Ask Your Question
0

After rebooting server not able to access the floating IP of the instance?

asked 2017-04-07 07:40:41 -0500

Harsha gravatar image

I'm running Openstack Newton, I configured a/c to Openstack documentation. Everything was working fine, I assigned floating IP to instances after that I'm able to ping and ssh into those instances. But after restarting the server I'm not able to ping into those instances( Floating IP was same ). I tried assigning other floating IP but after assigning also, I'm not able to communicate with those instances. External network:192.168.1.0/24 Internal network:10.10.11.0/24

Please help me resolve this issue!

edit retag flag offensive close merge delete

Comments

Did your instance retrain the internal network ip after reboot?

kuro gravatar imagekuro ( 2017-04-07 08:42:18 -0500 )edit

What is your configuration? Which server did you restart? A few simple checks: Are all services and network agents running? Any errors in the L3 agent log(s)?

Bernd Bausch gravatar imageBernd Bausch ( 2017-04-07 15:33:07 -0500 )edit

@kuro..yes both internal and external ip remains same but not able to ping into that.. But I can ping the router IP

Harsha gravatar imageHarsha ( 2017-04-08 02:24:06 -0500 )edit
add a comment

1 answer

Sort by ยป oldest newest most voted
0

answered 2017-04-07 08:29:55 -0500

bardhi232 gravatar image

Can you try to launch a cirros instance and log into it from the web console and see if you can communicate to the internet. For example ping 8.8.8.8 or your physical router? If that doesnt work then i would say that there is something wrong with your metadata server that is not giving the ip again after you reboot your instance. I had a similar issue myself and when i was checking the config files again i found out that i made a mistake there. After correcting the mistake everything was normal again.

edit flag offensive delete link more

Comments

Small correction: The IP comes from DHCP, not the metadata service.

Bernd Bausch gravatar imageBernd Bausch ( 2017-04-07 16:24:10 -0500 )edit

Yeah..I tried with the cirros instance itself was working fine am able to ssh into that instance before restart, but after restarting servers..I'm not able to ping into that instance.

Harsha gravatar imageHarsha ( 2017-04-08 02:35:18 -0500 )edit

What changes you made in config files?

Harsha gravatar imageHarsha ( 2017-04-10 06:17:37 -0500 )edit

I dont remember correctly now what changes i made to my config files because i did that over a year ago but as far as i can remember in my case i forgot to uncomment some config lines that had to do with the metadata service. if you could post in here your neutron logs so i could take a look

bardhi232 gravatar imagebardhi232 ( 2017-04-13 04:43:23 -0500 )edit

@Bernd Bausch off course IP comes from the DHCP service, but if the metadata server is not working your vm wont get the ip address. DCHP allocates the IP and then metadata writes it to the VM's network config files.

bardhi232 gravatar imagebardhi232 ( 2017-04-13 04:45:18 -0500 )edit
see more comments

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

subscribe to rss feed

Stats

Asked: 2017-04-07 07:40:41 -0500

Seen: 706 times

Last updated: Apr 07 '17

Related questions

openstack newton: linux bridge agent out of sync with plugin

Openstack Networking towards the internet

Could not able start neutron-lbaasv2-agent

Nova-Network on Hetzner. instance cant pass bridge.

HEAT syntax for multiple fixed_ips as a parameter

how to configure multiple external network in singlenode openstack juno +local

external network is not reachable from the router

Setting Firewall on OpenStack nodes

how to use CONFIG_NEUTRON_OVS_TUNNEL_IF in answer file?

Assign external IP to instance in Openstack Mirantis 6.1