Is Nagios The Future Of Monitoring For IBM i?
November 12, 2018 Alex Woodie
For decades, technology professionals turned to established frameworks from IBM Tivoli, CA, Hewlett Packard, and BMC (“The Big Four”) to monitor their hardware and software stacks. But the open source world has caught up with those closed monitoring environments, and a project dubbed Nagios is poised to be the go-to platform for IT monitoring, including on IBM i.
Nagios, if you are not familiar, is a free and open source software product that provides monitoring and alerting for servers, network gear, applications, and the array of services that organizations increasingly rely on. The software –backed by the obligatory recursive acronym “Nagios Ain’t Gonna Insist On Sainthood” – was initially developed by Ethan Galstad back in 2002 to run and monitor Linux systems, which was just starting to make headway in enterprise data centers.
Over the years, the C-based program has been extended to support an array of object types, including host resources like processor load, disk and RAM usage, and system logs for a majority of popular operating systems, including IBM i. The energetic Nagios community has created a wide range of Nagios agents that have extended the framework to an estimated 5,000 devices.
On the network side of things, Nagios Core (as the open source product is known) monitors routers, switches, and other gear, and supports SMTP, POP3, FTP, and SNMP protocols, among others. All remote monitoring can be encrypted, and it also offers a plug-in that lets users create their own monitors in a variety of languages.
Data collected by active and passive Nagios plug-ins is routed into the Nagios server, where the data is sorted and alerts are sent out. Users can consume alerts in a variety of ways, including through a Web browser, through SMS pagers, and through email. With a simple glance at the Nagios dashboard on a phone, a systems admins can see if critical servers, process, and services are alive or dead.
The Nagios Core server software also can work with a “round-robin database” (RRD) to simplify the storage and querying of time-series data. With RRD, admins can dive deeper into Nagios data and view graphs to see performance trends. And with a little setup ahead of time, Nagios can be configured to automatically respond to events as they occur, or escalate alerts to other individuals.
Organizations that want more than what Nagios Core has to offer can find it from Nagios Enterprises, the commercial outfit behind the open source project. The St. Paul, Minnesota-based company, which employs Galstad, extends Nagios Core with several offerings, including Nagio XI, which features a more extensible front-end and a MySQL-based back-end. Nagios XI also brings a better user experience with more powerful graphs and better dashboards.
Unfortunately, Nagios Enterprises doesn’t support IBM i. But you can get a Nagios distribution for IBM i from a little outfit called International Business Machines. In 2017, Big Blue shipped a beta version of an IBM i plug-in designed to feed data from the IBM i sever into the open source Nagios Core product or Nagios XI.
According to a developerWorks page, the IBM i Nagios plug-in collects 15 different data points from IBM i servers, including things like CPU and disk utilization, ASP usage, number of active jobs, and long-running SQL jobs, among others. Some of these services are supported via SNMP, while others require the open source JTOpen Java toolbox to be installed.
The Nagios server itself doesn’t run on IBM i – at least not yet. With the Nagios on IBM i plug-in, IBM provides users with instructions on how to install Nagios Core or Nagios XI on a Linux server. Technically, there should be nothing stopping from a user installing Nagios in a Linux LPAR on the Power Systems server – or even for IBM to port the software to the IBM i OS via PASE – but that has not happened yet.
IBM makes the case for IBM i shops to use Nagios in a PDF presentation on Nagios created by IBM i business architect Tim Rowe and Wang Yun, who developed the Nagios for IBM i plug-in. While Navigator for i provides good coverage of IBM i-centric queues and metrics, that product doesn’t provide customers with any visibility of what’s occurring outside the IBM i system.
Nagios, on the other hand, provides administrators and operators with a single centralized view of all IT assets, Rowe and Yun write. In addition to the core IBM i metrics that are now available via the Nagios plug-in, IBM i users can get trend and historical data, as well as the flexibility to configure metrics to their liking, not to mention the capability to create their own custom plug-ins.
IBM isn’t the only company doing Nagios work on IBM i. A few years ago, ShapPin Cheng launched a GitHub project called check_as400 that moved data from IBM i server logs into Nagios. Considering the extensibility and openness of Nagios, it wouldn’t be surprising to see independent software vendors in the IBM i community adopt the technology for their own systems monitoring offerings.
For now, IBM offers the Nagios plug-in as a beta. Don’t be surprised to see it become a fully supported offering in the months to come, especially considering the love affair that IBM has with open source software these days.