bug mail explaining cause of message has an inaccessible link when the the master is private and the subscription is is via duplicates

[Donnie Pennington]

I am subscribed to bug 121439 "from duplicates". I need to stop getting all bug email and successfully unsubscribed myself from several other bugs including at least one where I was subscribed "from duplicates". However, 121439 gives me this message and won't allow me to unsubscribe:

Not allowed here

Sorry, you don't have permission to access this page.

You are logged in as Donnie Pennington.

[Christian Reis (kiko)]

That's very odd. I wonder if it's because 121439 is itself a duplicate. Just to be clear, this is the URL that fails for you, right?

https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/121439/+subscribe

You're not by any chance not logged in, right? :-)

[Christian Reis (kiko)]

After answering that you can unsubscribe using this URL:

https://bugs.launchpad.net/ubuntu/+source/gnome-system-tools/+bug/130874/+subscribe

[Donnie Pennington (donniepennington)]

Yes, I am logged in, and, no, this URL does not fail. It's when I click 'Continue' from this URL that I get the 'Not allowed here' screen.

[Donnie Pennington (donniepennington)]

After reading my previous comment, it occurs to me that getting the 'Not allowed here' after clicking 'Continue' is what you meant by the URL's failing. In any case, the second URL you gave did successfully unsubscribe me from bug 121439. Thanks!

[Diogo Matsubara (matsubara)]

To reproduce on sample data:
1. Logged in as no-priv, subscribe to http://launchpad.dev/bugs/10
2. Set bug to private
3. Mark it as dupe of http://launchpad.dev/bugs/4
4. Open bug 4 and try to unsubscribe from it: http://bugs.launchpad.dev/firefox/+bug/4/+subscribe
5. When you choose "Unsubscribe me from this bug" and click Continue, you get a Not allowed here page

This might be a dupe of 89476

[Jeff Fortin Tam (kiddo)]

the problem is, how can I know which dupe to unsubsribe from, when there are *88* duplicates?

[Robert Collins (lifeless)]

I think we should not tackle this during disclosure. Its of low benefit and tricky to do. Implementation wise we're looking at either:
 - preventing the situation (there is a bug for this)
 - making private bugs metadata disclosable if they interact with other bugs via 'is a master bug for'

Either way involves some care, attention to detail and migration of existing data.