netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jesse Hathaway <jesse@mbuki-mvuki.org>
To: netdev@vger.kernel.org
Subject: Race condition in route lookup
Date: Wed, 9 Oct 2019 11:00:07 -0500	[thread overview]
Message-ID: <CANSNSoV1M9stB7CnUcEhsz3FHi4NV_yrBtpYsZ205+rqnvMbvA@mail.gmail.com> (raw)

We have been experiencing a route lookup race condition on our internet facing
Linux routers. I have been able to reproduce the issue, but would love more
help in isolating the cause.

Looking up a route found in the main table returns `*` rather than the directly
connected interface about once for every 10-20 million requests. From my
reading of the iproute2 source code an asterisk is indicative of the kernel
returning and interface index of 0 rather than the correct directly connected
interface.

This is reproducible with the following bash snippet on 5.4-rc2:

  $ cat route-race
  #!/bin/bash

  # Generate 50 million individual route gets to feed as batch input to `ip`
  function ip-cmds() {
          route_get='route get 192.168.11.142 from 192.168.180.10 iif vlan180'
          for ((i = 0; i < 50000000; i++)); do
                  printf '%s\n' "${route_get}"
          done

  }

  ip-cmds | ip -d -o -batch - | grep -E 'dev \*' | uniq -c

Example output:

  $ ./route-race
        6 unicast 192.168.11.142 from 192.168.180.10 dev * table main
\    cache iif vlan180

These routers have multiple routing tables and are ingesting full BGP routing
tables from multiple ISPs:

  $ ip route show table all | wc -l
  3105543

  $ ip route show table main | wc -l
  54

Please let me know what other information I can provide, thanks in advance,
Jesse Hathaway

             reply	other threads:[~2019-10-09 16:00 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-09 16:00 Jesse Hathaway [this message]
2019-10-10  8:31 ` Race condition in route lookup Ido Schimmel
2019-10-10  8:46   ` Ido Schimmel
2019-10-11 14:36   ` Jesse Hathaway
2019-10-11 15:42     ` Ido Schimmel
2019-10-11 16:09       ` Jesse Hathaway
2019-10-11 17:54       ` Wei Wang
2019-10-11 18:17         ` Ido Schimmel
2019-10-11 18:25           ` Ido Schimmel
2019-10-11 18:47             ` Wei Wang
2019-10-11 18:52               ` Ido Schimmel
2019-10-11 21:01                 ` Jesse Hathaway
2019-10-11 21:27                 ` David Ahern
2019-10-12  6:56         ` Martin Lau
2019-10-14  0:23           ` Wei Wang
2019-10-14 17:26             ` Martin Lau
2019-10-15 14:45               ` David Ahern
2019-10-15 16:42                 ` Wei Wang
2019-10-16  6:35                   ` Martin Lau
2019-10-15 14:29         ` Jesse Hathaway
2019-10-15 16:44           ` Wei Wang
2019-10-16  6:39             ` Martin Lau
2019-10-16 16:35               ` Wei Wang

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=CANSNSoV1M9stB7CnUcEhsz3FHi4NV_yrBtpYsZ205+rqnvMbvA@mail.gmail.com \
    --to=jesse@mbuki-mvuki.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).