Wouldn’t it be nice to know before a message queue fills up that it’s near capacity? Or to be able to see which apps are hogging CPU, and when?
Measurement is key to improvement, and visibility is essential in understanding issues and pain points. Monitoring is the nexus of those elements and necessarily critical part of highly available solutions.
Enterprise solutions are only as available as the underlying systems. Three-layered architecture introduces APIs across both the system-of-engagement and system-of-record. The benefit of this type of architecture is increased application flexibility along with the enablement of fast-paced development. But a flexible and cohesive design puts more emphasis on an organization’s need to improve governance, monitoring, and control.
MuleSoft has released the Anypoint Monitoring platform as part of their cloud offering. Anypoint Monitoring provides insights and visibility into the application network using monitoring capabilities, log management, and clean dashboards entirely “out of the box.”
Before Anypoint Monitoring, organizations either invested in external monitoring tools, like New Relic, or went without. While there is still a place for advanced monitoring solutions, Anypoint Monitoring provides an entry point into performance monitoring that MuleSoft customers might not otherwise have access to.
Anypoint Monitoring has four key components that improve visibility, accelerate problem identification, and create opportunities for proactive incident management and application tuning.
The monitoring element in Anypoint Monitoring is a 360-degree functional application performance monitoring tool that works “out of the box” from MuleSoft. One of the real benefits of using the Anypoint Monitoring tool is the real-time metrics tracking available as part of the built-in dashboards.
These dashboards can span multiple APIs and integrations. Without the need for external monitoring software, Anypoint Monitoring provides a view into the application architecture and visually represents the relevant pieces of data. Some of the key metrics that are captured in the built-in dashboards include:
- Total number of requests per application
- Average inbound and outbound response times
- CPU and memory utilization
If you’re looking for more than what the standard dashboards offer, you have the option to create custom dashboards. As a tailored solution, these dashboards can serve up application-level activity specific to your environment.
Having this information available and readily digestible helps your operational support teams pinpoint issues. Beyond that, because performance monitoring includes the ability to see trends and view historical insights, it can assist architects and developers in better tuning applications.
APM can also be effectively used as a log management tool to aggregate, search, alert, and analyze logs. With a log management tool, users can quickly search and find raw data points. This facilitates identifying issues and extracting meaningful insights from the logged data more efficiently. Anypoint Monitoring also includes a log tokenization service, which can obfuscate sensitive data like personally identifiable information (PII).
We talk at some length about the usefulness of log management within Anypoint Monitoring in our recent blog post, Traceability for Process Automation, and Improvement in API-led Connectivity. In that piece, you’ll see how we helped a customer use logs and a tracer ID to identify points of failure in a critical business process. Using that ID, we were able to watch messages move through the system, logging critical metrics with the idea and leveraging the logged data to create custom dashboards that provided the data relevant to the business processes.
Infrastructure and application downtime can be planned or unplanned and can happen at nearly any time, from any direction, and in a multitude of forms. The ability to maintain operations and recover quickly from an outage that sets an IT team apart.
System issues and application bugs can be mitigated, but they can’t be avoided altogether. Monitoring solutions play a significant role in helping teams identify a root cause and resolving issues with minimal downtime or business impact.
With Anypoint Monitoring, we have the tools to narrow down the time and effect of an incident. This capability is helpful to support and operational staff – they can pull in the right response teams and provide targeted data to support troubleshooting efforts.
The Anypoint Monitoring visualizer, in particular, is a powerful tool. With the visualizer, you can see an architectural level diagram of your systems and policies and the flow of messages through the applications.
These are all compelling reasons to adopt Anypoint Monitoring as part of your application network. Anypoint Monitoring provides one more tool to prove its worth – alerting. In Anypoint Monitoring, rule-based notifications can be configured to send alerts when systems or applications reach a defined threshold. In addition to “out of the box” alerts, Anypoint Monitoring provides functionality for operations and support teams to create custom alerts to monitor the application network proactively and notify the appropriate teams of potential issues. These alerts can be sent through a variety of communications tools you’re already using, including email, Slack, and text messaging.
Yes, there are systems and instances where advanced monitoring is called for. If you’re looking for cost-effective ways to set up essential monitoring, or just getting started with implementing monitoring and dashboards, Anypoint Monitoring offers a versatile toolset for MuleSoft customers with a lot of functionality straight out of the box. If you’re looking for ways to extend the functionality of Anypoint Monitoring even further, let us know. Big Compass has helped other companies discover the value the visibility, insight, and proactive information that Anypoint Monitoring can provide and can help you build on to that system to create customized alerting, tracking, and system reporting.