By user12607856 on Feb 17, 2009
For the past couple of weeks, I've been triaging bugs in the kernel and software sub-categories in the OpenSolaris Bugzilla bugs database.
Bugs that are filed in these two sub-categories tend to be catchalls for all the bugs that don't have another existing product/cat/sub-cat. defined that they could go into. They are also typically dumping grounds for bugs that are not being actively looked at by Sun engineers.
OpenSolaris releases are currently based on Nevada builds. We just take the SVR4 packages that the Nevada distributions create, and convert them into IPS packages and put those in network repositories or deliver via updated ISO images of the Live CD.
If we want these bugs to be fixed properly, then those fixes need to be applied to the Nevada sources. In order to get the Sun engineers to do that, they need to be aware of the bugs. That's why they are being transferred over to the Bugster database and closed here.
It's far from a perfect system, but hey, you work with what you've got.
In doing this bug transferring, I noticed that a lot of the new bugs that I'd filed in Bugster, were being actively picked up and worked on. That got me curious on just how successful this reverse transfer has been.
I wrote a small Python script that did a query to get a list of all the bugs that have been closed as TRACKEDINBUGSTER, and then scrapped this to get a list of all the BugsterCR values from their Whiteboard fields. I then used an internal Bugster->Web web site, to read each of those bug reports and extract the Status field. I then summarized the results.
(I fully appreciate that this would be much easier if I used SQL to integrate these bugs databases, but I don't know how to do that, or even where to go to find out how to do that).
Currently there are 400 bugs that have been closed in OpenSolaris Bugzilla as TRACKEDINBUGSTER. Here's their status:
1-Dispatched: 99 2-Incomplete: 19 3-Accepted: 59 4-Defer: 8 5-Cause Known: 4 6-Fix Understood: 3 7-Fix in Progress: 10 8-Fix Available: 6 10-Fix Delivered: 109 11-Closed: 87
The real bug that needs to be fixed here is to get the Sun engineers to actively monitor (and respond to) all bugs that are filed at defect.opensolaris.org (and I understand that that is being worked on), but it's nice to know that this triaging is having some effect, rather than all those open bugs just lying stagnant.