When it comes to managing critical Horizon management servers such as Connection Servers, Unified Access Gateways, AV Managers, etc… It is imperative that you properly monitor their services for availability and that none of them are in a failed state, as well as make sure there are no connectivity issues on internal networks that the Horizon Client traffic traverses. By leveraging the Telegraf agent HTTP Health Check and PING Check functionality with Aria Operations, we can provide additional insight into the Health, performance, and availability of these services.
This video is meant to provide high-level overview of:
Why you should leverage Telegraf with the Management Pack for Horizon
How to install and configure the Telegraf agents.
How to validate your settings after deployment.
Review of the dashboards that leverage Telegraf metrics.
As of 1.27.2022, the Management Pack for Horizon 2.0 is now released and available for on-prem or cloud customer deployments! This is an exciting release as it brings a number of groundbreaking features to further enhance our customer’s ability to successfully monitor, maintain, and optimize their Horizon deployments.
This list of enhancements include:
Primary use cases for Unified Access Gateway (UAG) monitoring include:
Primary use cases for Horizon Connection Server monitoring include:
Introducing Horizon User Capacity and Right-sizing metrics:
Introducing Horizon VDI Pool Capacity and Configuration optimization metrics:
Content Release Notes: This content release includes a number of Dashboards, Views, and Super Metrics designed to take advantage of the User Interface enhancements in the vROPs 8.x platform, providing richer and more visually impacting insight into the overall health, performance, and consumption of your Horizon View deployment.
Compatibility: All content was created on the latest vROP 8.0 platform, so it has only been tested to work with this specific version. The Super Metrics are most likely to work with older versions such as 7.5, but the Views specifically have configurations options that were not available prior to vROPs 8.0.
With vROPs 8.0, only Horizon Adapter 6.7.x is fully supported, so make sure that your Horizon View environment is compatible with the vROPs for Horizon 6.7.x adapter.
Conditions: All content provided is on an as-is basis, and should be validated and tested before use in production environments.
*It is recommended that you import and configure the provided Views and Super Metrics before importing the dashboards. Examples of how to do so are provided in previous posts.
As the number of user sites or locations increases, having good visibility into the overall quality of connectivity of those sites to your Horizon View data center(s) becomes increasingly important. Having worked with many customers on troubleshooting connectivity between such locations, it has become clear that monitoring only at the physical network layer is not sufficient to properly diagnose user connectivity issues impacting the display protocol. In fact, in most cases, it’s the configuration of the physical layer that causes the issue(s), and the device(s) in question do not have the ability to diagnose or detect their impact to the display protocol.
Alas, all hope is not lost! With vROPs for Horizon, we have the ability to monitor the key metrics related to protocol performance and can alert when those metrics have reached critical thresholds. We can also leverage a handy custom grouping feature to organize the remotely connected sessions into defined sites or locations, based on information available in the user’s session data. We can then leverage Super Metrics to calculate the overall health of the group of connected sessions from that site, and then display and alert when the health has dropped below our SLA thresholds.
Sound like something you want to take on? Then read on!
We need a strategy to capture and organize how we determine which connections are from which site. The most common approach is to leverage the known internal subnet IP range for that site, and set that as the primary filter for the custom group. However, additional metrics supplied by the user’s session data can be leveraged as well.
In this example, I will create a new Site labeled “External Users”, under the “Location” group, and filter on all users who have an internal subnet IP containing “192.”.
Navigate to Environment tab -> Location -> +Notice that both session types Desktop and Application are added separately. While not required it will impact which sessions are captured, so make sure to add the required session types.
Note: It may take some time before the group starts to populate the sessions, and as noted in the image above, population updates are every 20 minutes.
You will also want to create at least one more additional site for comparison purposes. In the example below, I have created an “Internal Users” site that filters on 10.x networks to separate the session traffic.
Second is just after first… as long as we’re counting up! 👆
Once your group has started to populate with sessions, it’s time to create meaningful Super Metrics on the group that will give us insight into how well that site is connected.
By default, vROPs 7.5+ will create population TotalCount and Health Criticality Total Count metrics. These will be useful in leveraging out-of-box KPI based alerts that track how well users are connected to the Horizon Environment and let us know the distribution of good to bad connected sessions from that site.
Out of box Health Criticality counts on the custom group.
Leveraging Super Metrics, it’s time to create additional site tracking metrics that include, but are not limited to, average site Latency, Packet Loss, Jitter, Frame Rate, and Transmitted Bytes.
Super Metric name is “Total Site VDI Avg Packet Loss”.Use the Avg function on the VDI Session Blast Packet Loss Uplink metric.Assign the Super Metric to the “Location” object.Select the active policy that will allow the Super Metric to collect.Verify that the “Location” Object Type has been assigned to the Super Metric.Verify that the active policy is associated with the Super Metric.
Thirdly Rinse and Repeat… ♻️
Continue to create the Super Metrics for the additional session metrics that you want to collect. As in our example below, you will need to do this for both the VDI and Application Sessions individually to capture the metrics from both types of sessions.
Completed Super Metrics list.
Now… what to do with all this stuff? 🤔
Now that we have meaningful Super Metrics on our custom group, it’s time to display that data in a way that it makes it easier to diagnose specific site connectivity issues, and compare that site against other sites that are connecting to our Horizon environment.
To do so, I’ve created a custom example Dashboard that allows you to select from the list of defined sites, and then display the relevant Alerts and Super Metrics that we defined on those sites. While you can choose to display the data in any manner you see fit, I’ve chosen to use a combination of Scoreboard, Distribution, and Heatmap widgets.
I’ve also chosen to label the widgets using an ordered numbering system with directions so that other users leveraging the dashboard will know how it’s meant to be consumed.
1) Select the site. 2a) Review the site alerts. 2b) Review the break down of criticality of alerts. 3a) Review Super Metrics that were created to analyze the site connectivity.3b) and 3c) Notice that clicking on a Distribution View will show you which session(s) fall under the defined performance “buckets”. 4a) Select a problem session. 4b) Review scoreboard of relevant KPIs impacting that session. KPI’s are colored to show level of impact to the user’s session. 4c) Review session related objects looking for other impacted objects with alerts.
Continue your analysis of the Application Sessions if relevant for your site. The workflow will be the same as the above VDI Session analysis.
5a) Review the App Session Super Metrics for anomalies. 5b) and 5c) Clicking the Distributions will show you the session(s) the fall into each performance “bucket”.6a) Select the session you want to review. 6b) Review the Scoreboard metrics for offending KPIs. 6c) Review the session relationship chart and look for additional objects with Alerts that may be impacting the user session.
Next steps… Download and enjoy! 🏁
Now that we’ve discussed how to create the content and the high-level strategy for performing the site connectivity analysis, it’s time to download and import the content . The zip file provided in this blog contains all of the content that is required to create the custom dashboard, except for the step on creating the custom group locations. That step will still be required for you to define what “Sites” you want to monitor for connectivity performance.
Contents of the zip include a Dashboard_Horizon Site Location Analysis.zip, Views_Horizon Site Location Analysis.zip, and Supermetric_Horizon Site Analysis.json. All of which need to be imported and at least one Location group defined before the Dashboard will function properly.
Enjoy! Please provide thoughts or feedback on how to make the dashboard and content better.
**This content is now included out-of-box in the vROPs for Horizon Adapter installation .pak file and dashboards download file available at: downloads.vmware.com
The latest vROPs for Horizon custom content includes an updated End User Experience and Help Desk dashboard, with Blast Session Metrics and Alerts, Horizon Client External IP address, as well as new Navigate to “Horizon Root Cause Analysis Dashboard” functionality.
End User Experience Dashboard:
To use the root cause analysis dashboard, select any object from a heat map or relationship chart on the End User Experience dashboard, then select the Navigate to “Horizon Root Cause Analysis” option to begin troubleshooting.
New Blast Session Metrics and Root Cause functionality.
Horizon Root Cause Analysis:
The Horizon Root Cause Analysis Dashboard allows you to quickly switch between correlated environmental objects, and see the best practice metrics and alerts on the selected item. Colors will change to indicate any problem metric, or simply provide visibility into metrics of “interest”.
Select the object you want additional information on, and the best practice metrics will be viewed in the Scoreboard below, as well as any alerts on the selected object. You can also click a Scoreboard metric to add it to the Selected Metric Chart widget, as well as switch objects, and continue to build out the metric chart with various environmental objects to put together a cross-environment “picture” for analysis.
New Root Cause Analysis Dashboard
Horizon Help Desk Dashboard:
The latest Help Desk Dashboard includes Blast Session metrics and alerts, as well as an updated Horizon Client Details widget to include the External/Internal IP of the users client connecting to the environment. The navigate to “Horizon Root Cause Analysis” Dashboard functionality is available on the user session and related objects widgets.
Updated Metric Config to drive the latest Scoreboard content:
To leverage the latest Scoreboard content to see Blast Session metrics, a new metric config labeled “Session_Desktop_stats_v2.xml” is included in the latest content pack. All three dashboards included in the content pack leverage this metric config file to highlight and display key offending metrics for ease of troubleshooting. To add the new metric config, navigate to Content -> Manage Metric Config, click add new and use the exact name of “Session_Desktop_stats_v2.xml”. Then copy and paste the content of the included metric config, and click Save.
If the dashboard fails to load the content correctly, click the edit icon on the Scoreboard widgets and make sure they are pointing to the correct file name created above.
New Help Desk Dashboard with Blast Metrics and External/Internal IP visibility.
To reduce the number of clicks required for administration and add additional visibility to end-user environmental issues, a new Host Metrics scoreboard has also been added to the Help Desk dashboard. The scoreboard will show best practice metrics in green, and any metrics of concern or interest will change color (yellow, orange, red) depending on the latest reported value.
See correlated best practice host metrics without additional navigation.
New Session Alert:
In addition to the new dashboard content, Blast and PCoIP session metrics have been added to a new single alert named “Session Protocol Performance Degraded”. This replaces the existing PCoIP Performance Degraded degraded alert. This will need to be imported under the Content -> Alerts menu.
New How To Widgets available on each dashboard:
Each dashboard now contains a “How to use…” widget that provides a brief description how to quickly operationalize the dashboards. The widgets leverage a URL link back to this website to display the latest instructions on how to use each dashboard. If the widget doesn’t display the text correctly, make sure that the URL referenced in the edit widget screen is resolvable.
**Note that most of this content will be available out-of-box in the vROPs for Horizon 6.4 release due Dec 2016. Additional custom content will only be available from this website.
Click to Download latest vROPs for Horizon Content 8.3 **download removed
