VMware Cloud Community
MrBigFeathas
Enthusiast
Enthusiast

vCenter server cannot ping ESXi 4 host, every other machine on network can ping it

I am having a very strange issue. I have had a 3 node vSphere cluster for about a year and a half. It has always worked fine. There have not been any changes to the cluster for months. Today when i checked it one of the three hosts is disconnected. I cannot ping the management IP or the vMotion IP for this server from the vCenter server. I can ping those IPs from any other server or client I have tried within my organization. Has anyone ever seen this? I am at a loss.

I can connect to the server with the vSphere client directly from other machines and there is nothing unusual in the logs.

Any help would be greatly appreciated.

Reply
0 Kudos
3 Replies
idle-jam
Immortal
Immortal

is your vcenter a physical or virtual machine? if it's physical and you have the luxury to do a back to back cable connection i would use a laptop try if there is anything wrong with the network card. i would also set my laptop to be those of the ESXi.

if that works, then you might need to look at the switch level or firewall level that maybe blocking or causing the problem.

Reply
0 Kudos
JustinBailey
Contributor
Contributor

I am having the exact same problem. If I try to ping from the vCenter server, the ESXi host will not respond; yet if I ping from my laptop or any other host (even hosts on the same subnet as the vCenter), the ESXi host will reply. This has happened at several locations where the hosts were responding and just recently stopped responding to the vCenter server. A traceroute from the vCenter server shows that the packet drops once crosses the WAN and hits the serial interface of the router; traceroute from a server on the same subnet as the vCenter server crosses the WAN and hits the router, then goes layer 2 to the host without fail. These hosts are trunked on the switch and all the necessary vlans are allowed. We know the hosts are on the network and they can be accessed by ssh from any other machine. We did resolve this at 1 location by changing the host ip to a different ip (I'm not convinced that the problem won't reoccur). We have checked both failing locations for duplicate ip addresses and this is not the case.

I'm hoping that someone else has found an answer to this issue.

Reply
0 Kudos
sflanders
Commander
Commander

IP or MAC address conflicts can cause these symptoms. I had also seen a similar issue with a VMkernel interface connected to a NAS datastore. The solution for me was to create a new interface with the same IP information, but a new MAC address and the issue never came back (http://sflanders.net/cloud/esx-nfs-datastores). Hope this helps.

Hope this helps! === If you find this information useful, please award points for "correct" or "helpful". ===
Reply
0 Kudos