my host/dom0: SunOS klaatu 5.11 snv_113 i86pc i386 i86xpv Solaris
i created a centos 5.3 VM using virt-install. i'm using a zfs fs for the image
-
rpool/centos-img 8.00G 222G 16K -
i may not be understanding the networking properly, so bear with me.
i have my own delegated /28 at home -
206.176.249.129, gateway
my dom0 machine has several IP's on it for service partitioning (mail, web,
etc):
root-klaatu /root% ifconfig -a
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232
index 1
inet 127.0.0.1 netmask ff000000
rge0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 206.176.249.135 netmask fffffff0 broadcast 206.176.249.143
ether 0:22:68:5c:38:4a
rge0:2: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 206.176.249.136 netmask fffffff0 broadcast 206.176.249.143
rge0:3: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 206.176.249.138 netmask fffffff0 broadcast 206.176.249.143
rge0:4: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 206.176.249.139 netmask fffffff0 broadcast 206.176.249.143
lo0: flags=2002000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv6,VIRTUAL> mtu 8252
index 1
inet6 ::1/128
i'm using one of my free/unallocated IP's for the centos vm - 206.176.249.134.
on the centos side, i have the following ifcfg-eth0 configured:
[r...@uvula ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
# Xen Virtual Ethernet
DEVICE=eth0
BOOTPROTO=static
IPADDR=206.176.249.134
NETMASK=255.255.255.240
GATEWAY=206.176.249.129
ONBOOT=yes
on my dom0, dladmin shows that the virtual interface is indeed there, with the
right mac address:
root-klaatu /root% dladm show-vnic
LINK OVER SPEED MACADDRESS MACADDRTYPE VID
xvm3_0 rge0 100 0:16:3e:52:c2:3b fixed 0
here's where things get weird. i can ping out from my centos VM to anyplace on
the net - no problem. from a shell account i have in another part of the
country, i can ping in to my VM no problem, too. however, from the 'outside', i
can't ssh into the VM. i can ssh to it from my local host machine, but not from
a remote host - even though a netstat shows an "ESTABLISHED" connection when i
try, the remote side is non-responsive, never showing a completed connection.
and when i attempt to download files from the net, the connection just hangs -
[r...@uvula src]# wget
http://downloads.sourceforge.net/zenoss/zenoss-2.4.1.el5.x86_64.rpm
--18:35:44--
http://downloads.sourceforge.net/zenoss/zenoss-2.4.1.el5.x86_64.rpm
Resolving downloads.sourceforge.net... 216.34.181.59
Connecting to downloads.sourceforge.net|216.34.181.59|:80... connected.
HTTP request sent, awaiting response...
i'm baffled. the firewall on the centos vm is disabled. a virsh dumpxml shows
that the interface is in bridged mode as it should be:
interface type='bridge'
source bridge='rge0'/
target dev='vif3.0'/
mac address='00:16:3e:52:c2:3b'/
script path='vif-vnic'/
/interface
(had to take out the angle brackets for that to show up in the post)
i hope i'm not missing something tremendously obvious. i sure would like to get
this going. it all arose because i spent a day and a half tearing my hair out
trying to get the source version of zenoss running on my solaris host, to no
avail (the zenoss source for solaris is a mess, as others on the net have
attested). i figured my 'easiest' (ha) way around the problem was to get a VM
going, as at work i'm running zenoss is a centos VM (albeit on a centos dom0).
--
This message posted from opensolaris.org
_______________________________________________
xen-discuss mailing list
[email protected]
