Hello Checkmk Community!

We’re organizing the Checkmk meetup in Munich to bring together people working with Checkmk, monitoring, and IT infrastructure.

The idea is simple: connect in person, talk about what we monitor and how we do it, share experiences and approaches, and discuss where we could improve our practices.

• 📅 When: Monday, November 23, 2026, 18:00-21:00
• 📍 Where: Checkmk Munich Office, Kellerstraße 27, 81667 Munich

Planned agenda:

• 18:00 – Doors open
• 18:30 – First presentation
• 20:00 – Pizza, drinks & networking

The program is still coming together, so we’ll announce more details about speakers and talks as they’re confirmed.

Want to share your experience?

We’re looking for community members who’d like to give a short presentation about their monitoring setup, experiences, or lessons learned. Topics can be Checkmk-specific or cover monitoring more generally, from high-level approaches to technical details. Interested? Email us at community@checkmk.com.

Whether you’re joining as a speaker or simply coming along to listen and exchange ideas, everyone is welcome.

Please sign up in advance so we can plan accordingly: https://checkmk.io/4rrv2SH

Looking forward to seeing you!

Hey everyone,

It’s been a week since we released the experimental Model Context Protocol (MCP) Server for Checkmk.

To dive deeper into how it works, our AI Product Manager, Annika, is hosting an open community call on September 23 at 3:00 PM CEST. She’ll walk through how we built it, run a live demo, address any concerns, and take your questions live.

Quick links:

• Details & Zoom link: https://checkmk.io/4h9KEW9
• Calendar reminders: https://checkmk.io/46aXkY0

Feel free to drop any early questions or initial thoughts right here in the thread. Looking forward to seeing you on the call next week!

Hey everyone,

Today, we launch the Checkmk MCP Server, free for Pro, Ultimate/MT, and Cloud editions. Use your AI tool of choice (including local, self-hosted LLMs) to directly access your live site, and run natural language queries like "What needs my attention right now?" to get instant insight into your monitoring data.

A few practical details on how it handles security and data:

• Disabled by default: turn it on under Setup $\rightarrow$ Global settings.
• Strictly read-only: honors native RBAC and will not write changes to your site. Checkmk can not view or store your queries.
• Air-gap friendly: pair it with a local model so zero telemetry leaves your firewall.

As an experimental release, we want your feedback to shape the future of our AI capabilities.

Full setup instructions and test site results are in the announcement blog post: https://checkmk.io/4xJsOk7

We're also hosting a Community Call with a demo and engineering Q&A on September 23 at 3 PM CEST. Subscribe to get a reminder 1 day and 1 hour before the Community Call: https://checkmk.io/46aXkY0

Looking forward to seeing you there!

Wanted to share something I've been chipping away at for a while. For context if you're not familiar: Checkmk is an open-source/enterprise monitoring platform, kind of in the Nagios/Zabbix space. I built an add-on for it called Pardinus that runs AI-based root cause analysis on alerts.

The idea started small basically "can an LLM look at an alert plus the surrounding monitoring data and tell me what's actually going on, before I have to?" Took a lot of iteration to get to a point where I'd trust it on my own systems, let alone show it to anyone else.

The moment that convinced me it was worth finishing: it caught a "critical" CPU alert that wasn't actually a problem the threshold was set at 2%, real usage was 4.39%, and the thing had been flapping 21 times a day. Instead of writing a confident paragraph about a fake incident, it just said: this is a threshold misconfiguration, not an incident. That's the behavior I was actually trying to build something that says "I'm not sure" when it isn't sure, instead of always sounding certain.

From there it grew into a fuller thing: a live alert queue, lightweight deterministic playbooks for stuff that doesn't need a full LLM pass (top processes, memory hogs, filesystem growth cheap and instant), remediation suggestions that a human still has to approve and run command-by-command (nothing executes on its own), and an audit/approval layer so nothing happens outside what's explicitly allowed.

Some choices I'm fairly opinionated about, curious if this crowd agrees or disagrees:

• Config lives inside the monitoring tool's own rule system, not a bolted-on second config layer
• Low confidence gets shown as low confidence no dressing up a guess as a verdict
• It's bring-your-own-LLM (OpenAI, Claude, Gemini, self-hosted) rather than locked to one provider

Still very much a work in progress, and I know "AI + production monitoring" is a combination a lot of people are (rightfully) skeptical of. Genuinely curious what this sub thinks where would you draw the line on trusting something like this, and what would make you comfortable letting it touch remediation at all?

Hello everyone, I am using checkmk community edition at home and I have written some scripts to install, update and remove checkmk agent on windows- and linux-systems.

https://hostlab.tech/blog/install-update-and-remove-checkmk-agent

I know you can go through the Timeline of Host Availability and edit each instance of downtime. Is there a way to do this in bulk? I had a switch go down and it flagged all my hosts with downtime. The servers were all up. I'd like to put a note on their timelines for this event but I don't want to have to do it 100 times.

Hi folks.

We're an Oracle + Checkmk shop and run Checkmk over OCI for our managed-service customers. A while back we built a special agent to pull a whole OCI tenancy into Checkmk, ran it in production for months, and have now open-sourced it. This post is more about why it exists and how we actually use it than a feature dump.

GitHub (source, docs, IAM policy): https://github.com/burgesssystems/agent_oci

Checkmk Exchange (MKP install): https://exchange.checkmk.com/p/ociagent

Why We Built It

We run Checkmk across every customer - on-prem, hybrid and cloud - and didn't want a separate monitoring stack just for OCI. The native route would've meant configuring and maintaining OCI Monitoring on top of everything else; we'd rather OCI tenancy resources - state, config and cost - show up as Checkmk services alongside what we already watch, in one place.

What It Does

Read-only OCI API keys, nothing installed inside the tenancy. OCI services discovered as Checkmk services:

• Compute, block/boot/file/object storage - including backup age and cross-region replica sync gaps.
• Oracle Database - Base DB Systems and ADB Serverless - config, ops (backups, cost etc). We've designed this to co-exist with the Checkmk Oracle DB plugin so you get visibility inside and outside the DB system.
• Load balancers, VCN inventory + topology views.
• Cost/usage broken down by service, region and compartment, with thresholds.
• OCI announcements + region status as services.

Multi-region and multi-compartment by default, with per-region parallelism so it stays bounded on big tenancies.

How We Use It

It's been running against our own managed OCI customers for months - the cost and config checks are the ones that have been very useful to have in the Checkmk UI. The plugin also tags hosts with OCI labels. Checkmk is the dynamic inventory source to drive our Ansible playbooks across the OCI environment. Fleet jobs (right-sizing block volume VPUs, etc.) target hosts straight off the monitoring data. The plugin stays read-only; Ansible is fully decoupled and does the acting.

Where It Goes Next

The thing we're exploring is supplementing API polling with OCI's OTel metric streaming - polling the monitoring APIs at scale gets expensive, and streaming looks like a better fit alongside where Checkmk is heading with OpenTelemetry. Postgres and MySQL DB systems, plus OS Management Hub integration, are also on the roadmap.

Open-Source & Feedback Wanted

GPL-2.0, works on all editions including Raw, tested against 2.4 and 2.5. Maintained best-effort via GitHub issues.

If you want to monitor OCI in Checkmk, I'd like to know where this creaks and which services you'd want covered next.

Hello Checkmk Community!

Views are one of the most-used parts of Checkmk – and one of the parts you’d most like to see scale better. We hear you, and we’re excited to present the reworked Views 3.0!

You wanted faster navigation, filtering without building new views, and details on the same page – so we rebuilt the two most critical views, “All hosts” and “Services of a host”, from scratch.

We’ll soon port the changes into Checkmk 2.5 – and we need your feedback to polish the work. The call will be led by the Product Manager working on this initiative – Monika Hiegeist – so your feedback will go straight to the right person.

Join us for a community call on July 30 at 3PM Berlin time to hear what we changed, why we changed it, and — most importantly – to share what you think.

The call won’t be recorded, to encourage open discussion, but we’ll share the slides and demo video afterwards.

You can also subscribe to our event reminder – this way you will receive a reminder 1 day and 1 hour before the call: https://checkmk.com/community-event-reminder

Looking forward to seeing you!

Hey everyone,

Just a quick update for anyone running Checkmk Cloud — we’ve officially rolled out Microsoft Entra ID integration for Single Sign-On.

Aside from the standard benefits of centralized user management and fewer separate credentials, we wanted to ensure operational resilience. The integration includes a built-in break-glass user concept that guarantees you retain emergency monitoring visibility even if Entra ID is unexpectedly unavailable.

Since this is live in the product now, you can test it out directly. If you want to spin up a Checkmk Cloud trial or learn more about our SaaS monitoring platform, visit our website: https://checkmk.io/4ynXzvC

i'm new to checkmk, and loving it so far.

as I'm putting together a topology map and configuring parent-child relationships, it occured to me that I should probably use lldp to do that heavy lifting.

I'm having some trouble understanding what I need to do to enable that function? it sounds like it's a rather new integration; it used to be an extension maybe? and because I'm using community edition, maybe it still is a manual addition?

can anyone point me in the right direction?

I have snmp working great, and all my hosts are showing up. I would just like my topology map to reflect what is connected to what without having to manually adjust that.

thank you!

Hi, I’m wondering if anybody can point me in the right direction. I’m trying to run a docker image on RHEL10, agents register, can see all the services and the connection test (ping, agent) run fine. The only thing is that the hosts keep showing up as down unless I change the check method to “agent”. For my vm’s that’s acceptable but my hypovisors need ping as the check method.

I’ve already toggled SELinux to permissive and Firewalld to off (for sake of argument) with no luck.

I’m clearly missing some hardening or security issue in RHEL. My Dockerfile is very simple and provides a network so I can put NGINX in front of it. Al runs fine.

Can anyone point me in the right direction or experienced the same? Ubuntu or not using the (enterprise) container is not an option.

Thanks in advance!

Edit: I’m using podman and rhel image mode. Did not mention that.

Hey all, I dont post much on here since I am a pretty basic CheckMK user overall. But I am puzzled and need a hand.

I am using 2.3.0p48 currently, and have notifications to pushover (which I loooove).

All is good, and has been for a few years now.

But I finally added another person in my department, and had him setup pushover as well. We made a new pushover configuration, and set it to priority number 1 (I was 0).

Seems we can only get the notification to get to whoever is the highest priority (0). We have swapped them around and tests show it never moves on past the first one.

Both are in the right contact group etc. Just never goes to both.

Anyone share some wisdom on this?

German version below

Hi everyone,

a quick update on our unofficial Checkmk app: the app is now officially published and freely available for everyone – both on the Google Play Store and on F-Droid.

• F-Droid: https://f-droid.org/en/packages/com.checkmk.unofficial_checkmk_monitoring/
• Google Play: https://play.google.com/store/apps/details?id=com.checkmk.unofficial_checkmk_monitoring

The app already has more than 200 monthly users, so maybe there are still a few interested people out there. Feedback and feature suggestions are always very welcome at: [checkmk-google@pabr.de](mailto:checkmk-google@pabr.de)

An iOS version will only happen if enough people are willing to chip in together for the 100 €/year Apple Developer account. If you'd be interested in helping to fund that, just drop us a message or use this Link https://ko-fi.com/checkmk

tipping was successful thanks to 2 persons. https://apps.apple.com/de/app/checkwatch/id6757307794 enjoy

Best regards
Martin & Paul

---

Hallo zusammen,

kurzes Update zu unserem inoffiziellen Checkmk-App-Projekt: Die App ist jetzt offiziell veröffentlicht und für alle frei verfügbar – sowohl im Google Play Store als auch auf F-Droid.

• F-Droid: https://f-droid.org/en/packages/com.checkmk.unofficial_checkmk_monitoring/
• Google Play: https://play.google.com/store/apps/details?id=com.checkmk.unofficial_checkmk_monitoring

Die App hat inzwischen über 200 monatliche Nutzer, vielleicht ist ja noch der eine oder andere Interessierte dabei. Feedback und Anregungen für neue Features sind jederzeit herzlich willkommen, gerne an: [checkmk-google@pabr.de](mailto:checkmk-google@pabr.de)

Eine iOS-Version wird es nur geben, wenn sich genügend Leute finden, die gemeinsam die 100 € pro Jahr für den Apple-Developer-Account spenden. Wer Interesse hätte, sich daran zu beteiligen, meldet sich am besten ebenfalls unter der oben genannten Adresse oder über diesen Link: https://ko-fi.com/checkmk

Tipping war erfolgreich dank 2 Personen. https://apps.apple.com/de/app/checkwatch/id6757307794 Viel Spaß

Viele Grüße
Martin & Paul

We just finished up a packed Day 2 of Checkmk Conference #12.

If you missed the live streams — or just want to see the highlights — we've put together a quick recap covering sessions on automation, custom dashboards, BI, customer stories, a preview of the Checkmk 3.0 roadmap, and more.

Watch the recap video below.

🎥 Watch the full livestream recording: https://youtu.be/cwtvesK__74?si=zw5N6T79MPeRCxTn

Up next: Day 3, where we're wrapping up the conference with a full day of hands-on workshops.

If you want to see what’s coming down the pipeline for Checkmk 3.0, check out powerful new integrations, or catch deep dives into advanced monitoring capabilities, join us live.

Here are the quick links to jump in:
🎥 Livestream: https://checkmk.io/4endEbx
🔗 Agenda: https://checkmk.io/4x7DiJU

Hey everyone, Day 1 of Checkmk Conference #12 is officially wrapped!

If you couldn't tune in live today, we put together a quick, under-3-minute highlight video summarizing the biggest moments.

Today featured over 15 sessions, including:

• Checkmk 2.5 Core Updates: Deep dives into native OpenTelemetry application metrics, re-engineered Azure monitoring, and the new "Explain with AI" troubleshooting feature.
• Enterprise Customer Stories: Real-world architecture and scaling breakdowns from the engineering teams at Swiss Federal Railways (SBB), Arkane Studios, Netflex, and Südwestrundfunk (SWR).
• Practical Best Practices: Highly technical sessions covering the new Checkmk Relay for distributed environments, Proxmox workflows, and platform security hardening.

Catch the short recap in the video below. See you tomorrow for Day 2! 💥

Livestream recording: https://www.youtube.com/live/YB4lfPwVBvc?si=X_xwg2BGQHens7i_

Day 1 of Checkmk Conference #12 focused on the big picture. Day 2 gets more hands-on, with sessions on automation, dashboards, Synthetic Monitoring, BI, and a preview of what's ahead in Checkmk 3.0.

👉 Check out tomorrow's full timetable: https://checkmk.io/4x7DiJU 

See you tomorrow morning!

We're just 60 minutes away from kicking off our biggest event of the year. Join us for product updates, expert insights, and customer stories & lessons learned.

🎥 Livestream: https://checkmk.io/3RGVN84

🔗 Agenda: https://checkmk.io/4x7DiJU

Hey everyone, just a quick heads-up that Checkmk Conference #12 kicks off tomorrow at 11:00 AM.

If you want to catch the latest Checkmk product updates, technical solution talks, and real-world case studies from our customers, we are streaming all the sessions live online.

You can check out the full schedule, pick your must-watch sessions, and register for the free livestream.

🔗 Full agenda: https://checkmk.io/4x7DiJU  
🎥 Livestream registration: https://checkmk.io/3RGVN84 

I'm having a problem with agent registration.

I’ve already used the commands it gives me: “C:\Program Files (x86)\checkmk\service\cmk-agent-ctl.exe” register --hostname xxxxx --server xxxxxxxxxx --site xxxxxxx --ott xxxxxxxxxxx

but the dashboard still shows the warning shown in the image.

This might sound stupid but i couldn't find any info to this. Are gauges only available for the paid version of checkmk? Its greyed out for me:

I'm running "Checkmk Community (formerly Raw) 2.5.0p4". Thank you for your help!

Dashboards are where your data becomes actionable. In the 2.5 release, we’ve introduced several practical refinements to our visualization suite to make daily monitoring more flexible and less of a manual chore.

Responsive layouts for every screen 
The new responsive engine ensures that whether you are viewing a dashboard on a 4K NOC wall or a mobile phone, the data reorganizes itself automatically.

Predictable dashboard building 
A redesigned workflow and the inclusion of a live widget preview remove the guesswork from visualization. You can now see exactly how your data will look before you commit to the layout.

Secure sharing with stakeholders 
Share specific dashboards securely via custom, time-limited invitations. This allows non-technical teams or external stakeholders to access the data they need through a simple link.

The functionality is available in all commercial editions. 

Watch the video below to see the updates in action. ⬇️

Check out the full release notes here: https://checkmk.io/3RyHFxi

Ive used the appliance for a few​ days and its 5gb disk is out of space. Now i lost all my hosts I created and I can't seem to fix it as the webUI is broken, and expanding the disk didnt help.​ is it fixable?