From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcus Zoller Subject: strange routing -- Date: 12 Jun 2002 20:21:19 +0200 Sender: netfilter-admin@lists.samba.org Message-ID: <1023906079.26607.73.camel@Snoopy> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: Errors-To: netfilter-admin@lists.samba.org List-Help: List-Post: List-Subscribe: , List-Id: List-Unsubscribe: , List-Archive: Content-Type: text/plain; charset="us-ascii" To: netfilter@lists.samba.org Cc: kai@plocki.org Hi all, this is not realy related to netfilter but maybe someone here can help me to find what I did wrong/forgot,...: I get the following in the logs on my netfilter gateway, about every 2 seconds from two machines behind of the gateway that are in one subnet: REJECT: IN=eth0 OUT=eth0 SRC=10.10.0.218 DST=10.10.0.200 LEN=56 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=UDP SPT=111 DPT=50916 LEN=36 I'am wondering why the packet is sent to the gateway instead of sending it direct? 10.10.0.218 is a kernel 2.4.18 NFS server running portmap, ... 10.10.0.200 is a linux box with mounted directorys from that host. All 3 machines are in the same subnet 10.10.0.0/24. The gateway logging the message is the default gateway for both. Running tcpdump -nvi eth0 on the gateway shows exactly what gets logged: > arp who-has 10.10.0.218 tell 10.10.0.254 < arp reply 10.10.0.218 is-at 0:50:ba:e9:7:d3 (0:80:c8:ca:ce:41) < 10.10.0.218.sunrpc > 10.10.0.200.50919: udp 28 (DF) (ttl 64, id 0) > 10.10.0.254 > 10.10.0.218: icmp: 10.10.0.200 udp port 50919 unreachable [tos 0xc0] (ttl 255, id 49752) I checked the interface mask, broadcast and routing entries on both machines and they are as they should be. The interesting thing is the routing cache on 10.10.0.218 (it has another interface with 10.10.0.219 but this address never appears in the logs): # route -Cn | grep 10.10.0.200 10.10.0.200 10.10.0.218 10.10.0.218 l 0 0 152 lo 10.10.0.200 10.10.0.218 10.10.0.218 l 0 0 32611 lo 10.10.0.219 10.10.0.200 10.10.0.200 0 0 25713 eth1 10.10.0.218 10.10.0.200 10.10.0.200 0 0 1 eth1 10.10.0.218 10.10.0.200 10.10.0.200 0 1 0 eth1 10.10.0.218 10.10.0.200 10.10.0.254 0 0 1304 eth0 Rebooting the machines helps for about 10 minutes then the last entry is again in the cache. Interface config: inet Adresse:10.10.0.218 Bcast:10.10.0.255 Maske:255.255.255.0 inet Adresse:10.10.0.219 Bcast:10.10.0.255 Maske:255.255.255.0 Many thanks for reading this and for any help/ideas!!! -marcus