First Cup of Ubuntu
I have several Ubuntu 12.04.2 64bit server VMs running under XenServer 6.1.
Whenever they are left for ~30 mins with no traffic over ipv6, the network becomes unavailable, requiring the restart of the networking service to bring it back up. IPv4 is still active.
They are able to ping other VMs running on the same Host OS, but nothing above the host, even though the host and other VMs have IPv6 access.Code::~# ping6 google.com connect: Network is unreachable
Any help is appreciated. Wader
Last edited by iWader; March 25th, 2013 at 12:18 AM.
Ubuntu addict and loving it
What kind of IPv6 are you using?
What is the output of:
Code:mtr -nrc2 ipv6.google.com dmesg | grep -i ipv6 ip -6 addr show
First Cup of Ubuntu
Its native IPv6 provided by OVH
mtr -nrc2 ipv6.google.com
dmesg | grep -i ipv6Code::~# mtr -nrc2 ipv6.google.com HOST: lithium.iwader.co.uk Loss% Snt Last Avg Best Wrst StDev
ip -6 addr showCode::~# dmesg | grep -i ipv6 [ 2.693689] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready [ 2.693694] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready
Code::~# ip -6 addr show 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qlen 1000 inet6 2607:5300:60:2f30::2/64 scope global valid_lft forever preferred_lft forever inet6 fe80::ff:fe5b:19b/64 scope link valid_lft forever preferred_lft forever
Last edited by iWader; March 25th, 2013 at 04:41 PM.
Ubuntu addict and loving it
What kind of network is between the host and the guest OS? NAT? Bridged? ... ?
Maybe the first next hop (host OS or router) forgets the route to guest OS. Or maybe the guest OS does not answer ICMP6-who-has requests?
First Cup of Ubuntu
Its running under routed networking. My host (OVH) doesn't allow bridged adapters.Originally Posted by sanderj
The kernel provided by OVH for the host os (XenServer 6.1) its pre-configured for both ipv4 and ipv6 routed networking, so i dont think the issues lies there.
Ubuntu addict and loving it
Routed? So the Host is taking care of *routing*?
Oh, wait: is the XenServer provided by OVH? If so, ask them what is going on.
Ubuntu addict and loving it
PS:
What does
tell you?Code:rdisc6 eth0
First Cup of Ubuntu
Yes, all traffic from the VMs must be routed through the host os rather than directly connecting to their network. I already have a post on their forums, not very active though, doubt i would get help via. ticket
Code::~# rdisc6 eth0 Soliciting ff02::2 (ff02::2) on eth0... Hop limit : 64 ( 0x40) Stateful address conf. : No Stateful other conf. : No Router preference : medium Router lifetime : 1800 (0x00000708) seconds Reachable time : unspecified (0x00000000) Retransmit time : unspecified (0x00000000) Source link-layer address: 6C:9C:ED:BA:EC:40 MTU : 1500 bytes (valid) Prefix : 2607:5300:60:2f00::/56 Valid time : 2592000 (0x00278d00) seconds Pref. time : 604800 (0x00093a80) seconds from fe80::6e9c:edff:feba:ec40 Hop limit : 64 ( 0x40) Stateful address conf. : No Stateful other conf. : No Router preference : medium Router lifetime : 1800 (0x00000708) seconds Reachable time : unspecified (0x00000000) Retransmit time : unspecified (0x00000000) Source link-layer address: 6C:9C:ED:BA:EB:40 MTU : 1500 bytes (valid) Prefix : 2607:5300:60:2f00::/56 Valid time : 2592000 (0x00278d00) seconds Pref. time : 604800 (0x00093a80) seconds from fe80::6e9c:edff:feba:eb40
Grande Half-n-Half Cinnamon Ubuntu
So, are there supposed to be two upstream routers?
Posting Permissions
Adv Reply
Bookmarks