Archive for November, 2008

Holiday

Thursday, November 13th, 2008

When I first started my own business around OpenNMS, I figured it would be a one man consulting shop. I’d work about half time and then take the rest of the time off to do other things.

It didn’t work out that way.

My clients wanted support, and that meant someone (i.e. me) had to be around all the time. Instead of having some time to myself I had no time. This went on for a number of years. I can remember missing the funeral of my favorite uncle because I had a client coming to town for training and I couldn’t cancel at that late date. It was just the way it was.

As OpenNMS grew I was able to take some time off. I hired other people and so I could get away for the occasional long weekend. Or, more accurately, I was able to take weekends off now and then. We have traditionally closed the office from Christmas to New Year’s Day, but even then I tend to work part days during that time just to stay caught up.

Last year was the turning point. We got a client in Hawaii, so I had to head out there for a week. My wife insisted on coming along, and after talking about it with the other folks in the company I decided to take my first two week vacation in … well … pretty much forever. No e-mail, no phone calls, no meetings. Just some time to relax and regroove my brain.

It was fun, so we decided to do it again this year. Since I travel often I get lots of frequent flyer miles, and I had enough for a trip to New Zealand for two. I’ve always wanted to go there. However, the seats go fast so you have to book as early as possible, so I booked a flight for this November back in December of 2007.

When the airline was routing us to New Zealand, they had us stopping over in a place called Nadi. Now, I don’t have a problem going to “naughty” on holiday, but I’d never heard of it before and asked where it was. It turns out that this is the major international airport for Fiji. Fiji sounded cool, so we decided to spend a few days there before heading on to New Zealand.

So starting tomorrow I will be gone for two weeks. As much as I love working on OpenNMS I will not be checking my e-mail, or the lists, or bugzilla, or anything remotely related to work. I’ll be snorkeling on a coral reef or trout fishing in a mountain stream but it won’t have anything to do with the OpenNMS project.

I’m even leaving my laptop at home. This does not mean that I will be without electronic gear, however. I am bringing at least two small gadgets with me.

The first is a Garmin GPS. One thing I like about the Garmin is that it is supported by the New Zealand Open GPS Maps Project. Instead of shelling out US$150 for maps I’m using the work of this group. It was pretty easy to install, and the addresses I looked up for the B&B in Auckland and Zorb were both there. I’ll report on how it goes. While I seem to make it a habit to visit countries where they drive on the left side of the road, this will be the first time I’ve actually done it. I learned from my trip to Australia not to follow the traffic circle directions from the Garmin (as they have you going counter-clockwise in these countries which will get you killed).

The second is a new iPod Touch. While I disagree with the degree of control Apple exercises with the App Store, I haven’t seen a better mobile browser, and I just can’t stay away from the ‘net for that long.



The best news is that the second generation Touch supports a microphone. I had been waiting and waiting for Apple to release its version, but with my trip just days away I went looking for other options.

I found it in a product from Fastmac. They sell an adapter that lets you insert a standard 3.5mm stereo headphone jack (with two rings) which terminates in a three ring connector that supports a built in microphone. Combined with Fring, a free app from the App Store that allows you to use Skype (among other things) I should be able to make Skype calls if need be. It arrived today and works fine, so I’m set.

To everyone in the community, hold down the fort while I’m gone. Try as I might, I won’t be able to keep my mind totally off the project, but I am going to give it my best shot.

Christmas Comes a Little Early

Thursday, November 13th, 2008

I just wanted to say thanks to one of our clients in Chicago who sent us this nice Cisco 48-port manageable switch for the OpenNMS lab.



I had posted a question on the discuss list about manageable switches and they had one they were phasing out. Now if anyone has a decent SPARC system they want to get rid of …

Lintopia

Wednesday, November 12th, 2008

When I was at the Linux Live show in London, I was introduced to yet another social networking site called Lintopia.



I don’t know much about it, but I registered and thought I’d share it. It aims to be a site for people involved in open source software. I haven’t had time to see what it offers over other sites that might be open source specific except for the “project” section (I added OpenNMS).

The only social networking site I somewhat frequent is LinkedIn (and to a lesser degree the more European-centic Xing) so I’m not sure what will happen with Lintopia, but its open source focus at least makes it interesting to me so I thought it might interest others..

Open Core Software

Tuesday, November 11th, 2008

I got to chat with Jay Lyman of the 451 Group today. Most analysts (with the exception of CotĂ©) tend to ignore our little ol’ project but the 451 Group likes to get in touch every several months or so.

It was fun to think and talk about the market for open source management tools for awhile, and I also wanted to find out who came up with the term “open core” (apparently it was Andrew Lampitt). I first saw it used in a 451 CAOS Theory blog post by Matthew Aslett.

As my three readers know, I have a bug up my skirt about the “hybrid” open source companies, where part of the code is closed off but part is open. I call it the “shareware” model where some of the functionality it open but you have to buy a commercial closed license to get all of the features.

I won’t rehash all of the reasons why I hate this model being referred to as “open source software” but I would like to focus on the term “open core software”. In my mind it is more descriptive of what the hybrid companies are doing than “open source” and it doesn’t have the negative connotation that “shareware” brings to mind.

So I plan to steal it.

I don’t think that you’ll see “open core network management” any time soon in the marketing literature of these companies. “Open source” is a much more valuable term, but “open core” is at least a more honest approach, meant to represent a company that releases the core of their product under an open source license but generates revenue as a vendor of proprietary additions to that core.

Jay asked me how our business was doing in the wake of the worldwide recession, and I said that on the whole things were okay. Fourth quarter is usually strong for us, and this year is no exception, but currently we have a lot of projects sitting in limbo. Everything has been approved but most of the large companies we are dealing with are holding their breath on spending in any form. We’re still seeing 40% to 45% year over year growth, which is a lot less than some of the open core companies are reporting, but then again we’ve been seeing that for four years now and they just haven’t been around that long.

He also asked what I thought about the business prospects for the open core guys and I had to admit that they don’t look that great from where I sit. Since they tend to compete on cost, there are at least three main pressures that are going to be bearing down on them in the next few months.

First, pressure from the established commercial players. Sure, the open source part of their product is free, and it is hard to beat free, but what about their commercial part – the part that they sell to keep the doors open? A small OpenNMS customer has a network of 1000 devices. Some of the open core vendors charge per device, on the order of US$250 to US$500. So this translates to US$250K to US$500K per year. One can usually get a Tivoli or OpenView solution for 1000 nodes at much less than that. Will someone go with the newcomer or the established player when the established player costs less?

Now, everyone realizes that there is no one out there paying those prices for the new guys. If you have a 1000 node network you can easily negotiate them down to a fraction of the list price as they are desperate to get new clients. And for some people getting 90% off of a price that was made up in the first place is a deal. But if what they are buying are “ease of use” features the second source of pressure will be from the smaller commercial players like Solarwinds. Why pay US$25K a year when you can get a complete out of the box solution for slightly less than that and only pay for it once? Orion runs on Windows, it is very pretty and it is reasonably priced. And you don’t have to deal with that pesky open source stuff.

The third source of pressure comes from the pure-play open source guys. A couple of years ago I had dinner with Peter Fenton, a smart VC with Benchmark. He told me that a product really needs to “own the bottom”. I didn’t agree with him and pointed out things like Mercedes Benz that sell to the top. I was told by others that I wasn’t supposed to argue with the VC, but I did anyway and in the end he convinced me he was right. After all, Mercedes had to seriously reinvent itself when Lexus came to market with a similar product at a lower price.

By having a totally free alternative, users of other products will constantly question their need to pay for commercial software. In many cases, like the small and medium businesses that use Orion, something like OpenNMS isn’t quite for them. But for others, products like OpenNMS provide a real alternative to their current solutions.

Most of the open core players are VC funded, and eventually those VCs will want to cash out. To do that they need to increase revenue. Thus you can expect those low, low, introductory prices to go away, and when companies find out the real cost for the open core software they will be looking for alternatives. They will be looking toward the bottom.

In the meantime we plan to keep running OpenNMS profitably. This means that year after year it will keep getting better, and slowly but surely the pressure on those other solutions will just keep building.

For those companies it is a race; a race to build themselves up and get acquired before they lose their customer base. It is hard to compete against the bottom, which was the point Peter was trying to make. I think Sun’s current lack of success with MySQL has made everyone a bit cautious about open source acquisitions, and with the current economic climate I don’t expect that to change any time soon. In that context the prospects for open core companies don’t look so rosy to me.

OpenNMS 1.6.0: Birthing an Elephant

Friday, November 7th, 2008

Sorry for the light blogging, but as you can imagine we’ve all been busy. We were finally able to release the next stable version of OpenNMS, 1.6.0, at the end of October, but I wasn’t able to write about it.

Getting a new stable release out can be painful. I often refer to it as “birthing an elephant”. We want our stable releases to be perfect, but as it has been said “perfect” is the enemy of “done”. When we get close to a stable there is always the driver of just one more fix, just one more feature, before we tag. I finally had to put a stake in the ground and say 1.6.0 will come out before 31 October, and we made the deadline by several days.

Unfortunately, I didn’t have any press ready at the same time. I’ve just gotten around to writing the press release, which I share here. I guess we care more about getting the code out than advertising it, but I’m not sure that is always a good thing.

Anyway, here’s a short overview of all the work that went in to 1.6.0. Hope you enjoy the release.

PITTSBORO, NORTH CAROLINA – The OpenNMS Group announces that OpenNMS, the world’s first enterprise-grade network management platform developed as 100% free and open software, has released a new version: 1.6.0. This is a stable, production release that incorporates nearly three years of development.

The last production version, 1.2.0, was aimed to compete squarely with Hewlett-Packard’s OpenView Network Node Manager product. This release builds upon that work to expand the reach of OpenNMS to other parts of the OpenView family as well as to provide an open source alternative to products such as Tivoli’s Netcool.

OpenNMS 1.6.0 sports a redesigned user interface, a number of scalability improvements and increased integration with other products. OpenNMS now runs on Windows, in addition to most flavors of Linux, Solaris and Mac OS X.

OpenNMS has four main functional areas: Automated Discovery and Provisioning, Event Management, Performance Data Collection and Service Monitoring. Each area has been greatly improved with this release.

While OpenNMS contains a robust automated discovery system, when managing tens of thousands of nodes it is often preferred to allow an external system to determine what OpenNMS is to monitor. Thus OpenNMS 1.6.0 contains a new “model importer” feature that allows node, interface and service information to be imported directly into the system using data in an XML format. One company, Swisscom Hospitality Services of Geneva, Switzerland, uses this method to manage over 70,000 devices with a single instance of OpenNMS.

Event management has been improved with the new Alarms subsystem. OpenNMS can receive events from a number of sources, such as SNMP traps, syslog, TL/1, and custom scripts. A key can be configured for each event that will allow it to be turned into an alarm. Thus if a device is generating multiple, identical events, their number will be reduced into just a single alarm. This greatly decreases the amount of event “noise” that operators see.

In addition, automated actions can be performed on alarms. For example, events that signal problem resolution, or “up” alarms, can be matched with “down” alarms to clear them. Event workflow can be built into the system by using these automations to manage the alarm list, thus freeing up the operators to focus on the most important issues.

Data collection saw many improvements as well. With the proper hardware, OpenNMS is able to collect over one million data points every five minutes. This data can be from SNMP (versions 1, 2c and 3), JMX, HTTP, or NSClient. The collected data can be exported via the web user interface. Reports showing the highest and lowest values for a particular set of data points (Top N Reports) can also be created, and 1.6.0 contains a vastly improved thresholding system. Thresholds can be generated on individual data points, combinations of data points, as well as a “relative or absolute change” such as when a value shows a sudden increase or decrease.

OpenNMS was originally designed for network service monitoring, and that functionality has been increased as well. New monitors for such things as Windows services are now available, as well as more advanced synthetic transactions. The Page Sequence Monitor was created to monitor a complete web-based transaction, while the Mail Transport Monitor determines the full round-trip availability of a mail service.

The new Event Translator feature allows for the creation of “passive services”. Passive services are those that are driven strictly by events versus an active poll. This is useful in monitoring devices that don’t support direct network connections and thus can be managed by a proxy device that then sends status events to OpenNMS.

Probably the biggest change was the development of remote monitoring. Using a small Java webstart application installed on a remote system, OpenNMS is able to monitor service availability from the point of view of the remote system. Combined with the Page Sequence Monitor one can measure the user’s experience when visiting a website from various remote locations. Papa Johns Pizza (headquartered in Louisville, Kentucky, USA) uses OpenNMS to monitor its online pizza ordering system which has been responsible for more than US$1 billion in revenue.

As OpenNMS was designed as a platform, there are numerous ways for external applications, both open and proprietary, to integrate with it. There is a new Trouble Ticketing API that allows for two-way communication between OpenNMS and a number of external ticketing systems such as Jira, Concursive (CentricCRM) and OTRS.

OpenNMS is free and open software, which means that all of the code is available at no cost and can be modified and distributed under the terms of the GNU General Public License. There are no commercial or proprietary versions of OpenNMS.

The application is maintained by The OpenNMS Project (http://www.opennms.org) and is supported worldwide by a large and active community. Commercial services, such as training, consulting, support and custom services, are available from The OpenNMS Group (http://www.opennms.com) which acts as a steward for the project.

About the OpenNMS Group, Inc.: The OpenNMS Group maintains The OpenNMS Project, the world’s first enterprise-grade network management platform developed under the open source model. They provide commercial services such as training, consulting, support and custom development around the OpenNMS application. This allows their clients to get all of the benefits of open source coupled with more security and accountability than they can expect from commercial software companies. OpenNMS Group consultants have years of experience with commercial management products from companies such as Hewlett-Packard, IBM, BMC Software and Computer Associates and they use this experience helping clients in over 20 countries to improve their management capabilities while reducing costs by migrating to OpenNMS.