Re: adaptation sections

On Tue, 2008-03-25 at 12:28 +0100, Kinkie wrote:
> On Tue, Mar 25, 2008 at 5:14 AM, Alex Rousskov
> <rousskov@measurement-factory.com> wrote:
> > Should I use the same debugging sections for ICAP, eCAP, and the
> > code they share? Or should we have three distinct debugging sections for
> > those three areas?
>
> IMO the most intuitive approach would be to have one section for ICAP,
> one for eCAP and the shared code should log at the highest level among
> those of ICAP and eCAP

That's a good idea! The level can also change depending on the caller
(e.g., use eCAP level if eCAP code is calling the shared code).

> (this would require extending the logging code
> somehow tho, for instance adding a third "behind the scenes" section).

... and maintaining its level whenever primary section levels change.

I think this is the right direction, but probably too much custom but
not-essential work for the first release. We need better scope- and
context-specific debugging control anyway, and that feature will
probably help here as well.

Since opinions on this thread differed, I will probably start with three
named constants for adaptation-related debugging sections. All constants
will have the same value (93) for now. Since we do not have support for
multiple adaptation services per HTTP message, most installations will
use either ICAP or eCAP until that support is added.

Thanks to everybody who commented on this thread,

Alex.
Received on Tue Mar 25 2008 - 09:09:40 MDT