Ergebnisse für " bernd "

SNMP Monitoring at scale

After a warm welcome by Bernd and the sponsor Eliatra the talks started. The first talk I was interested in was „SNMP Monitoring at scale“ by Rocco Pezzani from Würth Phoenix and Thomas Gelf from NETWAYS Professional Services. They teamed up to modernize SNMP monitoring and I was curious if the managed to do so. The talk started with anice summary of the problems you face in realtiy when using SNMP, and I think everyone using it practically known how complicated MIBs can be and how many of them are faulty. Their entry into the topic was to develop a MIB browser to be integrated into NetEye. One thing shown by this was the need of a new MIB parser, they plan for the future to relaxe parsing and compensate for errors as the number of faulty MIBs was too high. But they postponed this for the „funny“ part, a module to collect and inventarise SNMP data of an environment. They directly solved challenges other modules still have to solve like decentraled collection of data or other SNMP monitoring solutions face based on performance. In addition to active polling, they already created a first prototype for a trap reciever. Believing their explanations and shown error, I think they could really have managed the targeted modernisation, but decide by yourself as the first release will come soon.

Will ChatGPT take over my job?

The second talk I visited was the buzzword filled „Will ChatGPT take over my job?“ by Philipp Krenn. Of course, this was only his way to get people introduced to the topic. His main goal was to show the AI assitance integrated in the Elastic stack. While his demo was very nicely done and showed some great ideas, my personal feeling is that we are only at an early start, and it needs still needs improvement. This could be because it is a tech preview, or because of the exploratory mode of the AI used here. On the other hand, Philipp preferred this mode in contrast to the conservative mode which is used for the integration when it comes to security. But very important is still his warning that AI is just a tool you should learn to use, but not to trust at the current state. Based on the question of the audience, there was a big interest in the topic.

Icinga for Windows

Before the lunch break, Christian Stein gave his update on Icinga for Windows. He started with the things fixed and improved with the latest release and by giving a quick overview of Icinga for Windows. One big improvement was the introduction of the REST API, which limits concurrent checks and increases the execution performance while consume less resources. This requires the latest Icinga 2 version, as the required check command is integrated there for an easy integration for a good user experience. As a person interested in Security, I am also quite happy about his work on the JEA profiles. JEA stands for Just Enough Administration and is the Windows way of improving security. But this was only the current state, he also showed his visions for the future. The demo was about the improved management capabilities integrated in the console, allowing to remotely install on other servers. Other plans are a rewrite of some core features and to improve cluster monitoring, which will result in the next version being 2.0 next year. But there will be a 1.12 in between as preparation and for bug fixing.

The Ignite Talks

After everyone had his portion of the great lunch as desert, there were the Ignite talks. This format is perhaps well-known nowadays, but I always look up to those brave people who can do 20 slides with a fixed number of seconds each on automatic mode. In the first one „Honeypot Flavors: Open-Source Honeypots and their Use in the Automotive Industry“ Niclas Ilg gave a quick look inside Bosch, how they use Honeypots and how they participate in development. The second one was our Daniel Bodky with „Metrics, Margins, Mutiny – How to make your SREs (not) run away“ introducing the concept of Site Reliability Engineer to the audience. And last but not to least Felix Frank who always has the courage to tackle important but unpopular topics talked about „Your business isn’t Green enough„.

Impact assessment with Netbox Path

After those, I visited „Impact assessment with Netbox Path“ by Dave Kempe from the Australian Icinga partner Sol1. He shortly introduced NetBox, an Open Source CMDB, and the Icinga Business Process Module and what challenge they solve with it in a monitoring environment. NetBox Path is a plugin to manage dependencies between objects in the CMDB in a graphical way, but also via API. This plugin they utilized to close the gap by adding a notification plugin which shows the path to users to provide them context.

What’s new with Grafana Labs’s Open Source Observability stack

Really interested I was in „What’s new with Grafana Labs’s Open Source Observability stack“ by Sebastian Schubert from Grafana Labs. And it seemed that I am not the only one as the room was full and when asked if someone does not know Grafana anyone spoke up. He started with the improvements to Grafana which did not need any introduction. The second tool covered was Mimir, the metrics solution from Grafana Labs. After a short introduction, he showed the performance improvements and new features coming to in the last year. Next was the Log management solution Loki which works similar to Prometheus but for logs instead of metrics, again he introduced it shortly and show the improvements. He did the same for Tempo – the tracing solution, Beyla – another tracing solution which will be released in the near future, Faro – Browser or End User monitoring, and Pyroscope – the solution for profiling. I kept this summary quite short, as I plan to do another blogpost going into detail when the conference archive is online, so stay tuned.

Newest developments in Checkmk Raw – the open-source monitoring solution

It is always a good sign if you can not decide between talks as you want to be in both rooms, but as I had to decide I joined Lars Michelsen from Checkmk with „Newest developments in Checkmk Raw – the open-source monitoring solution“ because I have not had a look on it for a while. The focus of the talk was on the raw edition, which is the community edition. There is also a enterprise edition with performance improvements and convenience features, and a cloud edition for those who want the enterprise solution in a hosted version. The web interface was improved to be a more modernized version of the same design goal and looks quite nice, especially with the graphing solution switched to the one from the enterprise edition. Also, communication with the agent is now more secure and also security is active by default. API for the web and plugins were improved, too. And these were only the changes before the last version, in this release more UI improvements and improvements on graphing integration of Grafana are included with many more small things. Numbers and Tools developed by the community shown by Lars show the Checkmk ecosystem being wealthy and growing. The same goes for the numbers from behind the scene like growth of the company. A nice tool he showed was Robotmk which integrates the Robot Framework into Checkmk for End-to-End-monitoring.

Current state of Icinga

And as always Bernd finished with „Current state of Icinga“ the day. He started with a very short introduction, including some new and interesting success stories. Afterwards, he walked the audience through all the new module releases and the improvements which came with it. As this would be quite much, I will focus on the two new things.
Icinga Notifications will move the notification from the core to a new module, which will give you a much more granular control over notifications and escalations by incorporating schedules and giving all modules and in the future external solutions an endpoint to send notifications to. A usability test for it is going on during OSMC, a prerelease should happen later this year with the first release being planned for next year.
Icinga K8s Monitoring will replace all the current approaches to get Kubernetes to fit in Icinga’s data model with a much more fitting solution. This should be really helpful for users to get a much better look inside their environment. And it should be also available as a prerelease later in the year, with a final release next year.
He ended his talk with an outlook at upcoming events, just 15 minutes late, but right in the time for the evening event.

Dirk Götz

Principal Consultant

---

Dirk ist Red Hat Spezialist und arbeitet bei NETWAYS im Bereich Consulting für Icinga, Puppet, Ansible, Foreman und andere Systems-Management-Lösungen. Früher war er bei einem Träger der gesetzlichen Rentenversicherung als Senior Administrator beschäftigt und auch für die Ausbildung der Azubis verantwortlich wie nun bei NETWAYS.

Lies mehr von Dirk und triff unser Team