On Mon, Dec 19, 2005 at 08:44:55AM +0100, lukas.rueegg [pixworx multimedia] wrote: > hi > > we have a setup with kernel 2.6.14.4, just upgraded, patcht against > vserver 2.0.1. the hosts use bridged network interfaces vor L2 > redundancy. also, we are working with VLANs to separate traffic between > hosts and vservers. > > when we were installing util-vserver-0.30.209 every time we stopped a > vserver, not only his private interface got down, but the whole bridge > including all other interfaces of all vservers... > > after downgrading to util-vserver-0.30.207, everything works fine again. > > a sample configuration of a vserver: > $ more 0/* > :::::::::::::: > 0/dev > :::::::::::::: > br0.2 > :::::::::::::: > 0/ip > :::::::::::::: > 62.12.154.32 > :::::::::::::: > 0/name > :::::::::::::: > public > :::::::::::::: > 0/prefix > :::::::::::::: > 24 > > > the relevant part output part of ifconfig (on the host): > $ sudo ifconfig > > br0.2 Link encap:Ethernet HWaddr 00:11:43:E3:11:39 > inet addr:62.12.154.12 Bcast:62.255.255.255 Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:2808646 errors:0 dropped:0 overruns:0 frame:0 > TX packets:3383124 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:0 > RX bytes:424670986 (404.9 MiB) TX bytes:1168460377 (1.0 GiB) > > br0.2:pub Link encap:Ethernet HWaddr 00:11:43:E3:11:39 > inet addr:62.12.154.32 Bcast:62.12.154.255 Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > > the first is the host bridge br0.2. the second is the aliased interface > of the vserver. both are in VLAN 2. > > > output of ip addr list: > > $ ip addr list > 1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > inet 127.0.0.1/8 scope host lo > 2: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000 > link/ether 00:11:43:e3:11:39 brd ff:ff:ff:ff:ff:ff > 3: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000 > link/ether 00:11:43:e3:11:3a brd ff:ff:ff:ff:ff:ff > 4: eth2: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000 > link/ether 00:12:17:54:de:1e brd ff:ff:ff:ff:ff:ff > 5: br0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue > link/ether 00:11:43:e3:11:39 brd ff:ff:ff:ff:ff:ff > 6: br0.2: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue > link/ether 00:11:43:e3:11:39 brd ff:ff:ff:ff:ff:ff > inet 62.12.154.12/24 brd 62.255.255.255 scope global br0.2 > inet 62.12.154.102/24 brd 62.12.154.255 scope global secondary br0.2:0 > inet 62.12.154.57/24 brd 62.12.154.255 scope global secondary > br0.2:public > inet 62.12.154.56/24 brd 62.12.154.255 scope global secondary > br0.2:public > inet 62.12.154.32/24 brd 62.12.154.255 scope global secondary > br0.2:public
didn't even know that it was possible to do that? IMHO the interface names are supposed to be _unique_ which is definitely wrong here (for the aliases) I'd suggest to avoid the alias at all, unless you require them for very old tools (like ifconfig) > where all br0.2:public are vserver interfaces. > > we first where thinking that it could cause problems that all vserver > interface aliases are the same. but because also the host interface > (br0.2) gets down, we aren't convinced of this anymore... > > does anyone has a hint about this? well, as soon as the 62.12.154.12/24 IP is taken down, all others will follow. that's how the linux networking people consider it proper ... but as far as I understood, that one should not be used by a guest, or is it? if so, please assign a dummy IP to the host (e.g. 62.12.154.2/24) and be done ... HTH, Herbert > thanks and regards > lukas.rueegg > > > _______________________________________________ > Vserver mailing list > [email protected] > http://list.linux-vserver.org/mailman/listinfo/vserver _______________________________________________ Vserver mailing list [email protected] http://list.linux-vserver.org/mailman/listinfo/vserver
