Network stack unstable after arp flapping

K. Macy kmacy at freebsd.org
Mon Apr 4 10:25:39 UTC 2011


On Mon, Apr 4, 2011 at 12:22 PM, Frederique Rijsdijk
<frederique at isafeelin.org> wrote:
> Hi Kip,
>
> Thanks. Any idea btw why this would also effect hosts that run 6.x and 7.x RELEASE versions? These do not have the flowtable sysctl.


Uhm ... no. That would definitely be unrelated. I'm just taking their
word for it. And I can see how the flowtable might not handle ARP
flapping properly.

We'll need to do more diagnostics. So this is only IPv6 and you are using TCP?



-Kip

>
> -- Frederique
>
>
>
> On Mon, Apr 04, 2011 at 12:09:05PM +0200, K. Macy wrote:
>> Correct. Set it to zero and the flowtable will be bypassed.
>>
>> flowtable.c:
>> static uint32_t
>> ipv6_flow_lookup_hash_internal(
>>       struct sockaddr_in6 *ssin6, struct sockaddr_in6 *dsin6,
>>           uint32_t *key, uint16_t flags)
>> {
>>       uint16_t sport, dport;
>>       uint8_t proto;
>>       int offset = 0;
>>
>>       if ((V_flowtable_enable == 0) || (V_flowtable_ready == 0))
>>               return (0);
>>
>>
>> Ciao
>>
>> On Mon, Apr 4, 2011 at 11:53 AM, Frederique Rijsdijk
>> <frederique at isafeelin.org> wrote:
>> > Kip,
>> >
>> > Which sysctl are we talking about exactly? Just to be sure..
>> >
>> > net.inet.flowtable.enable ?
>> >
>> >
>> > -- Frederique
>> >
>> >
>> >
>> > On Sun, Apr 03, 2011 at 08:11:33PM +0200, K. Macy wrote:
>> >> I don't think it was properly tested when it was enabled for IPv6.
>> >> Given that I have been absentee it really should not be in the default
>> >> kernel or at least the sysctl should be off. Sorry for the
>> >> inconvenience. Additionally, you don't need to rebuild you can just
>> >> disable the sysctl.
>> >>
>> >>
>> >>  -Kip
>> >> On Sunday, April 3, 2011, Frederique Rijsdijk <frederique at isafeelin.org> wrote:
>> >> > Steve,
>> >> >
>> >> > On 01-04-11 16:50, Steve Polyack wrote:
>> >> >> On 04/01/11 10:16, Frederique Rijsdijk wrote:
>> >> > [ .. ]
>> >> >>> Mar 31 18:36:12 srv01 kernel: arp: x.x.x.1 moved from
>> >> >>> 00:00:0c:9f:f0:3d to 00:00:0c:07:ac:3d on bge0
>> >> > [ .. ]
>> >> >>> The result of that, is that loads of FreeBSD machines (6.x, 7.x and
>> >> >>> 8.x) developed serious network issues, mainly being no or slow traffic
>> >> > [ .. ]
>> >> >>> Any ideas anyone?
>> >> >> We experienced a similar issue here, but IIRC only on our 8.x systems
>> >> >> (we don't have any 7.x).  Disabling flowtable cleared everything up
>> >> >> immediately.  You can try that and see if it helps.
>> >> >
>> >> > AFAIK this feature was introduced in 8.x? Btw you are here referring to
>> >> > UDP, we had issues with TCP. It could still be related, perhaps I'll get
>> >> > around emulating the situation and see if I can reproduce it.
>> >> >
>> >> >
>> >> > -- Frederique
>> >> >
>> >> >
>> >> > _______________________________________________
>> >> > freebsd-net at freebsd.org mailing list
>> >> > http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> >> > To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>> >> >
>> >> _______________________________________________
>> >> freebsd-net at freebsd.org mailing list
>> >> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> >> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>> > _______________________________________________
>> > freebsd-net at freebsd.org mailing list
>> > http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> > To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>> >
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>


More information about the freebsd-net mailing list