unable to ping only one instance out of many

asked 2013-07-24 08:02:17 -0500

nishtha
61 ●4 ●6 ●9

updated 2013-11-18 06:43:26 -0500

I am unable to ping only a single instance although the status of all the instances is "running". I have 3 compute nodes installed and 1 controller node. My first instance's host is my 1st compute node. However, the other instance on the same compute node is not pinging. Why is this happening so?

edit retag flag offensive close merge delete

Comments

Could you clarify whether your setup uses neutron or nova-network? Also you may not have a route to ping a VM depending on where you are trying to ping from, so please clarify from where you are pinging.

annegentle ( 2013-12-02 16:30:44 -0500 )edit

add a comment

1

answered 2014-08-30 04:35:06 -0500

Germy Lure
31 ●2 ●5 ●6

Try to ping and then execute arp command. If arp show that your VM has not learned peer's MAC, I think you need check switch and ports. If it is OK, you should check routes.

Anyway, I suggest use tcpdump to dump arp or icmp packets. Good luck!