Windows Server Core & hardware monitoring

[AT1000]

Hi,

I'm running a single physical server (HPE ProLiant ML350 G9) running Windows Hyper-V Server 2016 (essentially the same as Windows Server 2016 Core with the HyperV role).  That in turn hosts a couple of Windows virtual servers (the usual - AD, file, print etc.).  I'm trying to find the best way to automate the monitoring of the underlying HPE hardware (e.g. looking out for failed disks, PSUs etc.).

I've got remote agents monitoring the virtual servers, and if the underlying hypervisor was VMWare running the HPE custom image and hence HP VIBs, I'd use standard CIM\WBEM\PowerCLI queries to interograte the VMWare host via the virtual machine.

With my HyperV setup I can install my remote agent directly into the host OS, but then need a way to query the hardware (i.e. some sort of HP tool\integration).  I've looked at various things like the Agentless Management Service (seems to be too big for what I need and relies on SNMP traps and\or email alerting) and Insight Management WBEM Providers (doesn't seem to want to install, so maybe isn't compatiable with a Core MS Server OS?).

I've got more experience with Dell server hardware and know that their OMSA tool would do the job beautifully - it can be installed on to the host OS (including Server Core), and will capture alerts\warnings and write them to the Windows Event Logs (which my remote agent will then pickup).  I'm hoping there's a similar HPE software suite available that I don't know about!

Any thoughts would be valued.
Thanks in advanced!

[NJK-Work]

I have never personally dealt with the ML models myself, I typically deal with the DL brand.  However, I am assuming the iLO in the ML model has the AlertMail feature?  If not, stop reading here :)

Since you have only one server, the AlertMail might be a good option.  It will send you, and others, an email whenever a hardware issue is detected by iLO.  You cannot do any filtering - its pretty much all or nothing.  For lots of servers and control over what get alerted on, you would use HPE SIM or OneView.  Personally, I use HPE SIM (hence this forum) and then use the Agentless Monitoring.  But since you only have one server, I am guessing you dont have or want/need SIM...and so the agentless monitor is not going to help.  So, try the AlertMail function of the iLO - that should get you some basic alerting on the hardware problems.

NK

[NJK-Work]

Here is a sample AlertMail email for a server that had a CPU fan failure:

NK

[AT1000]

Hi NK,

Cheers for the feedback and thoughts.  The challenge I've got with AlertMail is that I'm using Office365 for emails, and hence need to authenticate with a mailbox to allow the sending of emails (i.e. I've not got a local mail server which I can use as a relay with no authentication).  From looking at the AlertMail functionality (I tested it at another site with a local mail server), it doesn't seem to support authentication.

I could get round the above with a quick and cheerful IIS hosted SMTP relay on the VM (which would accept unauthenticated mail, and itself then authenticate with Office365), but I was hoping for a lightweight\minimal config solution.

I'm not against the idea of using SIM or another HP tool, but my focus is on keeping it lightweight if possible!