From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2) Gecko/20021202 Description of problem: We're observing a strange behavior of a server with 3 Ethernet-Interfaces (1 onboard, 2 PCI-Cards eepro100). On interface eth2 the system replies with its own ip address and the ip address of eth1 but the mac address of eth2. eth1 is connected to one loadbalancer (connected via switch) and eth2 is connected to another loadbalancer (connected via different switch). each interface has its own ip address. problem: a loadbalancer gets the same mac address for both different ip adresses. :-/ tcpdump: 16:00:08.715611 arp who-has lnxc-500.srv.domain.de tell prx-vtx-2-b.app.domain.de 16:00:08.715629 arp reply lnxc-500.srv.domain.de is-at 0:2:b3:b8:6c:c9 16:00:14.544623 arp who-has prx-vtx-2-b.app.domain.de tell prx-vtx-1-b.app.domain.de 16:00:14.544743 arp reply prx-vtx-2-b.app.domain.de is-at 0:2:b3:bb:f0:de 16:01:06.164925 arp who-has prx-vtx-2-b.app.domain.de tell prx-vtx-1-b.app.domain.de 16:01:06.165047 arp reply prx-vtx-2-b.app.domain.de is-at 0:2:b3:bb:f0:de 16:01:07.288439 arp who-has lnxc-500.srv.domain.de tell prx-vtx-2-b.app.domain.de ++> 16:01:07.288455 arp reply lnxc-500.srv.domain.de is-at 0:2:b3:b8:6c:c9 16:01:21.999273 arp who-has prx-vtx-1-b.app.domain.de tell x.y.240.161 ++> 16:01:21.999304 arp reply prx-vtx-1-b.app.domain.de is-at 0:2:b3:b8:6c:c9 see with ++> marked line # host lnxc-500.srv.domain.de lnxc-500.srv.domain.de has address x.y.240.164 # host prx-vtx-1-b.app.domain.de prx-vtx-1-b.app.domain.de has address x.y.240.165 eth1 Link encap:Ethernet HWaddr 00:02:B3:BB:ED:43 inet addr:x.y.240.164 Bcast:x.y.240.175 Mask:255.255.255.240 UP BROADCAST MULTICAST MTU:1500 Metric:1 eth2 Link encap:Ethernet HWaddr 00:02:B3:B8:6C:C9 inet addr:x.y.240.165 Bcast:x.y.240.175 Mask:255.255.255.240 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Is this a known kernel bug? Version-Release number of selected component (if applicable): kernel-2.4.18-19.7.x How reproducible: Always Steps to Reproduce: 1. power on switches 2. power on load-balancer 3. power on server Actual Results: after some time you see both ip addresses on one mac adresses (always that of eth2) Expected Results: only eth2's ip and mac on eth2 (at least not the ip of eth1) Additional info:
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/