multiple: Alarm 'Host connection and power state' on esxi changed from Red to Green

[aig]

Hi

I have a cluster with 2 ESXi 6.5u2 hosts on vcsa 6.7

one of the hosts is popping an error every ~60 seconds, Alarm:

'Host connection and power state' on esxi_name changed from Red to Green

and

Host esxi_name in Datacenter is not responding

i ping constantly to the host and no packet is lost.

[NathanosBlightc]

Did you add your hosts with FQDN? If so, is your DNS server respondent correctly to the vCenter server?!

Also, give more information about the failover order and configuration of your switches and uplink connectivity... Is there any special setting on physical switch ports that your ESXi host uplinks are connected to them?

[sk591]

Questions regarding this:

- Since when did the issue start occurring? Were there any driver/firmware updates done?

- Is the host showing up as Not Responding in the vSphere Web Client every 60 seconds? If yes, is the vpxd log on the vCenter showing the following log entry?

[<YYYY-MM-DD>T<time> verbose 'App'] [VpxdIntHost] Missed 2 heartbeats for host <HostName>

- If you run a packet capture on the esxi and the vcsa, are you seeing any packet drops for UDP port 902 from host to VC?

[aig]

Hi

it started since I added the ESX to an existirng cluster. the other ESX doesnt have the same problem

no driver updates.

Yes for an instance it shows as not responding and then the message disappear until 1 minute after.

the vpxd-61.log doesnt show those entries.

correction: my ESXi version is ESXi 6.5 Update 3

[aig · Accepted solution]

issue resolved.

it was a blocked UDP 902 port from the ESX to VC

thanks