[c-nsp] Questions on open caveats for 6500 IOS (12.2(33)SXJ train)

[Andy Ellsworth]

On Tue, Jun 26, 2012 at 10:31 AM, Phil Mayers <p.mayers at imperial.ac.uk> wrote:
> In short - take it with a big pinch of salt, and don't rely on it.

Thanks for the feedback. The Cisco development engineer I just spoke
to essentially confirmed this - the open and resolved caveat lists in
the release notes are automatically generated, and are therefore only
as good as the metadata in the bugs.

Focusing on one example (CSCsm59426), he was able to look at the
source code and confirm that the bug was in fact fixed in the SXJ
train, but simply not marked as fixed in that version due to an
oversight in the sequence of forks/commits that led from 12.2 to
12.2SX to 12.2SXJ to SXJ3.

He hinted at some internal improvements in the pipeline that will add
additional intelligence to the bug checking process - improvements
that might actually propagate to the bug navigator (and thus improve
the accuracy of caveats listed in the release notes). Until that
happens, if you've got enough leverage, it's possible to arm-twist
Cisco into checking specific open caveats you may have red-flagged to
determine if they are actual bugs in the train you're reviewing.

-Andy