[lvs-users] Timeouts after 5 seconds running IPVS-DR

Julien Cornuwel cornuwel at gmail.com
Tue Oct 7 09:45:13 BST 2008


2008/10/6 Laurentiu C. Badea (L.C.) <lc at waat.com>

> A tcpdump should tell you where the problem is.


Well, not really :(
I did a TCP dump on all 3 servers and here is what I see :

- On webservers, when it works, I see outgoing IP packets with the LB's
address as origin. When it doesn't, I just see nothing. About once per
second, the LB sends ARP requests trying to find both webservers (on their
real addresses), I never saw an ARP reply.
- On the load balancer, I see incomming requests from clients, and some
"ICMP host lb" not reachable sent to the client when it doesn't work.

Webservers should reply to ARP requests on their primary addresses, but they
don't :(

I missed the beginning of this thread so I'm just throwing this out
> there. When I played with LVS on the same network as the RIPs, I had
> send_redirects=0, otherwise the LVS would mislead the real servers that
> they should send their packets to the router instead (where they'd get
> dropped).


Were you using DirectReturn ?


> Oh, and I also had arp_ignore=2 though not sure if that made a difference.


Don't think so, this just limit ARP replies to requesters that are in the
same subnet...



More information about the lvs-users mailing list