Users not properly subscribed to bugs when their bug is marked as duplicate

[Tim Hull]

If a bug that a user reports is marked as a duplicate of another bug, the user isn't properly subscribed to the bug their bug was marked as a duplicate of. They sometimes appear subscribed to said bug, though a search for all subscribed bugs of that user does not display the bug in question. This can only be worked around by unsubscribing from the bug and then resubscribing to the bug.

[Skippy le Grand Gourou (lecotegougdelaforce)]

Is this bug still alive ? I just marked a bunch of bugs as duplicates, so I'd like to know if I have to manually unsubscribe/resubscribe every subscribers from duplicates for them to get mailed… For what I have seen the master bug doesn't appear in their list of subscribed bugs, which may be a bad hint.

[Karl Fogel (kfogel)]

This isn't exactly the same as subscription, but the bug-affects-person relationship is now capable of following dups; see the branch that fixed bug #505845. Some of the code there might be useful in resolving this too.

[Deryck Hodge (deryck)]

I believe this bug is describing bug 5977, Person's Bugs pages unexpectedly exclude closed bug reports. I'll dupe this and it's other dupes against 5977 unless anyone objects.

(If others come along to triage this bug, please don't move the duplicates across manually for me. While I normally appreciate the help, I need a bug to test the new auto-move dupes code when it's complete and this one would work.)

[Launchpad Janitor (janitor)]

[Expired for Launchpad itself because there has been no activity for 60 days.]

[Robert Collins (lifeless)]

I've duplicated this onto 5977 - for qa we have qastaging and can setup whatever sort of scenario we need there.