cancel
Showing results for 
Search instead for 
Did you mean: 

Historical Soft Endpoint WMI Metrics

Charles_Zhang
05 Base Camper

Historical Soft Endpoint WMI Metrics

IR Team,

Through multiple knowledge sharing sessions with our Ops teams (whom will be the primary users of Prognosis for Skype Monitoring), one of the top, if not the top asks was to make historical Soft Endpoint WMI metrics available, like how it is today with backend servers. Currently, Soft Endpoint WMI metrics are only available ad-hoc and in real-time (actually, the URL to initiate an endpoint WMI query is only available from a post-call session detail, which kind of defeats the purposes of using WMI to query an endpoint as a call is in-flight; the URL to initiate an endpoint WMI query is not available when viewing a session detail via SDN data but that's another conversation).

I can already predict push back on having Prognosis collect WMI metrics on endpoints due to the potential for exponential storage requirements (from 200K+ endpoints in our environment) but having this historical view from an endpoint perspective is critical from our Ops support perspective. The majority of tickets that our Ops team receive are post-call scenarios so having a historical view into an end user's endpoint can greatly assist our Ops team in determining root cause for poor quality Skype calls.

I wanted to throw this out there and see:
1. If there have been a similar requests for this from other clients of IR that is using Prognosis today for monitoring of their UC/Skype environment.
2. If Product Team have brainstormed about this? Just thinking aloud - perhaps Prognosis can selectively collect endpoint WMI metrics? Leverage SDN data, kick off the collection of endpoint WMI if we see that an active call is experiencing certain levels (thresholds can be definable) of degradation.

Thoughts?

Thanks,
-cz
Tags (1)
7 REPLIES 7

Re: Historical Soft Endpoint WMI Metrics

Hi Charles,

I had a look at this request and discussed briefly. As you indicated, 200K+ remote WMI collection points may not scale from a single or few monitoring nodes. I'll keep the conversation going with the product and R&D team on what limitations we have for this.

However, I do like the idea of selective monitoring for in-call endpoints having some sort of issues. Let's see what others think about it and if it is feasible to do so. I'll keep you posted if I have some news on this.

Regards,
Shoaib.

If my reply answered your question please click on the 'Accept as Solution' button to help others find the answer.
Thanks,
Shoaib
Charles_Zhang
05 Base Camper

Re: Historical Soft Endpoint WMI Metrics

Hi Shoaib,

Any other ideas that may have been thrown around internally?

I was also thinking why doesn't Prognosis show this specific client event metric that's captured in the QoE data?

CPU insufficient time: Percentage of time there was insufficient CPU for processing current modalities and applications, causing audio distortion

This metric gives some insight from what endpoint WMI would show, historically.

Thanks,
-cz

Re: Historical Soft Endpoint WMI Metrics

Charles,

internal discussion was also around scaling it, as this would not scale for 500+ endpoints as-is, let alone 1000s of endpoints. remote powershell is another discussed idea but nothing done practically on that front at the moment.

If my reply answered your question please click on the 'Accept as Solution' button to help others find the answer.
Thanks,
Shoaib
Andreas_B
05 Base Camper

Re: Historical Soft Endpoint WMI Metrics

[...]
However, I do like the idea of selective monitoring for in-call endpoints having some sort of issues. Let's see what others think about it and if it is feasible to do so. I'll keep you posted if I have some news on this.
[...]

Hi Shoaib,

 

I wsa just thinking about this kind of information.

We currently have some audio issues with one certain computer and I'd like to know what would be the best approach to nail down the root cause.

So my question: is it possible to grab WMI data from one endpoint (Windows 7, x64 client), reproduce the issue and see what could interfere?

 

Especially because some microphones seem to have a high "glitch rate", which according to Microsoft, might be caused by either driver or software like Virus scanner etc.

My aim is to find out with help of Prognosis what causes this kind of issue.

 

regards

Re: Historical Soft Endpoint WMI Metrics

Hi @Charles_Zhang and @Andreas_B,

 

we now have an extended solution, Remote Server Monitoring, that can collect remote data from soft phone hosts (windows devices) using either WMI or PowerShell. Using this solution, we can store historical data and look at what is causing bad experience from window's performance prospective. It may not scale for 1000s of devices, but for repeat offenders, they can be added to monitoring list to collect data historically.


If my reply answered your question please click on the 'Accept as Solution' button to help others find the answer.
Thanks,
Shoaib
Andreas_B
05 Base Camper

Re: Historical Soft Endpoint WMI Metrics

@Shoaib_Dilawar thanks for the response. Is there anything inside the online help which describes how to implement this?

 

kind regards

Re: Historical Soft Endpoint WMI Metrics

Hey @Andreas_B

 

It is an extended solution, so you'd need to get in touch with your account manager to discuss how to get it.


If my reply answered your question please click on the 'Accept as Solution' button to help others find the answer.
Thanks,
Shoaib