VMware Cloud Community
MobiusJB
Contributor
Contributor
Jump to solution

Disconnecting / Not Responding ESX server from Virtual Center

Hi, I am running Virtual Center 2 and I have two datacenters (each representing a different physical segment of my network).

Virtual Center is connecting to one of the ESX servers over the LAN (internal IP address) and connecting to the other ESX server over the via a public IP address. The local datacenter over the LAN is running ESX 3.x and the remote datacenter is running 2.5.x.

The ESX server that is remote (public IP, v.2.5.x) is having issues staying connected to my Virtual Center. For instance, I add the host, after adding the host I can see the server and virtual machines, shortly after, I'd say within 2-5 minutes, the server goes into a "Not Responding" mode. The only way I've really attempted to troubleshoot this was to disconnect and reconnect the server, which again, brings the host up and available for 2-5 minutes and then the same issue occurs.

I have used ping plotter to track packets to the source where my remote host is and there is very little packet loss, it is not consistent with the host server disconnecting, so I've ruled out that it is a problem with connecting over the public WAN. Is this a problem with ESX 2.5.x connecting to VC 2 maybe?

I am new to the forums and have not dug thru previous posts, but would someone please take a crack at this or point me to a resource / guide to resolve this issue?

Also, FYI, the remote ESX server runs just fine, I can use my Virtual Infrastructure Client to connect to it and monitor / edit etc without issue.

0 Kudos
1 Solution

Accepted Solutions
JonT
Enthusiast
Enthusiast
Jump to solution

I have read elsewhere that you may want to make sure there is no firewall intervention between your ESX 2 hosts and the VC server. Since you are going from public space to internal your firewall may be blocking some of the traffic.

View solution in original post

0 Kudos
4 Replies
akutz
Hot Shot
Hot Shot
Jump to solution

I was experiencing the exact same thing, and the problem was that the /var/core directory had filled up the 2 G dedicated to the /var filesystem (by VMware's design guidelines). Our solution was to delete old core files from /var/core and write a script that monitors the space available on the disk.

When the /var file system filled up, the vpxa daemon could not write its log and crashed. To fix this we input:

/etc/init.d/vmware-vpxa start

VC could now communicate with the ESX server.

0 Kudos
rminick
Contributor
Contributor
Jump to solution

MobiusJB - I am having a similar issue except all 7 ESX hosts are in 2 data centers, all 3.01 and on the internal LAN. The host that keeps showing up as disconnected is on the same Cisco switch as VC. It's reconnects itself after 4-5 minutes and does it all again about every half hour.

Richard J Minick, VCP
0 Kudos
MobiusJB
Contributor
Contributor
Jump to solution

Sorry for the delay in responding... that directory is fine on space. This has not been a pressing issue for me so I have not been looking too intently, but I did also read that you must have ports 80,443 and 902 opened both ways if your VC is on the other side of a FW from your ESX server, in case there are 2 FW's (geographical location) seperating VC from ESX. I have not had time to test these changes but I will in the next week and I'll post back if that works

0 Kudos
JonT
Enthusiast
Enthusiast
Jump to solution

I have read elsewhere that you may want to make sure there is no firewall intervention between your ESX 2 hosts and the VC server. Since you are going from public space to internal your firewall may be blocking some of the traffic.

0 Kudos