Long-haul problems - connections stuck in slow start

[Peter Wemm]

On Sat, Jan 4, 2014 at 11:20 AM, Peter Wemm <peter at wemm.org> wrote:
> On 1/4/14, 10:38 AM, Peter Wemm wrote:
>> We're seeing some unfortunate misbehavior with tcp over an intercontinental
>> link.
>
> I forgot to mention, the socket buffers have been generously tuned for these
> endpoints. They don't seem to be being used while the sender is in some sort
> of limited transmit mode.
>
> When reading the tcpdumps, keep in mind the window scaling factor is 11 and
> that's wasn't captured in the session.  You'll have to manually compensate
> when reading the mid-session dumps.
>
> XXX footnote:  It seems turning SACK off makes a huge difference for this
> connection.  The server must have been running with sack disabled when it
> was working and hadn't been saved in sysctl.conf. Turning sack off on the
> server again has raised the throughput from 8K/sec to 32MB/sec.  That's a
> nice 400x speedup.
>
> I'm investigating.
>

It's still not solved.  We just had a lucky break and the session ran
fast for a while.  SACK doesn't seem to be the variable.

I'm looking at the duplicate acks in this trace from each end..  any
clues where they're coming from?

http://people.freebsd.org/~peter/acks.txt

-- 
Peter Wemm - peter at wemm.org; peter at FreeBSD.org; peter at yahoo-inc.com; KI6FJV
Yes, I know, gmail sucks now. If you see this then I forgot. Habits
are hard to break.