Linux Today: Linux News On Internet Time.

OpenNMS Update v1.37

Dec 06, 2000, 00:01 (0 Talkback[s])
(Other stories by Shane O'Donnell)

Date: Tue, 5 Dec 2000 17:27:17 -0600 (CST)
From: announce-admin@opennms.org
To: announce@www.opennms.org
Subject: [opennms-announce] OpenNMS Update v1.37

OpenNMS Update
Vol 1., Issue 37

In this week's installment...

* Project Status

+ Announcing: Testdrive Release

+ New Release Strategy

+ Coding Projects Underway

* Out and About - Atlanta/Pittsburgh/NYC/Boston

* Have I mentioned Bugzilla?

* The Wish List

* Afterthoughts...

Project Status

Announcing: Testdrive Release:

Go ahead. Get in. Take 'er for a spin!

On behalf of the entire team at OpenNMS, I'm proud to announce the completion of the "Testdrive" release of the OpenNMS software!

The "Testdrive" release, aka Version 0.4.0, is the initial cut at a "single-box solution" that will be shipped (knock wood) on the upcoming release of one of the major Linux distributions. We worked hard to pack as much of the functionality from the first scheduled release (now slated for 1Q01) into this one as we could. Here's a run-down of the major components that are shipping, what they do, and what this means to you:

* What: SCM

What it Does: Service Control Manager. It provides a mechanism to control the start-up, shutdown, and status of some of the background processes.

Why You Care: SCM is user-configurable so you can include your own processes, exclude processes, etc, for your environment. But basically, the important part is now you have an easy way to start up the "guts" of OpenNMS

* What: SCMUI

What it Does: Service Control Manager User Interface. Effectively, this is your "eyes" into the current status of the background processes. It also provides an easy interface to start stopped processes, stop started processes, and shutdown the whole ball of wax if'n ya need to.

Why You Care: Starting and stopping applications and processes is a good thing.

* What: OpenNMS

What it Does: Basically, this is the user interface. A common login dialog is presented and depending who you log in as, you either get a user view of managed devices or an administrator interface.

Why You Care: Once you've got the processes started, it's nice to be able to both use them as well as configure them.

* What: Help

What it Does: Help is a Beatles' movie from 1965 in which Ringo is to be the human sacrifice for a cult and the rest of the Fab Four have to save him. It's also our graphical user interface to the included documentation.

Why You Care: I don't know about you, but on occasion, I actually RTFM. And in this case, there are four of them--a User Manual, an
Administrator Reference, an Overview Manual, and BAM! - A Quick Start Guide. If you choose to ignore these, at least check out the Quick Start Guide. It will be 10 minutes well spent.

* What: Various Background Processes

What They Do: Everything you want them to do, and more. Basically, this is where the work happens, including the processes responsible for network discovery, capabilities checking, receiving SNMP trap s, handling events, invoking automation (running scripts on event receipt, integrating with trouble ticketing and notification systems, etc), scheduling the next device polls, and determining which events impact which nodes and which users should be informed of them.

Why You Care: Basically, if you don't care, you should have stopped reading before now. Assuming you do, the SCMUI will give you a good cursory overview of the processes that are running, and the Admin Guide will give you an overview of what they actually do.

* What: Other External Requirements

What They Do: Oh, you know, little stuff like a database. We've built everything in this release to leverage Postgres on a Linux box, but there shouldn't be much keeping anyone from trying it on NT, either. Of course, that means that you'd have to WANT to run it on NT and if so, this might be a handy outlet for your BDSM tendencies...

Why You Care: Things just don't work without the external dependencies. For deeper insights into what you need, check out the Quick Start guide.

How do you get it? From the web site, my dear Watson. You can download the latest and greatest (Version 0.4.0-2) in either tarball or RPM format from http://www.opennms.org/files/releases/OpenNMS/ and if you are simply in a readin' mood, you can get our documentation separately at http://www.opennms.org/files/documentation in either HTML or PDF format.

If this is your intent, note the following file/directory names:

      adminref==Administrator's Guide;
      userguide==User's Guide;
      overview==Overview Guide;
      quick==Quick Start Guide;
Soon (if not by the time you are reading this), these files will be linked from our http://www.opennms.org/devdocs/ page.

And remember, download early and download often.

New Release Strategy:

With the "Testdrive"/0.4.0 release out of the nest and learning to fly, we've decided to add a little more method to our madness on upcoming releases.

Henceforth and hithermore, we shall release as stable production releases those with an even minor number (e.g., 0.4.0, 0.6.1, 0.8.3) and all new development will be done in releases with an odd minor number (e.g., 0.5.0, 0.7.0, etc). This is Linuxy and we like it.

This should help to add a little clarity to our release strategy, and we can all use a little clarity now and then.

The "stable" releases will be updated with bug fixes and will be released as a significant number of fixes are included, but all true enhancement will be pushed into the development branch and will be available only from CVS tree nightly snapshots (available in the Downloads section of the web site) or directly from CVS.

Hope this helps.

Coding Projects Underway:

This week, we get to kind of change things up, as we are in the transition from the "single box" development target to again refocusing on our distributed architecture. As such, the list will reflect specific new efforts going forward and will not mention bug-fixing.

* Maji Prelim Work -- Rick is active on the "events" mailing list.

* MIB Compiler -- Official research underway. Should have a direction to report next week.

* Distributed Architecture -- As soon as he's done fighting with home repairs and contractors, Weave will be back on the SOAP box.

Out and About - Atlanta/Pittsburgh/NYC/Boston

Just a heads-up to those of you in the Atlanta, Pittsburgh, NYC, and Greater Boston area that we've got some upcoming trips scheduled to head your way. If you'd like to hook up with us to discuss the project, potential partnership interests, employment opportunities, or simply ply us with alcohol, drop a line to either Luke (luke@opennms.org) or me (shaneo@opennms.org) and we'll be glad to share our itineraries and/or beverage preferences.

Our other goals for these trips are easy: Buy a Penguins jersey and find out what the hell a Bruin is.

You can find out the rest of the general details on our trips and other upcoming road trips on the web site at http://www.opennms.org/engage/

Have I Mentioned Bugzilla?

If you would indulge me, I'd like to take a minute to be frank.

   Frank: Please download the Testdrive release, try it out, and report
          your bugs to Bugzilla.
   You: Gee, Frank, that doesn't sound too hard.
   Frank: You're right! It's not! But when you find a bug, make sure you
          do a search of Bugzilla to make sure someone else hasn't reported
          it first.
   You: That only makes good sense, Frank. But what do I do if I find a
        typo or other such glitch in the documentation?
   Frank: How very nice of you to ask. We are tracking any defects in the
          documentation in Bugzilla as well, but we'd like to have all of
          those entered as additional descriptions in Bug #100 in Bugzilla.
   You: #100?!? That's easy to remember. Was that number fortuitous, or
        did you plan it that way?
   Frank: I'd rather be lucky than smart any day.
   Shane: Shut up, Frank, you dumbass.

The Wish List

For all of you who offered services testing OpenNMS in your environment--this is your cue. You're On!

Bug reports are great, bug fixes are better. Enhancement requests? Take a number like everybody else. And enhancement requests can be filed in Bugzilla, too. Never underestimate the power of Bugzilla!

Go, Go Bugzilla!


I've grown so accustomed to using the end of the Update as my own personal forum that I've decided to give this section a name. That way, in the future, you'll know when to stop reading...

I was wondering...have any of you ever come across any studies done that address the value of "maps" in visualizing networks and solving network-related problems? Here's why I ask...

Personally, I've never been a big fan of maps. In big environments, they are pretty much unusable and most of the people I've worked with who I considered good "router jocks", never really seemed to use the maps anyway. When it came time to fix a problem, they were driven by the events and then worked with the device.

It seems as if we're in an industry dominated by "mapping" tools, but no one seems to know why. I've come across one situation in which maps are almost mandatory, but only one, and it could probably be addressed with a different visual metaphor.

I'm interested in your feedback. I think there's probably a dissertation hidden in the answer to this question (or at least the quest for an answer...). You can send responses directly to me or post to the "Discuss" list. And thanks for your input.

Perplexed by the obvious,

Shane O.
Shane O'Donnell