Since about yesterday, all crash reports are reported as 'normal' bugs. Not sure if there is a change in DrKonqi or in Bugzilla.
The 'crash' type is a part of "Importance" which it was agreed that non-contributors shouldn't have the right to set. Therefore any attempt by Dr Konqi to set this will be ignored by Bugzilla.
It was agreed where? Sorry, this breaks DrKonqi backtrace searches. Also, people start reporting their wishes as "normal" bugs, and I have to go through them and change the severity and usually also the title (because they add "[Wish]" there instead).
The discussion took place on the kde-devel and plasma-devel mailing lists. Please see http://kde.6490.n7.nabble.com/Re-Closing-old-Plasma-4-bugs-td1674858.html It was agreed in that thread that Importance (priority and severity) should be restricted.
I indeed remember reading this thread, but did not see these consequences. As I understood it, we wanted to limit additional abilities for new triagers, but never remove existing abilities. Nate, can you clarify if the issues raised in comment #2 are intended? It is odd that people can close any bug, but cannot even set the severity of their own bugs.
An unfortunate consequence of how Bugzilla works is that the ability to change the Importance also impacts on the ability to set the Importance of a bug in the first instance (as Bugzilla treats this as a change from the default values)
Since there does not seem any solution to this, I re-titled the ticket.
In that case, users that don't have the right to change the severity shouldn't have the ability to attempt to set it when filing a bug through the "Enter Bug" form on website. I was very confused about why someone had to fix my severity even though I had (attempted to) set it correctly.
*** Bug 393346 has been marked as a duplicate of this bug. ***
Can you please post a screenshot and the link to the page which offers that so I can investigate? @Christoph: In regards to getting this changed, as the community has set the policy here it is not for me to overturn it. If you would like this overturned, please respond to that thread and we can go about getting the policy amended. I agree that restricting it (at least for initial creation on severity, for the normal/crash/wishlist states) appears to have had unintended consequences.
Created attachment 112151 [details] screenshot (In reply to Ben Cooksley from comment #9) > Can you please post a screenshot and the link to the page which offers that > so I can investigate? > users that don't have the right to change the severity shouldn't have the > ability to attempt to set it when filing a bug through the "Enter Bug" form > on website. URL: https://bugs.kde.org/enter_bug.cgi?product=elisa Here is a screenshot of my new bug page for all available products. This is a separate issue but I decided to comment here since it's related, however I could file a separate bug report for this.
Thanks. I've investigated and it looks like the bug entry form doesn't do any of the permission checks which the rest of Bugzilla does :( We'll defer this for now pending the result of the community review which I assume is going to be requested regarding this?
*** Bug 394108 has been marked as a duplicate of this bug. ***
Nate, any opinion on comment #4?
Before the change, did users previously have the ability to set the Severity? Is that why this was working before?
Yes, users could specify a severity when reporting tickets, both via DrKonqi and the web interface.
Maybe we could return to that, then? I don't recall it being a major problem before we removed the privilege.
Yes, please. That was the original intention of this ticket. Sysadmins did not want to re-introduce the permissions after the discussion from comment #3; that's why I was asking for clarification of said discussion and changed the title later. I have changed the title again to reflect both (common) issues, crashes and wishes. Ben, or would you prefer a new phab ticket?
Yeah, I don't think anyone ever intended to block DrKonqi from reporting crash severity correctly. That was unintentional fallout from the permissions change for the severity field. If we can't fix just that issue, I would support rolling back the permissions change for the severity field to un-break DrKonqi.
Please note that as the change to permissions was approved by the broader community (either via explicit approval, or through silence, in a public mailing list discussion) then further changes to those permissions must be approved by the broader community as well. This means you'll need to reopen the thread on kde-devel. While I don't expect there to be any opposition to the request change (to permit that specific item to be set on creation at a minimum) the broader community still has to be consulted as the broader community were the ones who requested and approved it being restricted in the first place.
Replied there.
This was recently fixed.