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
> Vserver@list.linux-vserver.org
> http://list.linux-vserver.org/mailman/listinfo/vserver
_______________________________________________
Vserver mailing list
Vserver@list.linux-vserver.org
http://list.linux-vserver.org/mailman/listinfo/vserver
Received on Mon Dec 19 16:19:44 2005