Is Bugzilla Broken?

[lazlow]

Hey Guys

I am not asking if the site or server is down. What I am asking is the concept broken?

In FC4 and FC5 days one could report a bug and get a response. While it was true that I did not always like or agree with the response, I did get one. With F7 it seems(at least to me) that if you do not get a response within a week, you will never get a response. I have posted several bugs in F7 that got responses and solutions within a week, these are the almost trivial bugs(I still appreciate the time spent by those packagers that helped me). All of the serious bugs I have posted have gone without response. Over the last few months I have posted more information to these bugs as updates have come out, still no response. If more information is needed fine, post that more information is needed. If it is a duplicate, then post that it is a duplicate. But just allowing the bug to sit there ignored is just wrong.

I do not know if it is a situation where the talent has been reassigned to work on F8, if it is the shear number of bugs in F7, a combination of the two, or some other combination of factors. It also seems to me that the types of bugs are different in F7 than in previous versions. F7 errors (at least mine) do not seem to keep the system from running, they just seem to keep the system from running properly. One example of this is that within system disk transfers are much slower in F7 than they were in FC6 or FC5 (8-40% depending on system). The hdparm tests are the same but " time cp" test show considerable reduction in transfer speed. Yet there is no response to the bug in bugzilla. For those of us that move 10gig data sets around several times an hour this is a significant bug.

I posted this basic question to the new Ask Fedora section. Their basic response was to go to the dev lists. Well if I have to go to the dev lists to get a bug fixed, what is the point of bugzilla?

Thanks
Lazlow

[Hlingler]

Hello:

Obviously, I'm not alone...I hate to say this, but I've begun to wonder if it's worth the time and effort to (properly) research and report a bug. And I assume you're referring to redhat.bugzilla. Still, I find it hard to complain about any FOSS, but I agree that redhat.bugzilla seems to have become...less responive.

Regards,
Vince

[lazlow]

Vince

Yes, I meant redhat.bugzilla.

Your description is far more tactful than mine. I wish I had more of that skill.


I sent the following (that I would essentially consider a repeat of above):

I posted to your ask Fedora section about bugzilla. I am assuming that the response given was from you. The response was pretty much passing the buck. If a bug sits there for 2 months plus and somebody has not commented on it, (need more info, wrong component, or whatever) there is something very wrong with the system. How do we (the average user) at least get the bug assigned to someone? At the rate things are going F8 will be out before somebody contacts me about my bugs with F7. So what is the point of filing a bug with bugzilla? At this point it seems to me that if one has not received some response to a bug within a week, one never will. Of the four (4) bugs that I currently have with bugzilla the only one that has even been assigned to anyone is the one that I suspect will never be solved (intermittent failure of K3b). Repeated posting with more information does not seem to be of any use. If I have to join a dev list to get a response, then again WHAT IS THE POINT OF BUGZILLA? It just seems silly to me that we are constantly asking users to post bugs to bugzill but then we ignore the report once it has been filed.

Any words of wisdom?

Lazlow

I almost fell out of my chair laughing at the response(the names are all that has been edited):

There is a team of people answering questions and the point of bugzilla reports is not a guarantee of fixes but a place to centrally keep track of issues.

Now any of us who have been using Linux for more than ten minutes knows that there are no guarantees of any sort. Those of us who have used bugzilla have read the first page:

Thank you for visiting Bugzilla. Bugzilla is the Red Hat bug-tracking system and is used to submit and review defects that have been found in Red Hat distributions. Bugzilla is not an avenue for technical assistance or support, but simply a bug tracking system. If you submit a defect, please provide detailed information in your submission after you have queried Bugzilla to ensure the defect has not been reported yet. Defects will go directly to the engineer responsible for the component you filed the defect against. Engineers have many responsibilities and will get to your defect in due time.

Most of us have been trained that when we find a bug (not user error, that is what the forum is for) we are to file a bug report with bugzilla so that the issue can be examined. In the past the system has worked rather well. You post an issue and someone would comment on it. Notice I did not say that they would fix anything (although they usually did). Sometimes it was as simple as, this component is being dropped in favor of some other component and no further work will be done on this issue. Which was fine, you knew that this was a dead end and it was time to find another option. There were a library of responses of a similar nature that while they were not entirely satisfying they were, for a lack of a better term, responsive. This is where I think bugzilla is now broken. Many of the bug reports (and there are a lot of them) just sit there and get no response. I guess most of us would assume that no one is in fact reading them. I have no evidence proving this, it is just that with so many reports (not just mine) just sitting there without response, it is the most logical conclusion that I can reach.

So the question becomes what part is broken? Is it of a technical nature and the software is overloaded to the point that a certain percentage of the new bugs just never get forwarded to the proper individual? Is it a failure in the transfer of ownership of certain packages (I have seen this one when an individual has left before a replacement could be found) on a larger scale than we have seen in the past? Is it a situation where the bugs are reviewed by someone and then are never assigned due to overloading, lack of knowledge or whatever?

Lazlow

[Blackspyder]

I posted a bug for F7 3 weeks ago and it hasent been touched. I would be happy with a email stating the it was being looked into or asking for more info but nada. Had to downgrade to fix my issue (FC6 works fine, Ubuntu works great, F7 worked horrible) .

[lazlow]

I got all excited today. In my email there was a notification from bugzilla. Unfortunately it was just somebody with the same problem. Still no response or recognition that anyone has looked at it.

[Hlingler]

Hello:

Well, I decided to go ahead and file another redhat.bugzilla last night. This brings the total number of open bugs I've started to four (oldest opened 7-10-2007, all status "NEW", not "ASSIGNED"). Plus three existing bugs I've commented on because I have the same problem (oldest opened 10-24-2006, one still "NEW", two "ASSIGNED"). I don't consider any of these to be trivial (e.g., components: kernel, firestarter, xorg-x11-xinit), and I did what I consider to be extensive research (including posting questions here on the forums) before filing them as bugs, plus the time to write a detailed and thorough bug report. Like lazlow, I've received quick responses, and closure, of a couple other bugs that were closed as "NOT-A-BUG", "CANT-FIX", etc.

I certainly don't want this to turn into a "scoring contest" ("I've got more than you! Nah-Nah-Nah") or a "RedHat-bashing fest", but for the sake of this discussion, I thought I'd mention these facts, without drawing any conclusions yet. I hope that this will help to give everyone, including me, a better idea of the state of things at bugzilla.redhat.com.

Thanx and Regards,
Vince

[lazlow]

Still no movement on any of my bugs. Anybody else seeing anything?

[giulix]

All the bugs I submitted got dug into and eventually solved. They were all reproducible. I suspect that submitting a bug that is not reproducible, or not confirmed by different sources, hampers its chances to get looked at by the developers. I don't know whether this is the case...

[lazlow]

On most of my bugs there are multiple submitters and reproducible. In most cases the bugs have not even been looked at as far as I can tell.

[Thetargos]

Laz, I have an issue opened in Bugzilla since FC5, all "NEEDINFO" requests have been swiftly answered and the bug still runs strong in F7...

Forgot to mention, that in my Fedora experience (since FC1), F7 has been the buggier of the bulk, that place was proudly held by FC3 (closely followed by FC2, but those issues were mainly due to the migration to 2.6.x kernels, which I did since FC1). Most of the issues I've encountered, have already been reported in Bugzilla, I've seldom added to them (especially if they've been looked into) and have reported a few... I'm certainly hoping for the best for F8.

[bob]

Hmmm...well there are mail lists to get on and possibly question the status of some of the more important bugs. That might get you answers or yelled at, depending on who you hook up with. There's also this: http://forums.fedoraforum.org/showthread.php?t=162409 which might be a vehicle to get some answers.

On 'Assigned' bugs, after a reasonable period of time, I've actually emailed the guy assigned and POLITELY asked if there was more info or any progress. Generally the responses have been good and some of the bugs were looked at very quickly.

Some things to consider, people.... There are a limited number of developers who are working on both the bugs and portions of the next version. Chances are that something they consider very important and time-sensitive has absorbed them. Now, if you REALLY want to make the system better and have some skills: http://fedoraproject.org/wiki/FAQ#he...94991ff841e658 you can help ease the load and possibly answer your own and other questions. Remember that all linux is a group enterprise and we can and should contribute as our skills allow.

[lazlow]

Thetargos

I just wish they would ask for some information. Then I would at least know that they have looked.

Thanks
Lazlow

[lazlow]

Bob

Post #3 is the general reply I got from ask Fedora.


I have emailed the assigned individuals with a simple "Just checking on the status of bug XXXXX, do you need any further information?". No reply from any of them.

Thanks
Lazlow

[Thetargos]

Originally Posted by bob

Hmmm...well there are mail lists to get on and possibly question the status of some of the more important bugs. That might get you answers or yelled at, depending on who you hook up with. There's also this: http://forums.fedoraforum.org/showthread.php?t=162409 which might be a vehicle to get some answers.

On 'Assigned' bugs, after a reasonable period of time, I've actually emailed the guy assigned and POLITELY asked if there was more info or any progress. Generally the responses have been good and some of the bugs were looked at very quickly.

Some things to consider, people.... There are a limited number of developers who are working on both the bugs and portions of the next version. Chances are that something they consider very important and time-sensitive has absorbed them. Now, if you REALLY want to make the system better and have some skills: http://fedoraproject.org/wiki/FAQ#he...94991ff841e658 you can help ease the load and possibly answer your own and other questions. Remember that all linux is a group enterprise and we can and should contribute as our skills allow.

I know and am aware of all of the above, and that's why I've pretty much ignored my issues and carried on. I wish I had the skills to look into most of the issues I've encountered and others have encountered, but the fact is I don't 1) have the skills, and 2) have the time to look as thoroughly as I'd like to... kind of pointless, since I don't have the skills, so...

At any rate, I'm not complaining as such, more like venting. One of the issues I've got has to do with DRI drivers, and I'm fairly aware that it is not "Fedora's fault" as such, but a Fedora developer has more chances to get in touch with the DRI/Xorg developer responsible for that particular driver (if you must really know is about image corruption in an S3 card after resume from suspend).

It certainly looks like at least currently, there's way too much traffic on Bugzilla than what the developers and possibly bugzilla itself can process. I've had a couple of bugs submissions marked as dupes (thankfully, as I also lack some search skills... known bug of mine), and I'm aware that maybe the issues I've reported have more to do with upstream than Fedora itself. I am at least aware of one such case, besides the driver issue mentioned above, Amarok, but I've already seen the fix is in the works in upstream, and as such it should cascade nicely to Fedora when it gets released.

[Hlingler]

Hello:

All I've seen since my last post is a couple added comments from people with the same/similar bugs and/or adding themselves to the CC list. However, whenever I'm tempted to get annoyed or feel ungrateful, it seems that I find myself forced to deal with Windoze at that exact moment. Tonight is such a time. After umpteen reboots due to minor software settings tweaks, cryptic error messages, and mysteriously disappearing/reappearing hardware, it's almost a relief to reboot into FC6, and watch the familiar "XError: BadMatch..." warning messages scroll by at a rapid clip in ~/.xsession-errors, on an otherwise functional operating system.

Thanx and Regards,
V
WINDOWS DELENDA EST