[Openstack] [Nova Network] Hosts lost connectivity but not instances
Naman Muley
naman.g.muley at gmail.com
Wed May 7 13:37:04 UTC 2014
Hello,
I have a multi-node install of openstack havana. Until yesterday I had all
the nodes and instances with proper connectivity, using FlatDHCPManager.
Today, when I tried to check connectivity, all compute hosts are unable to
send a ping outside the machine. Curiously, the instances are still able to
use the br100 and ping outside their respective hosts. Also, I am able to
ping the host from outside. (i.e. ping works in reverse direction).
=========
Ping from host(wind - eth0:10.10.1.13) to outside(10.10.1.10)
==========
xxxx at wind:~$ ping 10.10.1.10
PING 10.10.1.10 (10.10.1.10) 56(84) bytes of data.
>From 10.10.1.1 icmp_seq=1 Destination Host Unreachable
>From 10.10.1.1 icmp_seq=2 Destination Host Unreachable
>From 10.10.1.1 icmp_seq=3 Destination Host Unreachable
It is curious that ping says "From 10.10.1.1". I am unable to understand
this. It is as if br100 has taken over eth0 as the default interface.
===========
route -n on host(wind)
===========
ncsu at wind:~$ route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use
Iface
0.0.0.0 192.168.1.10 0.0.0.0 UG 100 0 0 eth1
10.0.3.0 0.0.0.0 255.255.255.0 U 0 0 0
lxcbr0
10.10.1.0 0.0.0.0 255.255.255.0 U 0 0 0
br100
10.10.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth1
==========================
relevant interface information for wind
=============================
eth0 - 10.10.1.13/24
br100 - 10.10.1.1
eth1 - 192.168.1.13 (mgmt interface)
I've tried restarting nova services and rebooting to no avail. I am unable
to troubleshoot it. Any ideas?
Naman
ps: I did notice that I was tinkering with NATs and iptables before this
happened.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20140507/9db4f058/attachment.html>
More information about the Openstack
mailing list