[ntp:hackers] Use MSG_CONFIRM flag to keep OS ARP table up to date?

Kurt Roeckx kurt at roeckx.be
Sun May 6 20:37:58 UTC 2012


On Wed, May 02, 2012 at 12:10:25PM +0200, Heiko Gerstung wrote:
> Hi!
> 
> We are currently running stress tests with more than 100k simulated NTP clients against one single NTP server and found out
> that Linux is sending ARP queries for every single NTP client every 30s, putting significant load onto the NTP server
> (running NTP 4.2.6p3) we use as the NTP reference for the simulated clients.

And those 100k NTP clients are all in the local subnet?  I only
have my gateway in the ARP table, and that's really all I need.

> This seems to be caused by Linux not automatically updating the ARP cache when NTP sends its NTP responses to the clients,
> resulting in all ARP table entries to become stale after a period of 30s (this can be changed but would require users to
> mess with their kernel / network configuration). This triggers a new ARP request for each stale table entry

Yes, that's normal and expected behaviour.  And if you want to
avoid ARP and other broadcast storms, I suggest you use smaller
subnets.


Kurt



More information about the hackers mailing list