"David S. Miller" wrote:
>
> Can you show the routing tables? They are all on the same network,
> so the kernel can choose whichever interface it pleases to send the
> packets out, and if your routing tables are how I believe they are,
> the kernel isn't making new 10.10.10.0 network destination routes
> for eth1 and eth2 because one already exists for eth0.
(routing table is first attachment, below)
Given that there is a network route for each eth interface I was
surprised when the traffic starting "migrating" to another interface.
The reason I think this is a problem is because it seems to break the
theory that multiple interfaces can help speed things up -- I ran
ApacheBench (ab) against 10.10.10.133 (eth1) and 10.10.10.134 (eth2),
but ifconfig only shows eth0 and eth1 as receiving packets.
Is there a sysctl or similar setting which will allow me to at least
advertise all three interfaces via arp? Since the machine advertises
the same ethernet address for .133 and .134, it is rather tough to test
10.10.10.134 (eth2) at all.
> I think you need to do a tiny bit of study on how IP routing
> works :-)
I don't doubt that at all..... :)
Though it seems really bogus to send packets to 10.10.10.134, eth2, only
to have them received by eth1. That would seem to break network load
balancing by admins who use multiple NICs.
Jeff
-- Jeff Garzik | "Vegetarian" is the Indian word Building 1024 | for 'lousy hunter.' MandrakeSoft, Inc. |
Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface 10.10.10.134 0.0.0.0 255.255.255.255 UH 40 0 0 eth2 10.10.10.133 0.0.0.0 255.255.255.255 UH 40 0 0 eth1 10.10.10.100 0.0.0.0 255.255.255.255 UH 40 0 0 eth0 10.10.10.0 0.0.0.0 255.255.255.0 U 40 0 0 eth0 10.10.10.0 0.0.0.0 255.255.255.0 U 40 0 0 eth1 10.10.10.0 0.0.0.0 255.255.255.0 U 40 0 0 eth2 127.0.0.0 0.0.0.0 255.0.0.0 U 40 0 0 lo 0.0.0.0 10.10.10.1 0.0.0.0 UG 40 0 0 eth0
eth0 Link encap:Ethernet HWaddr 00:40:F6:F4:FF:EF inet addr:10.10.10.100 Bcast:10.10.10.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:28441 errors:0 dropped:0 overruns:0 frame:0 TX packets:40635 errors:0 dropped:0 overruns:0 carrier:0 collisions:1401 txqueuelen:100 Interrupt:19
eth1 Link encap:Ethernet HWaddr 00:40:F6:F4:FF:73 inet addr:10.10.10.133 Bcast:10.10.10.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:13109 errors:0 dropped:0 overruns:0 frame:0 TX packets:32 errors:0 dropped:0 overruns:0 carrier:0 collisions:1 txqueuelen:100 Interrupt:18
eth2 Link encap:Ethernet HWaddr 00:00:21:DC:5A:9A inet addr:10.10.10.134 Bcast:10.10.10.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:304 errors:0 dropped:0 overruns:0 frame:0 TX packets:31 errors:0 dropped:0 overruns:0 carrier:0 collisions:1 txqueuelen:100 Interrupt:17
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:3924 Metric:1 RX packets:130 errors:0 dropped:0 overruns:0 frame:0 TX packets:130 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0
? (10.10.10.100) at 00:40:F6:F4:FF:EF [ether] on eth0 ? (10.10.10.1) at 00:40:05:17:92:0B [ether] on eth0 ? (10.10.10.166) at 00:A0:CC:50:E4:4C [ether] on eth0 ? (10.10.10.133) at 00:40:F6:F4:FF:73 [ether] on eth0 ? (10.10.10.134) at 00:40:F6:F4:FF:EF [ether] on eth0
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:24 EST