Google Cloud Platform is a serious alternative for SAP Cloud environments. Substantial advances have been made and it now has monitoring systems that enable improved supervision of our system.
Due to the critical nature of SAP environments, monitoring these systems has always been a priority in IT Operations. Traditionally, this monitoring has been satisfactorily resolved with generalist tools that included SAP modules or by complementing them with more specific tools such as Solution Manager or Focused Run.
Although these kinds of tools have worked well in the On-Premise world, the current tendency towards the Cloud world is also affecting SAP Solutions, with a growing number of clients using IaaS Cloud platforms to deploy their SAP systems.
In these environments, although traditional tools are compatible they present a series of disadvantages such as complex connectivity, increased costs due to having to deploy additional machines or more network traffic.
For SAP environments deployed in Cloud, the logical thing to do would be to use native Cloud tools that will integrate with the tools already available for IaaS monitoring, enabling full-stack monitoring in a single point.
In this article we talk about the solution offered by GCP to monitor SAP systems totally based on Stackdriver and, apart from offering a series of out-of-the-box metrics, it allows us to implement our own metrics.
Thus we will be able to integrate SAP metrics in the GCP global monitoring dashboards themselves. Also, following the Google multicloud philosophy, these tools will probably evolve for systems deployed in GCP and also be used with other Cloud service providers or even for On-Premise systems, and therefore we would expect a reduction in licensing costs if we substitute our traditional tools with the new options offered by GCP.
SAP-GCP Monitoring Agent
Google Cloud Platform and SAP have jointly developed a monitoring agent for SAP HANA, which is included in a development roadmap, and will therefore be improved over time with new functionalities. This agent is capable of compiling tailor-made metrics from the HANA instance and sending them to the Google monitoring console (Cloud Monitoring).
The agent comprises a configuration file where we can indicate the connection parameters and the method for compiling metrics. By default, the agent will compile a set of defined metrics from the SAP HANA database.
At the end of the article there is a link to Google Cloud Platform, where you can find information about the default metrics that this agent compiles.
In addition, we can configure YAML files with SQL queries to define our own custom metrics.
In the following image you can see an example of a custom metric defined to show the physical and virtual memory usage of a HANA database:
From the monitoring console we can create dashboards to display graphs of our metrics that can be filtered by different time intervals.
Cloud Monitoring allows to configure alerts based on the thresholds you define for the metrics, so that you can send, for example, an email or SMS to a support team to resolve a problem.
All notification channels are configurable from the Google Cloud monitoring console.
Another very useful feature of Google Cloud Monitoring is the Uptime Checks, an instance that we send to a resource (such as a SAP HANA instance or any of SAP NW) and check if we get an answer. With this we can determine the availability of an instance, a URL, a service or even an AWS load balancer.
Related to this checking, we can create an alerting policy to create an incident when the request fails. As with the agent, the policy can be configured to notify by email, SMS or any notification channel; including relevant information about the resource or the resolution procedure.
The following shows the “Uptime Checks” console with an example of an availability check of a HANA instance attacking the “indexserver” port.
Another need when trying to monitor a SAP HANA system is the use of the filesystem. Knowing whether the HANA data or log has been filled is necessary if we want to ensure the availability of our environment.
For this purpose, Google has developed a global monitoring agent, Cloud Monitoring Agent.
This agent is based on the “collectd Daemon” that collects system metrics from Google instances to send them to the Google monitoring console.
The following graph shows an example of a metric configured with the Cloud Monitoring Agent to show the usage of the filesystems:
On these metrics we can configure an alerting policy. In the following graph we can see an alert configured to notify in the event that the filesystem usage threshold exceeds 85%.
Moreover, this agent is able to monitor metrics such as disk, CPU, network traffic and other metrics that you can see in the following link:
Google provides us with a mobile app of the Cloud console in which we can manage the incidents caused by the alerts configured in the agents. From this app we can confirm and resolve incidents without having to turn on our computer. This app is available both in Google Play and in the App Store.
Ultimately, at present, Google Cloud Platform has a very comprehensive infrastructure, which is in continual development and continually evolves, to ensure the availability of our SAP environment, offering simple and efficient functionalities so that we are able to react to the incidents that our SAP environments generate.
GCP link with metrics obtained by the agent.