Should I make a tracking bug in fedora for problem reported upstream?

Adam Williamson awilliam at
Tue Aug 30 02:36:54 UTC 2011

On Sun, 2011-08-28 at 09:43 -0500, Bruno Wolff III wrote:
> I have reported for an
> issue with 3.1 kernels crashing, typically on the order of hours after
> booting. Based on some names in the traceback, I suspect this is MD raid
> or luks related.
> I reported this upstream, since that seems to be a better place to get
> attention for development kernel bugs, than Fedora's tracker. However,
> if this should be considered a beta blocker, then I should probably also
> make a Fedora bug entry so that it can be tracked as a blocker.
> However it doesn't seem to meet the explicit requirements for a beta blocker.
> The only category it might fit in is a high severity bug. But if it really
> limited to systems using MD raid or luks encryption, I am not sure if
> it would meet that standard.
> If it isn't a blocker, I'd rather not enter a duplicate bug in Fedora since
> that would just seem to be a waste of resources. (But maybe QA would rather
> see the bug entry anyway.)

You pretty much answered all your own questions. =) If you want to
propose it as a Fedora blocker formally then yes, you should create a
downstream bug. Since we have a while to go before Beta, if I were you,
I'd wait a bit longer until it's better understood and then bring it
downstream if appropriate, but there'd be nothing terrible about doing
it now. 

The kernel team is also generally pretty responsive in #fedora-kernel,
so you could simply flag up the upstream bug there and ask if anyone
wants to take a look and see how worried we should be.

As a general rule, only raise a bug in Fedora if it's needed for
something Fedora-specific: for the blocker process, or if you want to
request an upstream fix be brought downstream faster than it usually
would be (which varies from package to package of course).
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | adamwfedora

More information about the test mailing list