It has been almost a year since I wrote the Azure Portal monitoring basics guide. Given that this article is attracting quite a bit of traffic, I guess that many blog visitors are interested in reading more about this. Therefore, I’ve decided to write a little update covering significant recent Portal changes.
Azure Portal – Monitoring services
Before we dive in, it’s important to know what’s available when it comes to monitoring and diagnostics within Azure. Currently, there are three monitoring services available on Azure; Azure insights, Application Insights, and Log Analytics.
Note: lots of thing have changed in the portal, apart from simply migrating existing offerings from the classic portal or interface tweaks/improvements. However, that will be quite a large list to cover. I’m planning to focus on commonly used Monitoring and Diagnostic related changes only, unless something very important pops up.
Many of you might be familiar with Azure Application Insights, but never heard of Azure Insights. Azure insights are the core monitoring and alerting service, which includes standard Metrics, Alerts, Autoscale, Email Notifications, and Audit Logs.
The name Azure Insights is often referred to within the docs, and therefore helpful to remember this name.
A robust monitoring and logging framework for custom applications allowing you to detect, triage, and diagnose issues. I can recommend using Application Insights, even for existing applications not build to leverage all available features.
Log Analytics is part of the Operations Management Suite and ingests data from your Azure servers to provide rich reporting capabilities and advanced log search based alerting algorithms. This offering is more or less focusing on monitoring logs (system logs, web server logs, etc.)
At this point, I’ve briefly covered what’s available. For the remaining part of this post, I will focus on what’s available in Azure Insights only.
More on the Portal
Within the current Portal version, the following resources support Monitoring and Diagnostics (metrics, logs and alerts);
App Services (Web, Mobile, And API Apps), Cloud services, Data Factories, Load Balancers, Logic Apps, Network Security Groups, Redis Cache, Search service, SQL Databases, Storage Accounts, Stream Analytics Jobs and Virtual Machines
There are other offerings that do support monitoring, however not having all the rich features available.
Webhook and Alerts
Not long ago, the only way of notifying people regarding alerts was by sending an email notification. Email makes it unnecessary difficult when you would like to process the data within a different system. Webhooks (http request) will allow you to route the Azure Alert notifications to other systems for post-processing or custom notifications.
Webhooks are very simple to configure; just configure your alerts as before but in addition provide the URL pointing to your webhook endpoint (the system responsible for processing the data).
A detailed guide including advanced options can be found here: Azure insights webhook alerts: https://azure.microsoft.com/en-us/documentation/articles/insights-webhooks-alerts/
Trigger automation runbooks for VMs
For V2 Virtual machines, it’s even possible to configure running an Automation runbook in case an event has been met. For now, this only contains a list with predefined Runbook, but this might change in the future. After selecting a predefined runbook, the interface will guide you through the Azure Automation setup and deployment steps in case you never used this offering before.
Note: You aren’t limited to use predefined Automation workbooks only. You can simply trigger your custom runbooks using Webhooks. Just enable a webhook for your runbook via the Portal and paste your endpoint within the webhook notification section.
More about workbooks for runbooks can be found here: https://azure.microsoft.com/en-us/documentation/articles/automation-webhooks/.
Auto Scale Notifications
This feature will allow you to configure Scale Action notifications via Email and Webhooks and supports Cloud Services, Web Sites, and VM Scale Sets.
Like many others, you probably like to know when the Autoscale changes the number of active instances (up and down). Luckily this is very simple to set up. Just open the Scale-out option of your service, and you will find the new notification section within the bottom of the configuration blade.
App Service Advisor
App Service Advisor guides you through the process of keeping your services running optimally. The advisor will provide recommendations based on CPU, Memory, and Connections statistics.
New Dashboard features
Last but not least, some dashboard updates. I’m pretty excited about the new Dashboard capabilities, which I have requested about almost a year ago. It’s now possible to create multiple Dashboards, Share (including delegating access), clone, and open a dashboard in full screen like in many other systems containing a dashboard.
Instead of me going over this myself, I would recommend watching this video created by Trevor Sullivan.
Azure Insights is getting more and more powerful. Extending the notification system with webbooks enables some exciting possibilities (notifying Logic Apps, Runbooks, or custom apps). The new dashboard features make proactive monitoring a lot simpler. Other topics worth looking into are: Boot diagnostics, Changes in the Log backup and retention possibilities and Audit events.
Special thanks to Ashwin Kamath for sharing Azure Insights related information!