From: Neal Cardwell <ncardwell@google.com>
To: Apollon Oikonomopoulos <apoikos@dmesg.gr>
Cc: Yuchung Cheng <ycheng@google.com>,
Netdev <netdev@vger.kernel.org>,
Eric Dumazet <edumazet@google.com>,
Soheil Hassas Yeganeh <soheil@google.com>
Subject: Re: TCP sender stuck in persist despite peer advertising non-zero window
Date: Thu, 15 Oct 2020 18:37:37 -0400 [thread overview]
Message-ID: <CADVnQym2cJGRP8JnRAdzHfWEeEbZrmXd3eXD-nFP6pRNK7beWw@mail.gmail.com> (raw)
In-Reply-To: <87blh33zr7.fsf@marvin.dmesg.gr>
[-- Attachment #1: Type: text/plain, Size: 1690 bytes --]
On Thu, Oct 15, 2020 at 6:12 PM Apollon Oikonomopoulos <apoikos@dmesg.gr> wrote:
>
> Yuchung Cheng <ycheng@google.com> writes:
>
> > On Thu, Oct 15, 2020 at 1:22 PM Neal Cardwell <ncardwell@google.com> wrote:
> >>
> >> On Thu, Oct 15, 2020 at 2:31 PM Apollon Oikonomopoulos <apoikos@dmesg.gr> wrote:
> >> >
> >> > Hi,
> >> >
> >> > I'm trying to debug a (possible) TCP issue we have been encountering
> >> > sporadically during the past couple of years. Currently we're running
> >> > 4.9.144, but we've been observing this since at least 3.16.
> >> >
> >> > Tl;DR: I believe we are seeing a case where snd_wl1 fails to be properly
> >> > updated, leading to inability to recover from a TCP persist state and
> >> > would appreciate some help debugging this.
> >>
> >> Thanks for the detailed report and diagnosis. I think we may need a
> >> fix something like the following patch below.
>
> That was fast, thank you!
>
> >>
> >> Eric/Yuchung/Soheil, what do you think?
> > wow hard to believe how old this bug can be. The patch looks good but
> > can Apollon verify this patch fix the issue?
>
> Sure, I can give it a try and let the systems do their thing for a couple of
> days, which should be enough to see if it's fixed.
Great, thanks!
> Neal, would it be possible to re-send the patch as an attachment? The
> inlined version does not apply cleanly due to linewrapping and
> whitespace changes and, although I can re-type it, I would prefer to test
> the exact same thing that would be merged.
Sure, I have attached the "git format-patch" format of the commit. It
does seem to apply cleanly to the v4.9.144 kernel you mentioned you
are using.
Thanks for testing this!
best,
neal
[-- Attachment #2: 0001-tcp-fix-to-update-snd_wl1-in-bulk-receiver-fast-path.patch --]
[-- Type: application/x-patch, Size: 1363 bytes --]
next prev parent reply other threads:[~2020-10-15 22:37 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-15 18:23 TCP sender stuck in persist despite peer advertising non-zero window Apollon Oikonomopoulos
2020-10-15 20:22 ` Neal Cardwell
2020-10-15 20:27 ` Soheil Hassas Yeganeh
2020-10-15 21:39 ` Yuchung Cheng
2020-10-15 22:12 ` Apollon Oikonomopoulos
2020-10-15 22:37 ` Neal Cardwell [this message]
2020-10-16 7:35 ` Eric Dumazet
2020-10-16 16:57 ` Apollon Oikonomopoulos
2020-10-16 17:54 ` Neal Cardwell
2020-10-22 12:47 ` Apollon Oikonomopoulos
2020-10-22 14:39 ` Neal Cardwell
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=CADVnQym2cJGRP8JnRAdzHfWEeEbZrmXd3eXD-nFP6pRNK7beWw@mail.gmail.com \
--to=ncardwell@google.com \
--cc=apoikos@dmesg.gr \
--cc=edumazet@google.com \
--cc=netdev@vger.kernel.org \
--cc=soheil@google.com \
--cc=ycheng@google.com \
/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).