Project of the Month: NoMa up a notch

June 2010: In a large and distributed environment, multi channel alerts for hundreds of users can get complicated. Indeed that’s where NoMa came in handy for one of our longstanding clients. With no less than 60 alert definitions, NoMa had been managing Nagios alerts for the corporation’s 1000+ users dependent on central services, across their 80+ remote sites worldwide.
All was working well, but as always things can work better. One pesky characteristic was that if a highly interconnected component were to fail, not only would the admins receive that alert, but they would also receive individual alerts for all the other services and components that were subsequently affected. This domino effect inundated the admins with messages that ultimately stemmed from the same issue. So the proposal was put forward – how about bundling all related errors into one summary message, but still be able to individually (de)activate each alert?
Consultant William took on the challenge, thereby setting the foundations of NoMa’s latest release, version 1.0.5. By transferring the persistence coded in Perl memory to a database NoMa had improved stability and failover precaution built in. This enabled the new feature of central outage correlations and alert summaries. Further improving the user friendliness, the team also simplified the (de)activation of alerts to be executable on the web interface, without the need for restarts.
All in all, thanks to their sponsorship and close cooperation, NoMa will soon be available as v 1.0.5 equipped with greater flexibility. As the biggest contributor to NoMa, a company representative even presented the project at the OSMC 2009. Check out the presentation for more info and keep an eye on netways.org for the official release coming soon!

Weekly Snap: NETWAYSGrapherV2 update, World Cup telecoms, OpenITSM

24 – 28 May announced a NETWAYSGrapherV2 performance update, touched on open source IT Services Management, and looked forward to the World Cup from a telecommunications and NagVis perspective.
Karo got excited about the World Cup coming up in South Africa – and the telecommunications services that Deutsche Telekom customers will be able to enjoy during their stay over there. With Nagios monitoring their three roaming partners on the ground and special plugins to ensure internet login and connectivity, SMS, MMS and data transfers all should flow smoothly during the World Cup craze. To top it off, Deutsche Telekom operators will be able to constantly monitor the smooth running of communication channels through a new NagVis dashboard we recently created for them.
Following on the sevices theme, Julian shared his presentation on ‘IT Service Management with Open Source Software “OpenITSM”’ which he recently made at the Open Expo 2010 in Bern. The presentation covers various open source tools for Incident & Problem Management, Event Management, Operations Management, Service Desk and CMDB.
Finally Eric thanked the many people who gave us feedback since the release of the final NETWAYSGrapherV2 with a small update to improve the chart display in the dashboard and panels. No excuse to make a trip to the coffee machine, now that the performance is significantly better for large environments. As always, get the update from netways.org.

Weekly Snap: Nagios in Magna, Kempten, Saxony’s Public Attorney & T-Mobile plus NSClient++

weekly snapJan 18-22 was dominated by our project consulting team with Nagios monitoring left right and center.
Birger got around, visiting Magna in Thuringa, Kempten in Allgaeu and the Public Attorney in Saxony–Anhalt. In Thuringa, he threw praise on Magna’s orderly data center and software configurations during a Nagios consultation, and in Saxony-Anhalt did his bit for civil society in helping the Pubic Attorney get monitored by Nagios. In Kepmten Birger marvelled at their winter wonderland while on his way to fix a SNMP trap connection to EventDB by snmpd and snmptt.
With less travelling, but equal roaming, Manuela tipped off our latest success story on a massive network monitoring project achieved at T-Mobile. Check it out to see how Nagios is supporting your international data roaming on GSM when you travel.
To top off the Nagios theme, Michael introduced his NSClient++ blog series. He began with module groups: checkDisk (checks disk utilization), checkEventLog (checks entries in the eventlog), CheckSystem (checks processes, services, CPU utilization) and CheckHelpers (offers more ways of running checks). Each module is broken down into various sub-modules which carry out queries, and the parameterization of individual checks is similar. As he moves on with his series, Michael should cover installation, configuration and the running of several checks, which are communicated exclusively via encrypted NRPE to the monitored system. Keep posted!

Weekly Snap: OSMC and OSMC plus a project or two

weekly snapOct 26-31 was an intense week of blogging on the go at the OSMC. Only a couple strayed from the conference craze, such as Peter’s update on a Request Tracker project for the European Patent Office which will moved to our data center, boosted with NetApp SAN storage capacity and maintained by our managed services.
All else was OSMC. Karo kicked off the OSMC ticker on the eve of the conference, with a few impressions of the Linux live streaming and Thomas Krenn hardware setup.
Soon later, Bernd and Julian came in early with the first two 8am presentations: ‘Development of SNMP Plugins with Net::SNMP and Nagios::Plugin’ by Martin Hefter and Kristian Koehntopp’s Monitoring MySQL. Martin gave an introduction into Perl based development and use of available modules and libraries, offering Nagios/Icinga plugin developers a good guide to creating tools and extensions for existing implementations. Kristian covered areas of incident monitoring, problem detection, capacity monitoring, load and bottleneck forecasting and audit monitoring. He offered methods of attaining the required data, the available tools and their implementation with existing monitoring solutions.
Julian continued his coverage of the first OSMC day with Eric Pfaller’s presentation on an open source monitoring implementation at Audi, and a related presentation on the new LConf tool for the administration of monitoring configurations with LDAP. Here, our consultant Michael Steb took the reins with a live how-to on its integration into an existing monitoring system and administration in an LDAP front end such as Eclipse SDK.
Birger reported on a string of presentations, starting with PNP4Nagios Version 0.6 by Jeorg Linge. Jeorg announced a host of new innovations such as dashboard baskets, relative time ranges, new plugin templates and ‘dynamic series’ alongside multilingualism and a PHP 5.2 dependency. Birger then moved onto the Nagios Plugins presentation by Thomas Guyot-Sionnest who signalled new plugin projects and tools on the move, and ended the day with SNMPTT front end – NagTrap presentation by Michael Lubben. He noted Michael’s plans for future versions to include live views, database clean up, authorisation, hostname alias function and automatic status change for alerted traps.
Reporting on the second conference day, Bernd highlighted Ton Voon’s OpsView as a Nagios based tool to watch for extended coverage, web configuration and various distributed environment scenarios. Shortly after, Julian ended the presentation round with Bernd’s on ESB applications to distributed monitoring environments. Bernd offered Mule as an example of centralised transportation, router and transformation of check results in large heterogeneous monitoring systems.
Upon OSMC close we said our farewells and thankyous, and then hopped back into business with the project of October– a speedy, well coordinated Nagios expansion at BuW.
For impressions of our big yearly OSMC from the outside, a few reflections have made it to blogs from speakers and attendees that are worth checking out. So far:
http://www.adventuresinoss.com/?p=1189
http://blog.koehntopp.de/archives/2660-Ein-paar-Gedanken-zu-Zeitreihendaten.html
http://blog.koehntopp.de/archives/2659-Monitoring-MySQL.html
http://oerks.de/blog/2009/10/30/

Project of the month: Nagios to Oracle, SAP and AIX quick

buw_logoOctober 2009: When Bormann und Wulf from BuW first looked into Nagios they were looking for a quick and easy solution, out of the box. Upon meeting NETWAYS however, they decided against an appliance which promised plug and play monitoring for flexibility and independence in customization. After a streamline consulting visit from William to assist in system conception, the team at BuW were on their own till just this month.
They came this time to NETWAYS with a standard request – extend Nagios and quickly please. And this is what they got. Birger’s visit was short and smooth with all bases covered. From monitoring their Oracle and SAP clusters to AIX and installing EventDB and SNMP Trap Translator, all were integrated into a monitoring solution without a glitch.
The record speed project however, was only possible thanks to the fantastic coordination of the team at BuW. Upon Birger’s entrance to their Nagios scene, BuW already had good Debian foundations, an installed Oracle client, access rights and users set, as well as checks in CCMS for SAP fixed. With both sides well prepared, the entire project was achieved as fast as if it came straight out of the box.

Weekly snap: Jasper how-to, Startup, Nagios Voice Alerts & Workshop

weekly snap
Sep 28-Oct 2 was reporting week- on Startup successes, customer Nagios projects, tech workshops and of course more Jasper reporting tips.
To start, Peter shed light on our Nagios Voice Alert system recently implemented for the City of Augsburg. Per text to voice on a Starface appliance integrated by a NETWAYS module and managed by NoMa, the system calls and speaks alert details to the user, sending acknowledgements at the press of a button.
Another project was also brought into the limelight with the project of the month. For August, it was a large configuration restructure with template integration for a Nagios project at Schüco.
While Karo reported back on the success of the  Startup Weekend we supported, Bernd continued his reporting series with two Jasper how-tos. One on displaying current host status with a red/green icon, and another on linking in Jasper charts and content.
To top off the week, Bernd also reported on the important issues of our internal autumn tech workshop– from developments in NETWAYSGrapherV2, voice alert systems and LConf to check_interface_table and Open Source ESB in distributed environments. After such an agenda, we hope our techies came out loaded with new ideas to take on the upcoming projects.

Project of the month: Nagios templates for configuration clarity

September 2009: When Schüco approached NETWAYS, they had a basic, sluggish Nagios installation they wanted to expand and optimise. With over 46 locations around the globe and 650 devices to monitor, Schüco was looking for support in the coming configuration chaos. Inadvertently, config details were replicated many times over, sitting in large unstructured files so that Schüco admins simply had too much information and no overview.
The first task for consultant William was to restructure their config files into a tree and added template functionality. This suited their multiuser admin needs and simplifying their file structure in turn simplified configuration work. Instead of changing a config variable over 40 times, with templates they now only needed to do it twice.
On this new tidy configuration base, he could then expand the monitoring system’s capabilities to include the usual suspects: EventDB, PNP4Nagios, Nagvis, NETWAYSPortal, SMS alerts, business processes and even Jasper reports. As William customised the NETWAYSPortal for Schüco, he enhanced the SLA view to make target SLAs easier to compare with ‘actual availability’. After a good 20 days work on and offsite, their new state of the art monitoring system was up and running at optimum efficiency. Kudos to that.

sla_screenshot

Enhanced SLA View for easy target vs. actual availability comparison

Weekly snap: Training in Nagios, Reporting in Jasper & Free Streaming in OSMC

weekly snapAugust 31 – 4 Sep was a week dominated by reporting and events. Bernd E kicked off his new series on Jasper Reporting and Christian F gave us a few calendar highlights to look out for. On the side we had a look at the project of August – a Configurator for Nagios when there are simply too many servers and checks to do it by hand.
To open his new series, Bernd mused about the growing importance of professional reporting systems, where a spontaneously compiled Excel sheet just doesn’t hit the mark anymore. With a promised 10 posts to follow, he plans to cover everything from the installation of the server component, to the production of reports with dynamic parameters and the automated scheduling and sending of reports. Within the week he popped off his first entry on how to install JasperServer, child’s play as he says.
Christian F announced a few dates worth marking in the calendar for Nagios enthusiasts, with 6 courses on Nagios to come over 2009 and 2010. Around the corner, on 28 Oct 2009 at 9am he tipped off a once off, free live streaming of the OSMC presentation from MySQL guru Kristian Köhntopp on Monitoring MySQL – a great sneak peak into our legendary OSMC.