Update on bug system for OpenJDK (web-discuss)

[Phil Race]

A lot of the volume comes from consumer type end users
and one of those cited below is a good example of this :

http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7022357

They report bugs into our database because in large part because the hotspot
error directs them there :

# If you would like to submit a bug report, please visit
# http://java.sun.com/webapps/bugreport/crash.jsp

That does tend to result in a lot of the bugs being misfiled as
hotspot bugs even though they are not related ..

Also as you can see from the stack trace there, this is
very likely a bug in the application which has its own DLL ..

  V [jvm.dll+0xfce45]
  C [jaclib.dll+0x7cf5]
  C [jaclib.dll+0x7c03]
  C [jagdx.dll+0x17ef]
....

I'd like to think that if we continue to direct consumers there
and developers to System X that the signal to noise ratio will be
such that the bug database don't degenerate. Then we don't
have to worry so much about restricting access.

I suppose those hotspot error logs need to be updated to
document the new URLs at some point.

-phil.


On 6/1/2011 3:41 PM, Roger Calnan wrote:
>>> Surely it would not be the case that all bug reports disappear into a private
>>> queue viewable only by your internal team?
> today the bugs aren't in a private queue, all issues appear on bugs.sun.com
> within a day, for example:
>
> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7033945
> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7022357
>
> are incidents which haven't been picked up as bugs yet. This was done
> about a year ago to address the sense that incidents went
> into a black hole if they were not reviewed in a timely manner.
>
> 	To be clear the current model isn't ideal as the infrastructure
> on which it is based doesn't provide the filtering necessary so that
> it would be easy for developers to see a unified view, the goal is to
> have that as part of the new system.
>
> 	There were some early ideas sent out in April which go into
> more details:
>
> http://cr.openjdk.java.net/~rlewis/BugTracking/JDK%2bDefect%2bTracking.html
>
> looking at it today there are some terms that would need to be
> clarified when it comes to "Open JDK developer", which was
> raised earlier on the thread.  Also I'm assuming that the rules
> around if bugs should be triaged should depend on the project,
> some may well not want any,
>
> 	Roger