[gnome.org #14065] Fwd: problem with imported bugs: everconfirmed is 0

["Charles Lindsay via RT" <support gnome org>]

You are reading this message because you are a watcher of the General queue at GNOME.org Request Tracker.

Fri Jan 03 02:47:36 2014: Request 14065 was acted upon.
Transaction: Ticket created by chaz yorba org
       Queue: General
     Subject: Fwd: problem with imported bugs: everconfirmed is 0
       Owner: Nobody
  Requestors: chaz yorba org
      Status: new
 Ticket <URL: https://rt.gnome.org/Ticket/Display.html?id=14065 >

--- Original message follows: --------------------------------------------------------------
Hi guys, we've just become aware of some fallout of our bug migration
process.  If you remember, I asked you guys to modify the import script to
allow it to leave bugs status as specified in the xml: NEW (normally I
think it would set it to UNCONFIRMED regardless).  Well, it turns out this
means the bugs are in a state they don't normally get in, with status NEW
but "ever confirmed" still 0 (it's apparently normally set to 1 when the
bug enters NEW status).

This is causing contributors to be unable to make unrelated changes on some
bugs.  We had a user try to update the whiteboard status for a ticket, only
to get the error "You tried to change the Ever confirmed field from 0 to 1
, but only a user with the required permissions may change that field."  I
think what's going on is it's seeing that it's NEW and not "ever
confirmed", and updating "ever confirmed" to 1 along with the edit the user
is trying to make.  Apparently the canconfirm permission bit is necessary
to make that change, and not set by default.

I asked in #bugs (after asking in #sysadmin), and the consensus was to
mass-edit the affected bugs (
https://bugzilla.gnome.org/buglist.cgi?query_format=advanced&field0-0-0=everconfirmed&bug_status=NEW&type0-0-0=equals&value0-0-0=0)
with a trivial change, and let Bugzilla's natural edit behavior fix the
problem for us.  I'm glad I tested that with a handful of bugs, though,
because mass editing doesn't trigger the "ever confirmed" change like
editing individual tickets does.  And we did confirm that simply adding a
comment to a single ticket fixes the problem for that ticket.

So, I'm reaching out to you guys.  Any idea what we can do here to make
sure everyone with regular permissions can update our bugs?  I hate to
suggest hacking around our existing hacks, but do you think it'd work if we
simply updated the database?  If I've done that search correctly, it
appears that our imported bugs are the only ones with this problem.

Thanks, and happy new year to everyone!

Charles