Fw: [Bug 85091] New: neighbour table overflow is not reported and impacts localhost TCP connectivity

Fw: [Bug 85091] New: neighbour table overflow is not reported and impacts localhost TCP connectivity

[Stephen Hemminger]

Begin forwarded message:

Date: Wed, 24 Sep 2014 03:44:41 -0700
From: "bugzilla-daemon@bugzilla.kernel.org" <bugzilla-daemon@bugzilla.kernel.org>
To: "stephen@networkplumber.org" <stephen@networkplumber.org>
Subject: [Bug 85091] New: neighbour table overflow is not reported and impacts localhost TCP connectivity


https://bugzilla.kernel.org/show_bug.cgi?id=85091

            Bug ID: 85091
           Summary: neighbour table overflow is not reported and impacts
                    localhost TCP connectivity
           Product: Networking
           Version: 2.5
    Kernel Version: 3.8 - 3.17-rc6
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: IPV4
          Assignee: shemminger@linux-foundation.org
          Reporter: aschultz@tpip.net
        Regression: No

With the default gc_thresh values and a busy /16 network attached, the
neighbour cache can overflow. No indication is given that this happens and it
does impact TCP on localhost.

Test setup:

* about 16k (simulated IP/MAC's) on one interface
* web server behind on second interface
* routing between the two
* HTTP benchmark from the 16k IP's to the web server
* for localhost connectivity verification a netperf instance is run on
localhost like so: 'netperf -D 1 -l 600 127.0.0.1'

Result:

Kernel has to learn the 16k IP/MAC combinations, as soon as gc_thresh3 is hit,
netperf stalls, no syslog/kernel message indicates the problem.

The only indication are log entries like this:

  "net_ratelimit: 1464 callbacks suppressed"

No other messages are logged.

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

Re: Fw: [Bug 85091] New: neighbour table overflow is not reported and impacts localhost TCP connectivity

[Cong Wang]

On Wed, Sep 24, 2014 at 9:32 AM, Stephen Hemminger
<stephen@networkplumber.org> wrote:
> Result:
>
> Kernel has to learn the 16k IP/MAC combinations, as soon as gc_thresh3 is hit,
> netperf stalls, no syslog/kernel message indicates the problem.
>
> The only indication are log entries like this:
>
>   "net_ratelimit: 1464 callbacks suppressed"
>
> No other messages are logged.
>

Looks like we have a race condition between we check gc_thresh3
and add the new entry into the table after locking in __neigh_create().
Otherwise the application layer should get EINVAL to stop.