Re: [PATCH 3/3] pkt-line: support tracing verbatim pack contents

On Tue, Jun 16, 2015 at 1:18 PM, Jeff King [off-list ref] wrote:

On Tue, Jun 16, 2015 at 01:14:03PM -0400, Augie Fackler wrote:

quoted

Yeah, not having it for the push side is a slight bummer, but in
general I haven't had problems debugging git clients pushing bogus
data in the same way that I've had problems with weirdness in new
server features.

Being in charge of a large git server farm, I think I have the opposite
problem. :)

I've got plenty of servers too, but we haven't typically seen
push-time problems. Maybe we're lucky, and now that I've said
something and tempted Murphy I'll suffer torment. :)

quoted

quoted

Here's a rough cut at the "trace stdin" idea I mentioned earlier (which
is essentially an internal "tee"). You can collect the incoming pack
like:

Neat, but not sure I like the extra overhead of having to grab the
full trace and then reconstruct some arguments to be able to diagnose
the pack. Having the verbatim pack just land on disk is really handy,
because then any existing tools one has cooked up (my team has a few
weird one-off ones by now) just work without extra fussing or looking
up steps to reconstruct the whole file.

I guess there is really room for both. Just because you _can_ accomplish
the same thing with both does not mean we cannot have two ways to do it
(an easy way, and a harder, more flexible way).

*nod* that might make the most sense - given that we both seem to have
use cases in mind for verbatim packs on pulls, that seems like a good
thing to have easy to deploy.

(It still seems to me more likely to write custom servers than
clients, but then again custom VCS servers has been my life for a
while, so I likely have a weird perspective on things.)

-Peff