ESX3.0.2 not pinging

[kevin79]

I have an ESX 3.0.2 host that I was trying to reconfigure HA on. During the reconfiguration, the host became unpingable. Is there a way to make it pingable without rebooting the host? Since I can't ping it, I can't vmotion the guests over and thus would have to reboot each guest, which I would rather not do if at all possible.

[bolsen]

What's unpingable? The SC or Kernel?

[kevin79]

The service console. What do you mean Kernel? Does that actually have an IP address? Isn't it just the SC? All of the VM's are still functioning perfectly.

[bolsen]

Kernel = vmotion, iscsi, etc.

From the console, can you ping out? How is the SC configured?

Shot in the dark here, but perhaps you could reconfig the network adapter duplex settings (or just leave them at auto) and let the nics reset?

[kevin79]

Here is the configuration: the SC and Kernel both share two physical NIC's that are shared with a vSwitch. The Kernel is using VLAN ID to route traffic, the SC is not. I can ping the Kernel IP but not the SC IP. I believe the NICs are already set at auto netogiate as the NIC's are gig nics.

On the console when I try to ping out it doesn't work. I can't try to ping by name because it things my DNS servers are unavailable. If I try to ping by IP address I get "ping: sendmsg: Operation not permitted".

Any idea's?

[tgradig]

I believe the command to restart the service console is: service mgmt-vmware restart

[TomHowarth]

You should have two diferent IP addresss, one for the SC and one for VMKernel.

Tom Howarth

VMware Communities User Moderator

[kevin79]

Will that cause my VM's to restart?

[kevin79]

I do have two different IP's but they are both part of the same vSwitch.

[tgradig]

It doesn't restart the guests. It will disconnect it out of VC until the service is restarted and automatically reconnect afterwards.

[kevin79]

Thanks, I tried the service mgmt-vmware restart and it still isn't pinging.

[tgradig]

Check your /etc/hosts Settings. If you configure a server for HA, you need to have the Host in lower case with the FQDN also. If might of mess with it.

Then try to restart the mgmt-vmware service again.

[kevin79]

My hosts file looks good. It has IP address FQDN (all lower case) hostname (lowercase).

[bolsen]

What I ment is if you open and then close the NIC duplex setting page it will renegotiate.

[TomHowarth]

can you post the output of esxcfg-vswitch -l, esxcfg-nics -l and esxcfg-vswif- l

Tom Howarth

VMware Communities User Moderator

[kevin79]

Ok, I have it pinging again. I restarted the iptables service. Now I have another issue, or rather, I think it is a continuation of my first issue. If I run the command iptables -L I get the following output:

Chain INPUT (policy ACCEPT)

target prot opt source destination

Chain FORWARD (policy ACCEPT)

target prot opt source destination

Chain OUTPUT (policy ACCEPT)

target prot opt source destination

I get a lot more if I run the command on my ESX box that has always been working. How do I reload the iptables rules back into the firewall?

[kevin79]

Anyone?

[PerryWhittle]

login to the SC either directly or via Putty,etc

run the following command to load firewall settings

esxcfg-firewall --load

(note the double - then load )

regards

Pez

#######################################

please award points if you find this helpful/useful

[kevin79]

Ok, I got it working. I ended up rebooting the server. Now, I can't configure HA on the server. I get "an error occurred during configuration of the ha agent on the host" when I try to enable HA. In the events tab in VC, I see ": cmd remove failed:" Any ideas?

[rlabhart]

I had the same problem with 2 of our ESX 3.0.1 servers after upgrading to VC 2.5.

I ran the following:

service iptables restart

esxcfg-firewall --load

That resolved our problem without rebooting.

For the HA, I went to the cluster name --> Edit Settings.

Uncheck Enable HA. Hit OK.

Now re-enable HA. Fixed.