The Nagios Conference just ended with a question and answer session and a promise for a larger conference next year.  Everyone I talked to was extremely pleased with the Conference.  One thing that everyone emphasizd was the time people had to network and get tips from other administrators deploying Nagios.  There certainly was a wide range of experience at the conference.  I talked with admins who had deployed Nagios 2 days before the conference as well as admins who had used the first version of Nagios way back at NetSaint.

Here are the biggest highlights that I experienced at the conference:

The most informative session I went to was from an admin at a large insurance company who detailed, page after page, all of the small things that they did to enhance performance in a large environment.  This kind of information could not be gathered on the Interent or in any other way because of the detail and the scale that it required to attain.  I will certainly put this in place even in small settings just to prepare for the option of scaling.

The time I was able to talk with other admins working on Nagios was really beneficial as I had a programmer give me a tip on perl that I would have not figured out on my own.  I had admins confirm things that I had taught in classes but in a training situation you cannot always try things on the same scale as some of these people.  I have worked with Nagios Enterprises directly in the training area for over a year now which gave me a great foundation to help people with specific problems that they were facing in their environment.  And it was great to see students who had been in classes for Nagios training that were now at the conference.

It was interesting to be able to talk face to face with the Nagios staff who I send email back and forth with daily.  I had beneficial conversations with Ethan Galstad about some of the specifics we are facing with training and some of the things happening to Nagios XI development.

This Nagios Conference was certainly worth the time and effort to attend.

One of the highlights of the conference was to listen to some of the new features and get a review of current features with Nagios Core and Nagios XI.  Ethan Galstad presented an interesting session that demonstrated some of the upgrades that will occur with both Core and XI.  Most of the people that I have talked with here are using Nagios Core, which is interesting.   So most of these people have good skills at the command line and have been working with Core for a long time.  It was encouraging to see that Nagios is putting new emphasis on developing a number of issues with Core, mostly under the hood stuff, that will increase speed and reduce resource usage.  This is something everyone wants when they begin to scale Nagios.

The Nagios XI feature that I hear most people request is automated reporting.  This is where you can set up who, when and what people will get automated PDFs sent to them to review.  It is great news to hear this is in the testing phase and will soon be rolled out.

I also talked with Ethan about updates on a bug issue or two that have been resolved and will be fixed in the next release so that is also good news on the XI front.

Understanding better the Nagios XI reporting and alerting features will be a project for me in the near future as the presentation clearly demonstrated aspects that I did not know existed.   So I will document these and pass them on to students.

Again, it is great to be able to talk shop about Nagios to everyone at the conference.

I did two sessions this morning which provided some interesting feedback and I learned a few things from people as well.  I did a session on monitoring Linux mail servers and a session on criteria for choosing Nagios plugins.  One thing I learned from a programmer was the difference in impact on a Nagios server whether you are using the standard perl of embedded perl.  Really stuff I should know but didn’t.

One of the great things about the conference has been the ability to walk into any small group and immediately start up a conversation about Nagios and how it is implemented.  There are so many eager people here talking about Nagios it is really a unique situation. I guess I was just surprised at the level of interest.  Having taught Nagios for years, it has given me a foundation for discussions about a lot of aspects about Nagios.  No doubt about it, I am learning a lot as well.

During lunch I had the opportunity to talk with the staff at Nagios.  I just sat down at a table and started talking not realizing who they were.  Of course I exchange email and communicate with them on a regular basis but never had seen them face to face.  Interesting to talk with the staff about Nagios growth and the processes they go through to support Nagios.

The Nagios Conference registration was a great opportunity to get to talk with people.  I spent a lot of time talking with a person from a large company that was testing one of the addons for Nagios Mod-Gearman.  Mod-Gearman is an application that helps you offload the work load on your Nagios server to other workers.  It is all about the ability to scale Nagios in a large environment.  This is one the major advantages of a conference,  rubbing shoulders with people evaluating the same issues that you are because you are bound to learn some valuable information.  There is just so much to know about Nagios.

This company is performing scale as they are dealing with 3 million checks with their Nagios deployment.  It was encouraging to hear that on a large scale deployment of Nagios using Mod-Gearman, it was getting the same positive response that I have found while teaching how to use it in the Advanced Nagios class.  One of the great things about Mod-Gearman is the ease of set-up and the ability to use older hardware as workers.  The other feature that is great about Mod-Gearman is that it is very easy to set up and use.  This provides a rapid method of scaling for situations that occur.

The three hours I spent talking with people at registration was time well spent.  As a trainer it is always important to listen to what people want to learn, the problems they are having and the solutions they have discovered because those are the things we want to be teaching in the classes.

After a 2.5 hour flight to Minneapolis/St. Paul for the Nagios Conference, I had a couple hours this afternoon after arriving to work on a few changes for my presentations.  I am using a laptop to virtualize Nagios Core, Nagios XI and a Postfix mail server for the first presentation.  This can offer a few challenges in terms of resources for a laptop but it is working OK. I am providing a slide presentation with demonstrations from the live Nagios Core and XI instances.  My goal in this presentation is to demonstrate the flexibility that Nagios has in terms of how to monitor critical services on a mail server.

These type of training sessions are tough because you never know the level of skill that may be in the class and the classes do not last that long, only 50 minutes.

Since this is the first Nagios conference in the US there is a lot of anticipation to see how it goes.  I know attendees will get a lot of resources because they are providing PDFs of the conference and access to videos of all of the sessions. That will be a great resource as there are 4 sessions at a time during the conference.

Well, I can hear people in the atrium as they prepare to go to registration.  Sounds like people very interested in the Nagios conference.  The conference should be full as they sold out.

Administrators who will be monitoring their networks using SNMP (Simple Network Management Protocol) usually struggle with the initial stages of SNMP.  It is not easy to be productive until you have struggled through some of the basics.
The the 1st Nagios World Conference in North America being held in Saint Paul, Minnesota (USA) September 27th-29th, 2011 provides an avenue for us to provide training for those trying to get started with SNMP.  The Nagios Conference offers multiple options for training and networking with other administrators.  One of the sessions we will be providing is outline below.

SNMP Library: Trying to Locate Relevant Information
When you enter a library to locate a book typically you browse the books that are available, trying to survey your options.  Once you have a topic in mind you begin to move to a section that is close to the book you are interested in.  For example, if you were interested in a book on ironclads during the Civil War you would go to the American History section.  Within the American History section you would find a section called Civil War and then you would start looking for an individual title that would provide the content you were interested in learning.

In the same way, SNMP provides a large library of information that contains titles on specific topics.  The SNMP library is divided into sections just like your favorite public library.  Within those sections may be sub-sections that ultimately lead to specific titles which in the library are called books but within the SNMP library they are called OIDs or Object Identifiers.

SNMP Library Organization by Numbers
Obviously to find  a book in a large public library means that the library, in fact all libraries, must be organized in the same way.  So, if you know how to find a book on cooking fish in your small hometown library, the process of finding a similar book in a large library has to be similar.  In fact this is true as public libraries have historically used the Dewey Decimal system or the Library of Congress to number books throughout the library.

Just like your public library SNMP is numbered so it is easier to find what you are looking for.  The SNMP library starts with the number 1 and flows outward like a tree with sub-numbers.

The sections in the SNMP library are labeled with the term MIB (Management Information Base).  The MIB structure is required of every device using the SNMP library so as a Nagios administrator you know you have a structure or tree that will be similar on all devices.  One of the most significant MIBs or sections is MIB-II which has sub-sections just like the “history” section or your public library has “american history”, “greek history”, “european history”, etc.  This MIB-II section includes information that is very valuable when monitoring a device with Nagios.  For example, one of the sub-sections is interfaces (mib-2.2) which provides you information about each of the interfaces on the machine you want to monitor.

SNMP Library Discovery Tools
When you go to a large library you can waste a lot of time “browsing” when in fact there are specific tools that can be used to help you locate just the book you are looking for.  In a similar way, in order to access the SNMP library you will need a tool like snmpwalk or snmpget  from the command line.  These commands allow you to access the library and choose a section to browse and ultimately locate the OID (book) you are looking for.  The tool snmpwalk, walks or traverses the entire tree or library.  The tool snmpget looks only for one book in the library.

The session will demonstrate the several available tools and how to create Nagios checks using the tools so that you can leave the presentation with the ability to start monitoring with SNMP.

Nagios Enterprises, LLC is providing the 1st Nagios World Conference in North America being held in Saint Paul, Minnesota (USA) September 27th-29th, 2011.  One of the training sessions which we will be providing is the project of monitoring a Postfix mail server with Nagios.  Below is an outline of the topics that will be discussed.

Host Services Monitored
CPU
* percent used by users, system,iowait and idle time
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts
System Load
* load averages over 1,5,15 minute periods
* email notification when warning or critcal load averages exceeded
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts
Drive Space
* list disk free space and inodes on specific partitions
* warning and critical notification for levels exceeded
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts
System Memory
* listing free memory, used memory, shared, buffers and cached
* warning and critical notification for levels exceeded
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts
Processes
* total number of processes on server listed
* warning and critical notification for levels exceeded
* zombie processes listed
* warning and critical notification for levels exceeded
* verify xinetd daemon is running
* warning and critical notification for levels exceeded
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts

Postfix Services Monitored
Port Access Speeds
* publicly accessible ports 25,993,995 or 143,110 monitored to verify access times for clients
* warning and critical level notification when time periods are exceeded
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts

Queue and Mailbox Management
* verification that email can be delivered
* verification of access time to account
* number of email in a given account
* read header and content of test email to verify integrity of system
* graphing charts of 4 hour, 25 hour, 1 week, 1 month, year, custom
* PDF download of charts

AntiVirus Warnings
* report on the number of viruses captured
* report on the last date of a virus captured
* verifcation that AntiVirus program is running
* last date of virus database update
* list of the number of AntiVirus signatures

CyberMontana Inc. will be providing several training sessions for the conference.

If you are interested in the Conference and would like a $300.00 discount please click on the link below for more information.

Conference Discount

Now part of the learning benefits of taking the Official Nagios Training is that all students will have a one hour a week support class available for no cost. This support class is designed to provide answers to questions as students progress in their learning experience with Nagios.

In order for students to have access to these free classes they must complete the Nagios live training course.

Are you running out of resources for your Nagios Server? Do you wish there was an easy way to create a “cluster” for spreading checks over server Nagios servers? If so then DNX is for you. We will explore how to offload Nagios service checks to other Nagios servers which will help you expand your capacity with Nagios.

Past Classes

SSH Checks
Learn how to use SSH connections in NagiosXI to perform checks on remote Linux servers.  These are the types of checks you would want to perform in a hostile environment as the connections and the data transfer are encrypted.  This class will also show you how to create your own commands for checks that NagiosXI does not perform by default.

Webinject
An introduction to webinject.  This project allows you to monitor web transactions.  We will also be looking at parsing logs for text strings to create your own plugins for applications you may be using.  We will look at doing this with local checks and NRPE.

Nagvis Gadgets
We will be looking at how to set up Nagvis Gadgets and incorporating them within your Nagvis maps with Nagios.

Nagios Fusion
The Nagios Support Class will demonstrate Nagios Fusion which is an interface that allows you to monitor and mange many NagiosXI servers from one server. This represents a distributed model for many many Nagios servers.

Nagios Backups
We will also look at doing backups for your Nagios server both Core and NagiosXI

Nagios Security with ModSecurity
Because the use of passive checks requires additional security consideration for the Nagios server.  This class will explore important developments for protecting the Nagios server.  The security features we will explore are writing an iptables firewall, tcp_wrappers, ModSecurity, AIDE and rkhunter.  ModSecurity provides protection from zero-day exploits and  is a part of the Advanced Nagios Class.

Setting Up Passive Checks on the Server
This class will focus on setting up the Nagios box to receive messages from clients.  We will set up NSCA to receive those messages and communicate securely with the client.

Setting Up Passive Checks on the Client
This class will take you through the steps of setting up a bash shell script to create a check and then use NSCA (Nagios Service Check Adaptor).  You will also see how to set up NSCA on the client so that you can use an encrypted message to connect to the Nagios server.

Understanding Eventhandlers – working with self-healing scripts to restart services when they fail

Extending NagiosXI
Adding Plugins from NagiosExchange – You can see how to implement your own plugins or upload plugins that you need.
Adding Components – Business Process Intelligence, Active Directory,Auto-Discovery, LDAP, integration with Google Mapping,etc
Adding Wizards – There are 33 wizards that can be added to Nagios currently, including mapping, auto-discovery, DHCP, Dell OpenManage for SNMP, DNS, EMail, Sensors, FTP, etc.
Install Dashlets – SANS, Google Map, Health report, etc.

Customized Linux Classes

Over the last year we have been working with a large company that is moving all of their servers from Microsoft to Linux, specifically CentOS. This is an international company that is leveraging their growth by cutting costs on the server operations. Cost cutting was not the only reason they selected CentOS as the operating system, it was also based on stability and security. Down time for this company, as with most companies, is not an option, therefore stability is a major consideration. When approached by this company they had specific requirements for training all of their administrators and engineers. Basically, their entire server support staff were without Linux knowledge with little or no experience. No doubt moving an international company to Linux in one year is a major undertaking.

We took the specific requirements that they had for their administrators and built a curriculum from information that we had and based on special needs they requested. At that point we created an evaluation class held as a virtual class that students could log in and interact in the class with an instructor using VOIP and web conferencing tools. This process allows an instructor to talk, interact, demonstrate and actually access student desktops to help them with the labs.

Evaluation Class

The initial class of 25 included those who would adjust the content so the company could allow students to be trained in 30 hours instead of the standard 40 for Linux Server Administration. This allowed for two, 3 hour sessions each day with a 2 hour break in between. A unique way of providing training to provide optimum learning for students. We provided a 500 page manual for each student as well as 200 videos for students to review online if they needed additional help during or after the class. Each student had a virtual CentOS server provided to them so they could use a CentOS server during the class on their Windows desktop. The instructor discussed a topic, illustrated it and then had students perform one of 175 labs that we provided for CentOS. The focus was on instruction followed by application and interaction with the student.

Additional Linux Classes

At this point we have been providing monthly classes for students in the organization for almost a year. Classes have been live Linux training on location and live Linus training using the virtual desktop. This kind of flexibility allows the company to choose the delivery method that best works for the students. It also provides flexibility because of the International aspect of this company as they have offices most major world countries.

The live online training provided access for students in any country and times for classes were modified so that it was more convenient for students. Our instructors started virtual classes day and night providing any time request made by the company. All of these factors in Linux training demonstrate that live training can be provided internationally with live classrooms on a virtual desktop or instructors can physically be where a company needs training, meeting the needs of companies migrating to Linux.

The Ubuntu server is the most innovative distribution for servers. It enables administrators to gain access to cutting edge technology and implement that with new ideas. If you are looking for the latest and greatest, this is the server option for you.

Focus: rating 9
The Ubuntu focus is to provide an Enterprise level server that is cutting edge in technology and innovative in ideas doing this all with an interest in simplicity. The stated goal of the Ubuntu Server Team is “working on simplifying the system administrator’s life, while focusing on advanced features, rock-solid stability, and high performance”(https://wiki.ubuntu.com/ServerTeam/).

Source: rating 9
Debian was the source of the Ubuntu Server but in time, the distance between these two have increased as innovation has brought new ideas and technologies to Ubuntu. The distance is also created by Canonical’s determination to make two releases every year, on time no matter what. At this point Ubuntu is independent in terms of where it is going and so has no real dependency upon any other distributions and their development. Having a history rooted in Debian’s huge application base fits well with the Ubuntu Server focus and has brought many experienced administrators from the Debian camp to show interest in Ubuntu.

Stability: rating 8
Stability is based upon a user base of installations that provide feedback as well as a determined testing cycle. Unfortunately, Ubuntu cannot be cutting edge and innovative and be as stable as administrator’s would like. The biggest negative feedback that I have received from administrator’s who run Ubuntu in production environments is the number of bugs that seem to creep into the system. If you are unwilling or unprepared to solve small bugs on a regular basis, choose another distribution. Now, not to say that Ubuntu is not stable, it is, but you will consistently be nagged by small issues. The 6 month release cycle does not contribute to stability, other distributions like Debian and Red Hat have learned that.

Security: rating 8
Security and stability work together. As an administrator, you may be concerned about the small bugs and what they do with security. That is certainly something to consider. Again, Canonical cannot provide innovative ideas and hardware device drivers that are cutting edge and at the same time provide the highest level of security, it just is not possible to have both worlds. Developers have tried to implement security principles like logging into the system as an unprivileged user. Complex aspects of security that an administrator may choose like SELinux and ACLs are available but not installed by default. If your organization wants to employ SELinux choose CentOS/RHEL which is designed from the beginning for that purpose.

The Uncomplicated Firewall is designed to be an easier way to manage a firewall from the command line. Unfortunately it is not activated on start up so your server has no network protection initially. The Ubuntu 9.10 server brings three new features to the UFW firewall; outgoing filtering update, filtering by interface and bash completion. This now brings a total of 12 new features since the UFW was first released in version 8.04. Finally, the UFW is reaching a mature stage where you can use it instead of writing rules with iptables. This is a good example of an innovative security option that adds value to the system but was in an immature state for 2 years as it developed.

AppArmor enables protection against those zero day bugs, which is great but, AppArmor has several serious issues. First the future is in question for this application as the developers now work for Microsoft and second, AppArmor requires a creation of a security profile for each application. This creation process is limited by the amount of time an administrator needs to use to implement the application and will never be system wide in scope like SELinux. That said, more profiles are available all of the time.

Blocking the loading of additional modules into the kernel after boot is a security feature added to Ubuntu server among many others.

Cost: rating 9
Access to the install CD and repositories of course is free. Documentation from the community is better than most distributions. Ubuntu as a whole has a fabulous community and they are known for their contribution to documentation and forums. However, any consideration of cost must include the cost of solving small issues that arise with cutting edge software so be prepared to allow administrators time to work with these issues. Also, consider Ubuntu training options for administrators to enhance their ability to understand the Ubuntu system and how to use it effectively.

On the flip side of the coin, developers have been true to their goal of making the Linux server easier to deploy. Without actually installing a wide range of Ubuntu servers you would not be able to understand the hard work that has gone into the Ubuntu server config files. For example, when you install Postfix mail server the default configuration already has keys and TLS set up to make your work a little easier. These innovations are small points that Ubuntu servers do not always get credit for and they certainly will save an administrator time.

Features: 10
In terms of available features Ubuntu servers certainly are rated 10 for ideas that you may want to deploy like, Cloud Computing. The Ubuntu server supports the latest apache, MySQL, PHP, FTP, etc. This distribution is an excellent choice if you need cutting edge versions. Again, what Ubuntu gives up with security and stability it makes up for with features and versions.

JeOS provides a version of the Ubuntu server that is created for appliances. This provides an efficient way to use technology on those lightweight devices that so many organization depend upon.

Landscape, for a fee, allows you to manage a large number of servers from one location.

Ease of Management: rating 10
The use of apt-get is legendary and is a great tool for management. Innovative additions to applications like the configuration offered for Postfix make administration easier and installation process provides simple ways to install applications with security enhancements like MySQL, PHP and apache. Overall, management of the Ubuntu server is made easier by the concepts built into the server install.

Support: rating 9
Support provided by the community in bug reports, forums, and documentation cannot be beat by any distribution. Commercial support is provided by Canonical for those difficult issues that you may be facing.

Summary:
If you are using the latest hardware or if you want cutting edge technology supported by innovative ideas, Ubuntu server is the choice for you. One of the reasons we provide Ubuntu server training is that it provides a real contrast to that of CentOS.