Actually, I found that the tcpdump -i 'tapf9884dcb-e5' got dhcp reply. However,
tcpdump -i br-int only has dhcp request.
Also, when I tried to get ip address on the dhcp agent, I was not able to see
'tapf9884dcb-e5'
ip netns exec qdhcp-7e73481a-d100-46d4-8dd1-9b3bb7949333
No cmd specified
[root@blade04 quantum]# ip netns exec
qdhcp-7e73481a-d100-46d4-8dd1-9b3bb7949333 ip address
15: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
18: ns-f9884dcb-e5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP qlen 1000
link/ether fa:16:3e:a1:d5:e2 brd ff:ff:ff:ff:ff:ff
inet 30.0.0.3/24 brd 30.0.0.255 scope global ns-f9884dcb-e5
inet6 fe80::f816:3eff:fea1:d5e2/64 scope link
valid_lft forever preferred_lft forever
Would this be an issue?
Thanks,
YuLing
From: C, Yuling
Sent: Tuesday, December 17, 2013 9:37 PM
To: '[email protected]'
Subject: RE: vm instances could not get ip address from dhcp agent
B.T.W, I just noticed that 'brctl show' command does not show any attached
interface on this test bed. Could this be the root cause?
# brctl show
bridge name bridge id STP enabled interfaces
virbr0 8000.525400046ff3 yes virbr0-nic
Also, I saw error message as '2013-12-17 21:33:47.586 32628 ERROR
nova.virt.libvirt.driver [-] Getting disk size of foreman: [Errno 2] No such
file or directory: '/var/lib/libvirt/images/foreman.img''.
Any clue?
Thanks,
YuLing
From: C, Yuling
Sent: Tuesday, December 17, 2013 6:47 PM
To: [email protected]<mailto:[email protected]>
Subject: vm instances could not get ip address from dhcp agent
Hi All,
I have an openstack installation environment of two nodes. One node contains
Controller, Network, Nova, Glance, Keystone, Client, and Cinder. The other node
is just a nova_compute node. Everything worked very well before. Suddenly we
hit the issue that vm instances could not get ip addresses from dhcp agent. I
looked into dhcp-agent.log, but didn't find much useful information. Just
wondering if someone could shed some light on this issue.
The behavior is that even the VMs created on node1, which contains Controller,
Network, Nova, Glance, keystone, Client and Cinder could not get the IP from
dhcp agent. When I use tcpdump -i br-ehter1, or tcpdump -i < physical Ethernet
port id>, I was able to see the DHCP requests, but no reply from dhcp agent.
Any idea?
Thanks,
YuLing
_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : [email protected]
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
