On Tue, Oct 10, 2000 at 06:03:44PM +0200, bert hubert wrote: > On Tue, Oct 10, 2000 at 04:21:06PM +0200, Daniel Bergqvist wrote: > > The logs are at http://www.bergqvist.se/teql/. > > > > The speed is about 1.3Mbit/s. > > There is lots of packetloss. Also, it is obvious from this dump that > lan_router is sending both over eth1 and eth2, but that the wan router is > only receiving on eth2, and not on eth1, or that your dump failed. On second thought, there is no packet loss. This is expected behaviour, it appears, see http://www.kernelnotes.de/kt/latest.html: Alexey Kuznetsov was critical of this explanation, and said that multipath routing worked "perfectly when you need to split load on servers talking to enough large number of clients. Any http server is good example." He added that Andi's suggestion of the existing eql, teql and bonding devices, would introcude "even worse problem of strong tcp reordering. Actually, experiments show that load balancing works only in the situations, when congestion window is bounded by 3 packets. If it is not made artificially, it occurs automatically on each connection after some amount of excessive retransmissions. Total single TCP connection throughput is never better in this case. Actually, it hints to the thought that "true load blalancing" has to involve tracking connections and avoiding reordering TCP packets." There was no reply to this, but there was a bit of implementation discussion elsewhere, along the lines of Andi's explanations. ---- You might consider using google a bit to find out about packet reordering - packets arrive out of sequence on eth1 and eth2, which the kernel interprets as packetloss. Regards, bert hubert -- PowerDNS Versatile DNS Services Trilab The Technology People 'SYN! .. SYN|ACK! .. ACK!' - the mating call of the internet
