netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: netdev@vger.kernel.org
Subject: Fw: [Bug 195617] New: TCP may send small segments limited to initial window size
Date: Sat, 29 Apr 2017 23:01:15 -0700	[thread overview]
Message-ID: <20170429230115.6b30ef13@xeon-e3> (raw)



Begin forwarded message:

Date: Sun, 30 Apr 2017 00:50:22 +0000
From: bugzilla-daemon@bugzilla.kernel.org
To: stephen@networkplumber.org
Subject: [Bug 195617] New: TCP may send small segments limited to initial window size


https://bugzilla.kernel.org/show_bug.cgi?id=195617

            Bug ID: 195617
           Summary: TCP may send small segments limited to initial window
                    size
           Product: Networking
           Version: 2.5
    Kernel Version: 4.9.24
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: IPV4
          Assignee: stephen@networkplumber.org
          Reporter: ambrop7@gmail.com
        Regression: No

Created attachment 256141
  --> https://bugzilla.kernel.org/attachment.cgi?id=256141&action=edit  
packet trace

I am seeing what I believe is faulty behavior when a Linux client connects to a
TCP/IPv4 server which advertises a very small initial window. After the client
sends a small segment limited to this small window, the server ACKs the data
and extends the window considerably, but the client proceeds to send very small
segments.

Here is a summary of the packets:
- Client: SYN
- Server: SYN,ACK,Win=1
- Client: ACK
- Client: ACK,Data=1B
- Server: ACK,Win=49984
- Client: ACK,Data=47B
- Client: ACK,Data=48B
- Client: ACK,Data=48B
- Client: ACK,Data=48B
...

This example was with initially advertised window (WIN) equal to 1, but
generally it appears that Linux will limit outgoing segments to no more than
max(48,WIN).

I have tested the server with a Windows 10 client which does not have this
problem, the client starts sending fully sized segments when the window is
extended.

-- 
You are receiving this mail because:
You are the assignee for the bug.

                 reply	other threads:[~2017-04-30  6:01 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20170429230115.6b30ef13@xeon-e3 \
    --to=stephen@networkplumber.org \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
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;
as well as URLs for NNTP newsgroup(s).