SMT2200RM2U PowerChute UPS Communications Lost

[Anonymous user]

Hello,

I have two SMT2200RM2U UPSs each setup and connected through the RJ45-Serial connector to a ESXi host server. I had been playing around trying to setup the PowerChute software on UPS2, which is connected to ESXi2. In the first iteration of the install, I was able to get the PowerChute agent running using the documentation provided on the CD, and could login and view UPS statistics. I was trying to configure advanced settings when the UPS and server shutdown for some reason. I could power everything back on, but Powerchute now only shows 'UPS Communication Lost' for the system status. I uninstalled and re-installed the PBEAgent several times with different options, tried using the serial-RJ45 from the other UPS. I can't get the original Powerchute setup with green UI and all the UPS stats. I can re-install and use the simple signaling mode, with the blue GUI, but that tells me nothing about the UPS.

I did a fresh setup for UPS1 and ESXi1, and can see the UPS information, and have only applied the initialization configuration.

Where the heck did UPS2/ESXi2 go wrong?

What should I try to get this working or troubleshoot further? I've reset UPS2 to factory settings, and will try deleting the VMware appliance image and creating a new one on ESXi2.

Thanks.

[BillP]

You should verify that the serial port is set for Connect at power on and that no other VM is attempting to access the serial port.

Something else that could cause the issue is if you have the hosts set up in a cluster and you are accessing the host via vCenter Server. When the vMA was powered down it associated itself with the other host so there is no physical connection between the host the vMA is running on and the UPS.

If none of the above helps I would uninstall the vMA and reinstall.

[Anonymous user]

About 1 hour after I had setup Powerchute with vMA1 and UPS1, the event log lists Communication Lost. After rebooting the vMA, the powerchute event log shows "Communication Not Established" after the monitoring service starts. The only other VMs running are 2x Windows 2012 domain controllers, one for each ESXi server, which do not have the serial ports added to the configuration.

I had deleted and created vMA2 multiple times, and re-installed Powerchute multiple times as well. I'm probably missing something, but as long as the servers can shutdown gracefully on power loss, and power back on with line power back up, I don't care if I get UPS stats or not. Using the simple signaling setup for Powerchute seems to only tell me that the UPS is online (but it's more than "communication lost"). Would that be all I need to have my VMs shutdown gracefully and power back on when the UPSs are on line? 

[Anonymous user · Accepted solution]

I found an old thread about a similar issue, with some of your responses, Bill.

http://forums.apc.com/spaces/7/ups-management-devices-powerchute-software/forums/general/3206/ups-co...

Specifically the one recommending HP configuration, and this link:

http://h20565.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c01139344〈=en&cc=us

I have newer gen HP dl360s, but it appears the same iLO settings have carried forward. I've disabled only the Serial CLI Status for iLO, and left the RBSU setting alone, rebooted one of the servers, and PowerChute seems to be working as originally expected. I'll be doing the same to the other server, and we'll see what happens over to the next day.

[Anonymous user]

It appears there have been no incidents since disabling Serial CLI Status on my servers, and the UPSs are communicating properly through the serial/RJ45 link.