Alarm Host connection state after installation of Veeam One

[mvirata]

Hello All,

We are currently running ESX 3.5 with four ESX hosts and a physical vCenter. I installed Veeam One 6 (free version) and we get the following false alarm from VMware that all four of our hosts are down. I originally saw the message when I first installed Veeam One 6. I realized it was a false alarm and disregarded it as a symptom of the Veeam One initial install.

This morning at 3 am, VMware complained again that all four hosts were down. Our hosts are not down , so the alarm is confirmed as a false positive. Here is the alarm:

[VMware VirtualCenter] Alarm - Host connection state

Target: VMserver.ourcompany.com
Old Status: Green
New Status: Red

Current value:
Host connection state - (State = Not responding)

Alarm: Host connection state
([Red State Is Equal To notResponding])

Description:
Alarm Host connection state on VMserver.ourcompany.com changed from Green to Red

I checked all four hosts and this alarm goes off right after a Task:Refresh Hardware Information on our ESX hosts. This happened on all four hosts simultaneously at 2:47am which caused these VMware alarms at 3:00am. Since it is all four at the same time, I am thinking it has something to with Veeam One, since we haven't done anything to the hosts, or seen this happen before.

I am disabling the Veeam One service for now. Any ideas? Could this be harmful to our vCenter in any way?

[Vitaly.Mzokov]

Mike,

Could you please contact our support team, as it is really hard to troubleshoot this issue via forums.
Your issue might require advanced troubleshooting and, also, our support team should at least take a closer look at the problem you have.

p.s. Post your support case number here, so I would be able to track your support case and assist my colleagues if necessary.

Thanks,
VM.

[Vitaliy S.]

Hi Mike,

I don't believe it can be harmful to your vCenter Server, as Task:Refresh Hardware Information is a widely used API call which just refreshes hardware sensors on the ESX(i) hosts.

Do you have any other events happening at the same time? Btw, what about Veeam ONE alarms? Do you see the same alarm triggered in Veeam ONE Monitor?

Anyway, as Vitaliy has suggested, I would appreciate if you could let our support team have a look at this issue live.

Thanks!

[mvirata]

Thank you for the responses,

I will turn the services back on and verify if the same alarm triggered in Veeam One Monitor. I don't recall that happanening, but I will confirm. I'll also give support a call if we continue to have issues.

[mvirata]

Actually, now that I have turned on the Veeam One services and looked through the history, there was a Veeam One alarm that triggered, and the conection was lost on all four ESX hosts simultaneosly.

In both cases (right after the Task:Refresh Hardware Information), the alarm was triggered on VMware and Veeam One. Obviously a false positive, but it did trigger. In the logs, about 1 minute later, connection status was restored.

I have forced a hardware refresh on all four of our hosts and no alarm has triggered. I will keep an eye out for it and then contact support if I continue to see issues.

[cgrebing]

Hi,

I have the same problem at a customer. After he removed a SAN Volume from all of his vCenter hosts he gets the message "Refresh Hardware Information" every 10 seconds for all of his hosts.
before the removal of the SAN volume everythig was OK.Do I need to reset/refresh something to remove this messages.

Thanks

[Vitaliy S.]

Hi Christian,

Veeam ONE does not trigger these events with this interval, the default interval is either 12 or 24 hours (need to double check). I'm not sure there is anything you can reset to stop it.

I would suggest either waiting for the new release of Veeam ONE 6.5 (should happen very soon), which uses CIM over XML protocol to retrieve host hardware sensors, or contact our support team and investigate your current issue via debug log files.

Thanks!