On Tue, Mar 24, 2009 at 03:20:48PM -0000, martin wrote:
> It's useful to keep track of how often each oops happens and if it's
> suddenly fixed etc. Certain oops might actually be one off events that
> happen due to some obscure race conditions while others happens all the
> time on boot or while doing a certain action. For triage, it's important
> to be able to keep track of these. I agree that it needs to be as
> unobtrusive as possible though.
For kerneloops.org, I agree, every oops should be reported as they are
measuring frequency.
For Launchpad bugs, these should only be reported once per unique oops. For
bonus points, if there is an existing bug for the oops, kerneloops could
increment the "affects me too" counter.
On Tue, Mar 24, 2009 at 03:20:48PM -0000, martin wrote:
> It's useful to keep track of how often each oops happens and if it's
> suddenly fixed etc. Certain oops might actually be one off events that
> happen due to some obscure race conditions while others happens all the
> time on boot or while doing a certain action. For triage, it's important
> to be able to keep track of these. I agree that it needs to be as
> unobtrusive as possible though.
For kerneloops.org, I agree, every oops should be reported as they are
measuring frequency.
For Launchpad bugs, these should only be reported once per unique oops. For
bonus points, if there is an existing bug for the oops, kerneloops could
increment the "affects me too" counter.
--
- mdz