From: Jeff King <peff@peff•net>
To: Junio C Hamano <gitster@pobox•com>
Cc: Dave Borowitz <dborowitz@google•com>, git@vger•kernel.org
Subject: Re: [PATCH 3/7] pack-protocol.txt: Mark all LFs in push-cert as required
Date: Fri, 3 Jul 2015 14:07:19 -0400 [thread overview]
Message-ID: <20150703180718.GB9223@peff.net> (raw)
In-Reply-To: <xmqq38155e3s.fsf@gitster.dls.corp.google.com>
On Fri, Jul 03, 2015 at 10:45:59AM -0700, Junio C Hamano wrote:
> > Usually flush packets are "0000", and an empty data packet
> > is "0004". Or are you talking about some kind of flush inside the
> > pkt-data stream?
>
> Neither. At the wire level there is a difference, but the callers
> of most often used function in pkt-line API, packet_read(), says
>
> while (1) {
> len = packet_read();
> if (!len) {
> /* flush */
> break;
> }
> ... do things on the "len" bytes received ...
> ... and then on to the next packet ...
> }
Ah, I see. Yeah, that is a problem. The solutions you proposed seem like
good workarounds to me, but we are unfortunately stuck with existing
clients and servers which did not behave that way.
I wondered briefly whether this impacted the keepalives we added to
`upload-pack` in 05e9515; those are implemented as 0-byte data packets,
which we send during the potentially long counting/delta-compression
phase before we send out pack data. It works there because the packets
actually contain a single sideband byte, so they are never mistaken for
a flush packet.
Related, I recently ran into a case where I think we should do the same
for pushes. After receiving the pack, `index-pack` may chew on the
result for literally minutes (try pushing up the entire linux.git
history sometime). We say nothing at all on the wire until we've
finished that, run check_everything_connected, and run all hooks. Some
clients (or intermediates on the connection) may give up after a few
minutes of silence.
I think we should have:
1. Some progress eye-candy from the server to tell us that something
is happening, and how close we are to finishing (basically
"index-pack -v").
2. When progress is disabled, similar keepalive packets saying "nope,
no output yet".
For (2), hopefully we can implement it in the same way, and rely on
empty sideband-0 packets. I haven't tested it in practice, though (I
have some very rough patches for (1) already).
-Peff
next prev parent reply other threads:[~2015-07-03 18:07 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-01 18:08 [PATCH 0/7] Clarify signed push protocol documentation Dave Borowitz
2015-07-01 18:08 ` [PATCH 1/7] pack-protocol.txt: Add warning about protocol inaccuracies Dave Borowitz
2015-07-01 19:39 ` Jonathan Nieder
2015-07-01 19:52 ` Junio C Hamano
2015-07-01 19:56 ` Dave Borowitz
2015-07-01 18:08 ` [PATCH 2/7] pack-protocol.txt: Mark LF in command-list as optional Dave Borowitz
2015-07-01 18:21 ` Stefan Beller
2015-07-01 18:46 ` Dave Borowitz
2015-07-01 18:08 ` [PATCH 3/7] pack-protocol.txt: Mark all LFs in push-cert as required Dave Borowitz
2015-07-01 20:00 ` Junio C Hamano
2015-07-01 20:07 ` Dave Borowitz
2015-07-01 20:49 ` Junio C Hamano
2015-07-06 14:46 ` Dave Borowitz
2015-07-06 15:22 ` Dave Borowitz
2015-07-06 15:27 ` Dave Borowitz
2015-07-06 15:29 ` Dave Borowitz
2015-07-06 15:35 ` Dave Borowitz
2015-07-06 16:12 ` Junio C Hamano
2015-07-06 15:46 ` Shawn Pearce
2015-07-06 16:28 ` Junio C Hamano
2015-07-06 16:28 ` Junio C Hamano
2015-07-06 16:38 ` Dave Borowitz
2015-07-06 16:57 ` Junio C Hamano
2015-07-06 17:11 ` Dave Borowitz
2015-07-06 17:18 ` Dave Borowitz
2015-07-06 17:34 ` Junio C Hamano
2015-07-06 17:38 ` Dave Borowitz
2015-07-06 18:06 ` Junio C Hamano
2015-07-06 18:08 ` Dave Borowitz
2015-07-06 18:23 ` Junio C Hamano
2015-07-06 17:30 ` Junio C Hamano
2015-07-06 17:35 ` Dave Borowitz
2015-07-06 17:59 ` Junio C Hamano
2015-07-01 20:36 ` Junio C Hamano
2015-07-01 20:39 ` Junio C Hamano
2015-07-02 13:53 ` Jeff King
2015-07-03 17:45 ` Junio C Hamano
2015-07-03 18:07 ` Jeff King [this message]
2015-07-03 18:43 ` Shawn Pearce
2015-07-03 18:46 ` Jeff King
2015-07-01 18:08 ` [PATCH 4/7] pack-protocol.txt: Elaborate on pusher identity Dave Borowitz
2015-07-01 18:58 ` Junio C Hamano
2015-07-01 18:08 ` [PATCH 5/7] pack-protocol.txt: Be more precise about pusher-key relationship Dave Borowitz
2015-07-01 18:08 ` [PATCH 6/7] pack-protocol.txt: Mark pushee field as optional Dave Borowitz
2015-07-01 18:56 ` Junio C Hamano
2015-07-01 19:06 ` Dave Borowitz
2015-07-01 19:07 ` Junio C Hamano
2015-07-01 19:08 ` Junio C Hamano
2015-07-01 19:31 ` Dave Borowitz
2015-07-01 18:08 ` [PATCH 7/7] send-pack.c: Die if the nonce is empty Dave Borowitz
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150703180718.GB9223@peff.net \
--to=peff@peff$(echo .)net \
--cc=dborowitz@google$(echo .)com \
--cc=git@vger$(echo .)kernel.org \
--cc=gitster@pobox$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox