“You would be crazy not to use the potential of Open Source”
Interview with Gerhard Lausser, Nagios expert, plugin developer (check_logfiles und check_oracle_health) and author of various Nagios publications.
In 2009 you’ve published a book on Nagios features and functionalities: which were the main focus points and what’s your experience in relation to the plugins you’ve developed?
The goal of my book was not to be another recapitulation of the official Nagios documentation. My target audience are system administrators who already have enough knowledge to run a Nagios installation. I wanted to write a cookbook which helps them to go the next step: From a basic monitoring to a monitoring which takes care of the needs of big companies. Today is is not sufficient to just look at network connectivity and disk space usage. No more than is is enough to check if your database or webserver is answering requests. Business is time-critical, so I described how to deep check the performance of these systems. My book focuses mainly on the usage of plugins for enterprise-relevant monitoring items like databases, webservers, SAP, logs. Some time ago I started to write plugins which cover these fields, so i could use my “insider knowledge” to help the readers get out maximum benefit for his company.
Let’s think from an IT manager point of view: which are the advantages of implementing an Open Source based monitoring? What’s your professional experiences on that?
You don’t have to think like an IT manager here. Implementing an Open Source monitoring solution gets you the same value for less money. Except when you have special requirement that only a proprietary product can solve, you would be crazy not to use the potential of Open Source. If you have skilled IT staff, then all you have to invest is time. If you don’t have the time, then there are providers who will gladly help you implement and customize your Open Source Monitoring solution. And if don´t you want to operate the monitoring yourself, these providers will do it for you. You can even have a complete package with support and automatically software updates.
So you may ask: “where’s the benefit?”. You don’t have to pay for a software license. And, considering the prices per node per month, we are talking about thousands of Euros, sometimes hundreds of thousands. My personal experience is: whenever we replaced proprietary monitoring installations with an Open Source solution based on Nagios, our customers couldn’t believe how stupid they were. Spending so much money just for software licenses. Open Source monitoring solves most of your problems out of the box for free. The rest can be implemented with a fraction of the amount of money you spent for licenses.
Open Source and big corporates have been two unfitting worlds for too many years, but now something is changing. What do you think about the current situation and the new prospectives for the future?
I have worked for big companies before and met a lot of Open Source there. But these were sometimes “illegal” installations. Admins always knew Open Source, but management didn’t know how to handle it. Today these admins are the managers and most of the decision makers grew up with Linux, so they have an idea what Open Source is all about. It will be very difficult for proprietary monitoring solutions to compete with the free opponents, because also for the latter you can get service and support at a very high level.
Concerning this, some of your plugins were thought to check Oracle DB and SQL server environments in complex infrastructures . How was your experience in this direction?
The first of the database plugin family was check_oracle_health. I did not write it just for fun. Some years ago i had a customer who wanted to replace Oracle Enterprise Manager which he used to monitor all the Oracle database servers. He had two things in mind: to save money of course. And the second idea, which i found more interesting, to standardize his monitoring landscape. There was already a running Nagios installation for hardware and os-level checks and he didn’t want to switch the tool just to have a look ath the databases. So he asked me to implement as much as possible of Oracle Enterprise manager’s functionality in Nagios. That’s how the plugin check_oracle_health came into being. Today I receive mail from users from all over the world and I looks like check_oracle_health became a de-facto standard when it comes to monitoring Oracle databases.
This motivated me to re-use the code and write plugins for MySQL, DB2 and MS SQL Server. They also became a big success with a broad user base. Especially check_mssql_health is very popular, as it allows you to build a bridge from Nagios to the Microsoft world at enterprise database level.
What do you think about the potential of Nagios compared to ‘’younger’’ solutions like Icinga or Shinken?
I think the development of the Nagios core has come to an end. At least I would not expect any new features. The overall concept of Nagios has been tried and tested for several years now and missing functionality has been implemented in AddOns. What i see today is that Nagios becomes the core of several commercial monitoring solutions which consist of compiled and packaged Open Source components and a support option, like your NetEye or OP5’s Monitor. There’s nothing wrong with this, because companies are willing to spend money on a combination of Open Source and support.
On the other side, there are other, not nagios-compatible Open Source monitoring solutions with an increasing market share. If the Nagios ecosystem wants to stay on top of the list, there has to be innovation in the long term. And in order to send strong signals to potential users, the innovation mustn’t take only place in the add-ons around the core, it must happen in the core.
That’s where Icinga and Shinken enter the game. Icinga focuses more on the overall package. A lot of work is invested in the mighty interface “Icinga Web”. Shinken on the other side is a completely new monitoring system. It’s a rewrite of Nagios in a modern programming language with the goal to make further development very easy. And it’s a redesign with the goal to make distributed, loadbalanced and fail-safe setups a breeze. Regardless of the new techniques used, Shinken is still fully compatible to Nagios, which means you don’t have to modify your configuration and plugins.
All of Nagios, Icinga and Shinken have their advantages, maybe they will coexist, maybe they will inspire themselves, maybe there will be one winner in the end. It’s the Open Source community who will decide.
Many Open Source oriented companies use to embrace the User Group formula to establish a more direct relationship with their customers. What do you think about this kind of strategy?
Being open and talking to people at eye level is the only way to establish trust in you and your product today. Customers are smart, they don’t spend money on promises, they compare, they want to feel comfortable. Buying decisions are no longer based on fancy glossy brochures. Organizing or taking part in Open Source user events shows the community that you’re not some evil company just looking for money. Instead you have the opportunity to listen to users’ problems and ideas and to immediately give a feedback. By exchanging knowledge and stories they will learn that you have problems too, but are working hard to solve them. Formerly companies were the experts and customers were the novices. Today you have highly skilled people among your customers, who will not respect you because of your knowledge. But as soon as you talk shop with them, they will respect you because you make your customer’s life easier by doing the work he could do himself but has no time for.
You took part to the Nagios conference in Bolzano with the possibility of getting some information about the IT monitoring NetEye. Which was your impression and what do you think about this solution?
First of all, let me thank you for organizing this event. I enjoyed coming to Bolzano two times not only because you have such fine, sunny weather here. Like I said, Nagios is becoming more and more the basis of out-of-the-box monitoring solutions. NetEye, being one of them, saves the admin a lot of time because it has everything on board you would otherwise have to download, compile, install and maintain. As NetEye user you have monitoring from the hardware up to the business process level, you have reporting, inventory, trouble-tickets and configuration in one tool. And you have a partner who can assist you if you need some special features. This is how system monitoring and management should work today.
Mr. Laußer, thank you very much for this interview
Oreste Attanasio
Author
