about something. Would it somehow make the project better to open bug reports on all those just to turn around and close them as “not a bug” all the time?
SÃ¸ren: What do you mean with ‘bugzilla don’t support comments via email’? Bugzilla supports comments via email, if the Bugzilla was setup that way… and you’re not using some old version.
(I’m a Bugzilla developer)
> The project has recently been adding some infrastructure to address this area.
Greg, that’s good to hear.
> You can check the status of submitted patches at http://wiki.postgresql.org/wiki/Development_information
However, that is still rather suboptimal. Even if the links were more precise, and if it was easier to follow mailing list discussions via the archive, the use of a human instead of a database increases latency and introduces doubt. That doesn’t help day to day development.
> but given the low number of bugs in PostgreSQL theyâ€™re doing something right.
PostgeSQL is indeed very high quality, but claims of low bugs by projects that make it hard to submit or track bugs are suspect.
> A high percentage of submissions via the web form are in fact not bugs, just people confused about something.
> Would it somehow make the project better to open bug reports on all those just to turn around and close them
> as â€œnot a bugâ€ all the time?
This is often true no matter how you allow people to submit bugs. It’s just something that you have to deal with when you make it easy to submit bugs. And you do need to make it easy to submit (and track) bugs, so you discover all bugs, and everything about those bugs.
At least with a bugzilla (or similar), it’s easier to share out the work of triaging all that information. See the GNOME Bugsquad, for instance.
It’s a pity that there’s no real bugzilla hosting service – an equivalent of wordpress.com for wordpress.org. It’s something that I’d like Openismus to offer to large projects, and I hope that our bugmaster service for maemo.org (to start very soon) can be the start of that.
Yeah, dogfooding is a good thing (given that Bugzilla can run with PostgreSQL, not only MySQL).
You know, I’m hearing a lot of people saying that PostgreSQL has a low number of bugs. And how, pray tell, do people know this? Just because you haven’t reported the bug does not mean that it’s not there.
You’re correct: the accurate thing is to say that PostgreSQL has an extremely low number of reported, outstanding bugs. The general policy at postgresql.org is that all bugs get fixed in 72 hours, or less (of course some really difficult ones take longer). Thus there’s no reason to track them in order to fix them — they’re already fixed.
There are other reasons why tracking would be valuable, but we’ve had a difficult time picking a system which allows good tracking *without* getting in the bug-fixers’ way. Plus everyone has their favorite pet system, so the debates ensue.
:-) Don’t get me wrong, I have no issues with PostgreSQL. That 72 hour turn around is actually pretty impressive!
What were the problems with picking a bug tracking system? Is there any reason why Bugzilla isn’t good enough? It seems to do the Mozilla devs all right… but I do understand that you have different needs.
Yeah, but mozilla has thousands of open bugs