Skip to content

Commit

Permalink
Merge pull request #10452 from MicrosoftDocs/main
Browse files Browse the repository at this point in the history
main to live publish, 12/6/24, 3:30 PM PT
  • Loading branch information
rjagiewich authored Dec 6, 2024
2 parents 621cc41 + 4b0d73c commit de87bbc
Show file tree
Hide file tree
Showing 5 changed files with 130 additions and 25 deletions.
8 changes: 8 additions & 0 deletions windows/deployment/do/mcc-ent-deploy-to-linux.md
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,10 @@ Before deploying Connected Cache to a Linux host machine, ensure that the host m
1. Within the Azure portal, navigate to the **Provisioning** tab of your cache node and copy the provisioning command.
1. Download the provisioning package using the option at the top of the Cache Node Configuration page and extract the package onto the host machine.
1. Open a command line window *as administrator* on the host machine, then change directory to the extracted provisioning package.

>[!Note]
>* If you are deploying your cache node to a Linux host machine that uses a TLS-inspecting proxy (e.g. ZScaler), ensure that you've [configured the proxy settings](mcc-ent-create-resource-and-cache.md#proxy-settings) for your cache node, then place the proxy certificate file (.pem) in the extracted provisioning package directory and add `proxyTlsCertificatePath="/path/to/pem/file"` to the provisioning command.
1. Set access permissions to allow the `provisionmcc.sh` script within the provisioning package directory to execute.
1. Run the provisioning command on the host machine.

Expand All @@ -42,6 +46,10 @@ To deploy a cache node programmatically, you'll need to use Azure CLI to get the
1. Save the resulting output. These values will be passed as parameters within the provisioning command.
1. Download and extract the [Connected Cache provisioning package for Linux](https://aka.ms/MCC-Ent-InstallScript-Linux) to your host machine.
1. Open a command line window *as administrator* on the host machine, then change directory to the extracted provisioning package.

>[!Note]
>* If you are deploying your cache node to a host machine that uses a TLS-inspecting proxy (e.g. ZScaler), ensure that you've [configured the proxy settings](mcc-ent-create-resource-and-cache.md#proxy-settings) for your cache node, then place the proxy certificate file (.pem) in the extracted provisioning package directory and add `proxyTlsCertificatePath="/path/to/pem/file"` to the provisioning command.
1. Set access permissions to allow the `provisionmcc.sh` script within the provisioning package directory to execute.
1. Replace the values in the following provisioning command before running it on the host machine.

Expand Down
44 changes: 33 additions & 11 deletions windows/deployment/do/mcc-ent-deploy-to-windows.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@ appliesto:

This article describes how to deploy Microsoft Connected Cache for Enterprise and Education caching software to a Windows host machine.

Deploying Connected Cache to a Windows host machine requires designating a [Group Managed Service Account (gMSA)](/windows-server/security/group-managed-service-accounts/getting-started-with-group-managed-service-accounts) or a [Local User Account](https://support.microsoft.com/windows/create-a-local-user-or-administrator-account-in-windows-20de74e0-ac7f-3502-a866-32915af2a34d) as the Connected Cache runtime account. This prevents tampering with the Connected Cache container and the cached content on the host machine.
Deploying Connected Cache to a Windows host machine requires designating a [Group Managed Service Account (gMSA)](/windows-server/security/group-managed-service-accounts/getting-started-with-group-managed-service-accounts) or a [local user account](https://support.microsoft.com/topic/20de74e0-ac7f-3502-a866-32915af2a34d) as the Connected Cache runtime account. This prevents tampering with the Connected Cache container and the cached content on the host machine.

Before deploying Connected Cache to a Windows host machine, ensure that the host machine meets all [requirements](mcc-ent-prerequisites.md), and that you have [created and configured your Connected Cache Azure resource](mcc-ent-create-resource-and-cache.md).

Expand All @@ -26,14 +26,25 @@ Before deploying Connected Cache to a Windows host machine, ensure that the host
# [Azure portal](#tab/portal)

1. Within the Azure portal, navigate to the **Provisioning** tab of your cache node and copy the provisioning command.
1. Download the provisioning package using the option at the top of the Cache Node Configuration page and extract the package onto the host machine. **Note**: The installer should be in a folder that isn't synced to OneDrive, as this will interfere with the installation process.
1. Download the provisioning package using the option at the top of the Cache Node Configuration page and extract the archive onto the host machine.

>[!Note]
>* The provisioning package should be extracted to a directory that isn't synced to OneDrive, as the sychronization process will interfere with the installation. It is recommended to extract the provisioning package to the root directory of the host machine (e.g. C:\mccInstaller)
1. Open a PowerShell window *as administrator* on the host machine, then change directory to the extracted provisioning package.

>[!Note]
>* If you are deploying your cache node to a Windows host machine that uses a TLS-inspecting proxy (e.g. ZScaler), ensure that you've [configured the proxy settings](mcc-ent-create-resource-and-cache.md#proxy-settings) for your cache node, then place the proxy certificate file (.pem) in the extracted provisioning package directory and add `-proxyTlsCertificatePemFileName "mycert.pem"` to the provisioning command.
1. Set the Execution Policy to *Unrestricted* to allow the provisioning scripts to run.
1. Create a `$User` environment variable containing the username of the account you intend to designate as the Connected Cache runtime account.
1. Create a `$User` PowerShell variable containing the username of the account you intend to designate as the Connected Cache runtime account.

For gMSAs, the `$User` PowerShell variable should be formatted as `"Domain\Username$"`. For local user accounts, `$User` PowerShell variable should be formatted as `"LocalMachineName\Username"`.

For gMSAs, the value should be formatted as `"Domain\Username$"`. For Local User accounts, `$User` should be formatted as `"LocalMachineName\Username"`.
If you're using a local user account as the Connected Cache runtime account, you'll also need to create a [PSCredential Object](/dotnet/api/system.management.automation.pscredential) named `$myLocalAccountCredential`.

If you're using a Local User account as the Connected Cache runtime account, you'll also need to create a [PSCredential Object](/dotnet/api/system.management.automation.pscredential) named `$myLocalAccountCredential`. **Note**: You'll need to apply a local security policy to permit the Local User account to `Log on as a batch job`.
>[!Note]
>* You'll need to apply a local security policy to permit the local user account to `Log on as a batch job`.
1. Run the provisioning command on the host machine.

Expand All @@ -48,22 +59,33 @@ To deploy a cache node programmatically, you'll need to use Azure CLI to get the
```

1. Save the resulting output. These values will be passed as parameters within the provisioning command.
1. Download and extract the [Connected Cache provisioning package for Windows](https://aka.ms/MCC-Ent-InstallScript-WSL) to your host machine. **Note**: The installer should be in a folder that isn't synced to OneDrive, as this will interfere with the installation process.
1. Download and extract the [Connected Cache provisioning package for Windows](https://aka.ms/MCC-Ent-InstallScript-WSL) to your host machine.

>[!Note]
>* The provisioning package should be extracted to a directory that isn't synced to OneDrive, as the sychronization process will interfere with the installation. It is recommended to extract the provisioning package to the root directory of the host machine (e.g. C:\mccInstaller)
1. Open a PowerShell window *as administrator* on the host machine, then change directory to the extracted provisioning package.

>[!Note]
>* If you are deploying your cache node to a host machine that uses a TLS-inspecting proxy (e.g. ZScaler), ensure that you've [configured the proxy settings](mcc-ent-create-resource-and-cache.md#proxy-settings) for your cache node, then place the proxy certificate file (.pem) in the extracted provisioning package directory and add `-proxyTlsCertificatePath "path/to/pem/file"` to the provisioning command.
1. Set the Execution Policy to *Unrestricted* to allow the provisioning scripts to run.
1. Create a `$User` environment variable containing the username of the account you intend to designate as the Connected Cache runtime account.
1. Create a `$User` PowerShell variable containing the username of the account you intend to designate as the Connected Cache runtime account.

For gMSAs, the value should be formatted as `"Domain\Username$"`. For Local User accounts, `$User` should be formatted as `"LocalMachineName\Username"`.
For gMSAs, the `$User` PowerShell variable should be formatted as `"Domain\Username$"`. For local user accounts, the `$User` PowerShell variable should be formatted as `"LocalMachineName\Username"`.

If you're using a Local User account as the Connected Cache runtime account, you'll also need to create a [PSCredential Object](/dotnet/api/system.management.automation.pscredential) named `$myLocalAccountCredential`. **Note**: You'll need to apply a local security policy to permit the Local User account to `Log on as a batch job`.
If you're using a local user account as the Connected Cache runtime account, you'll also need to create a [PSCredential Object](/dotnet/api/system.management.automation.pscredential) named `$myLocalAccountCredential`.

1. Replace the values in the following provisioning command before running it on the host machine. **Note**: `-mccLocalAccountCredential $myLocalAccountCredential` is only needed if you're using a Local User account as the Connected Cache runtime account.
>[!Note]
>* You'll need to apply a local security policy to permit the local user account to `Log on as a batch job`.
1. Replace the values in the following provisioning command before running it on the host machine.

```powershell-interactive
./provisionmcconwsl.ps1 -installationFolder c:\mccwsl01 -customerid [enter mccResourceId here] -cachenodeid [enter cacheNodeId here] -customerkey [enter customerKey here] -registrationkey [enter registration key] -cacheDrives "/var/mcc,enter drive size" -shouldUseProxy [enter true if present, enter false if not] -proxyurl "http://[enter proxy host name]:[enter port]" -mccRunTimeAccount $User -mccLocalAccountCredential $myLocalAccountCredential
```

---
---

## Steps to point Windows client devices at Connected Cache node

Expand Down
32 changes: 23 additions & 9 deletions windows/deployment/do/mcc-ent-monitoring.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,25 +18,25 @@ ms.date: 10/30/2024

Tracking the status and performance of your Connected Cache node is essential to making sure you're getting the most out of the service.

For basic monitoring, navigate to the **Overview** tab. Here you'll be able to view a collection of predefined metrics and charts. All the monitoring in this section will function right after your Connected Cache node has been deployed.
For basic monitoring, navigate to the **Overview** tab. Here you can view a collection of predefined metrics and charts. All the monitoring in this section will function right after your Connected Cache node has been deployed. You can view more details about each cache node by navigating to the **Cache Nodes** section under the **Cache Node Management** tab. This page displays cache node information such as Status, Host machine OS, Software Version, and Cache Node ID.

For advanced monitoring, navigate to the **Metrics** section under the **Monitoring** tab. Here you'll be able to access more sampled metrics (hits, misses, inbound traffic) and specify different aggregations (count, avg, min, max, sum). You can then use this data to create customized charts and configure alerts.
For advanced monitoring, navigate to the **Metrics** section under the **Monitoring** tab. Here you can access more sampled metrics (hits, misses, inbound traffic) and specify different aggregations (count, avg, min, max, sum). You can then use this data to create customized charts and configure alerts.

Between the two monitoring sections, you'll be able to gather essential insights into the health, performance, and efficiency of your Connected Cache nodes.
Using the two monitoring sections, you can gather essential insights into the health, performance, and efficiency of your Connected Cache nodes.

## Basic Monitoring

### Cache node summary

Below are the metrics you'll find in the **Cache Node Summary** dashboard, along with their descriptions. This dashboard only reflects data received from cache nodes in the last 24 hours.
Below are the metrics found in the **Cache Node Summary** dashboard, along with their descriptions. This dashboard only reflects data received from cache nodes in the last 24 hours.

![Screenshot of cache node summary in the Azure portal interface.](../images/mcc-ent-cache-node-summary.png)

| Metric | Description |
| --- | --- |
| Healthy nodes | Your Connected Cache node will periodically send heartbeat messages to the Connected Cache service. If the Connected Cache service has received a heartbeat message from your Connected Cache node in the last 24 hours, the node will be labeled as healthy. |
| Unhealthy nodes | If the Connected Cache service hasn't received a heartbeat message from your Connected Cache node in the last 24 hours, the node will be labeled as unhealthy. |
| Max in | The maximum ingress in Megabits per second (Mbps) that your node has pulled from CDN endpoints in the last 24 hours. |
| Healthy nodes | Your Connected Cache node will periodically send heartbeat messages to the Connected Cache service. If the Connected Cache service has received a heartbeat message from your Connected Cache node in the last 24 hours, the node is labeled as healthy. |
| Unhealthy nodes | If the Connected Cache service hasn't received a heartbeat message from your Connected Cache node in the last 24 hours, the node is labeled as unhealthy. |
| Max in | The maximum ingress in Megabits per second (Mbps) that your node has pulled from Content Delivery Network (CDN) endpoints in the last 24 hours. |
| Max out | The minimum egress in Mbps that your node has sent to Windows devices in its network over the last 24 hours. |
| Average in | The average ingress in Mbps that your node has pulled from CDN endpoints in the last 24 hours. |
| Average out | The average egress in Mbps that your node has sent to Windows devices in its network over the last 24 hours. |
Expand Down Expand Up @@ -65,6 +65,20 @@ This chart displays the volume of each supported content type in bytes (B) that

The content types displayed in the chart each have a distinct color and are sorted in descending order of volume. The bar chart is stacked such that you can visually compare total volume being delivered at different points in time.

### Cache node details

The **Cache Nodes** section under the **Cache Node Management** tab displays cache node information such as Status, Host machine OS, Software Version, and Cache Node ID.

![Screenshot of cache node details in the Azure portal interface.](../images/mcc-ent-cache-node-details.png)

| Metric | Description |
| --- | --- |
| Cache node name | The user-defined name of the cache node. |
| Status | The heartbeat status of the cache node. |
| OS | The host machine OS that this cache node is compatible with. |
| Software version | The version number of the cache node's Connected Cache container. |
| Cache node ID | The unique identifier of the cache node. |

## Advanced Monitoring

To expand upon the metrics shown in the Overview tab, navigate to the **Metrics** tab in the left side toolbar of Azure portal.
Expand All @@ -79,13 +93,13 @@ Listed below are the metrics you can access in this section:

### Customizable Dashboards

Once you select the charts you would like to track, you can save them to a personalized dashboard. You can configure the chart title, filters, range, legend, and more. You can also use this personalized dashboard to set up alerts that will notify you if your Connected Cache node dips in performance.
Once you select the charts you would like to track, you can save them to a personalized dashboard. You can configure the chart title, filters, range, legend, and more. You can also use this personalized dashboard to set up alerts that notify you if your Connected Cache node dips in performance.

Some example scenarios where you would want to set up a custom alert:

- My Connected Cache node is being shown as unhealthy and I want to know exactly when it stopped egressing last
- A new Microsoft Word update was released last night and I want to know if my Connected Cache node is helping deliver this content to my Windows devices

## Additional Metrics
## Client-Side Metrics

Your Connected Cache node can keep track of how much content has been sent to requesting Windows devices, but the node can't track whether the content was successfully received by the device. For more information on accessing client-side data from your Windows devices, see [Monitor Delivery Optimization](waas-delivery-optimization-monitor.md).
Loading

0 comments on commit de87bbc

Please sign in to comment.