Few general advises required

Hello Guys,

I have a few questions to ask about Check_Mk, and I hope some experts will help me to understand it or to shed some light on these parts of CheckMk.

  1. Does CheckMk provides Acknowledgement and Comment feature? I have gone through the WATO but didn’t find these options anywhere there. I could see the Comment page when clicking on an event, but there is no button to add one or I may have look at a wrong place.

  2. When we add a host, and after the service discovery process it shows a lot of monitor services in which I only require a few (about 5 out of 30+). I have set the required 5 ones to monitored status, and activated the changes to make it visible on the panel. However, does the CheckMK check all those 30+ services at a regular interval or only the ones selected? It may cause performance issues if all those undecided services status fetch by the system.

  3. Is there any way to disable the periodic service discovery for all the hosts we would add? We only have to discover the services ones when we add the host, and no need to run the discovery process periodically.

Please advise on the above if possible.

Thanks, :slight_smile:

Yes, it does. There is a little hammer button to do that. There you can set acknowledgements/downtimes etc… Take a look at the screenshot.

Checkmk takes everything you give it. After that it inventorizes the data and only shows you the things you want, but in the background it checks “everything”. I never saw any performance issues caused by a checkmk agent. The agent is only a script, that you can execute as much as you want.Just take a look at it. If you monitor something with only SNMP then checkmk will be using a snmpwalk to gather the data.

Sure, but I would not recommend it because you want to know if something changed.
There is a rule for the periodic service discovery found in
rules -> Monitoring Configuration -> Inventory and Check_MK Settings -> Periodic Service Discovery
There you can find the default rule. Just edit or disable it and you are good to go.

I hope it helps you. If you have any more questions just let me know. :wink:

2 Likes

Thanks @Norm for your excellent explanation, and I found it all and made the necessary changes. I have one more doubt, can you advise me on this when you are free to keep me on track :slightly_smiling_face:

  • I have set the “Normal check interval for host/service checks” to 1 minute, and “Retry check interval for host/service checks” to 1 minute with “Maximum number of check attempts for host/service” to 3. Does this mean, I will only get the notification if a service/host status went warn/crit or wise-versa for 3 minute continuously? If not, is there any way to make it possible? Means, I will only have to generate the notification crit/warn/ok if the host/service stay on that status for 3 minutes. Else it may cause some false-positive due to check failures/latency during each checks.

Please advise me on this too, thanks :innocent:

Glad to hear that it helped you. :smile:

First things first. You are correct. :partying_face:
But let me explain it for you.

Checkmk is working with states. This is explained here: https://checkmk.com/cms_monitoring_basics.html#States%20and%20events

To simplify it:
There are soft and hard states.

For example:
a service changed it’s state from OK to WARN or CRIT and your check intervall is 1 minute and your maximum check attempts are 5. Therefore is the changed service in a soft state for the time your maximum check attempts has not reached 5 and will not trigger a notification.
But if the service stays in the changed state for more than 5 minutes (5 Check attempts), checkmk registers this and the service is put in a hard state.
Hard states can trigger a notification or can be used for filtering in dashboards etc.

You can also see this behaviour in the host & service notification tab.


As soon as the service changed it’s state into a hard state a notification is triggered. (TYPE = HARD)

You should be good to go with your current configuration. But please keep in mind that you have to or can configure this behaviour for service checks and for hosts checks. There are rules for booth.

The soft and hard states behaviour in notifications is also explained here: https://checkmk.com/cms_notifications.html#Repeated%20check%20attempts

Happy Monitoring :blush:

1 Like

@Norm thank you so much for the good explanation, how do you got to that page :smiley: ?

Thank you :blush:

Click on the desired hosts and use the “Host/Svc history” button.


There you can see the history of service and hosts notifications.

1 Like

Much appreciated your explanation @Norm

What is the usage of “Retry check interval for host/service checks” rule then? To trigger a notification after 5 consecutive checks (5 minutes), I only have to set “Normal check interval for host/service checks” to 1 minute, and “Maximum number of check attempts for host/service” to 5 right?

I have also set the retry check interval, so as per your description there is no need of this rule right?

Please correct me if I’m wrong on this :slight_smile:

Sorry for all the questions, I’m trying to familiarize with CheckMK to make it adapt for our usages.

I need one more assistance with the user privileges. I cloned "Normal Monitoring user’ roles and made some changes on for our staffs access. However, I would like to give them access to Host creation/modification permission. I set the privilege to yes (those related options I see when goes through the roles) but the privilege is not there when access to panel as that user. Also, add comments/acknowledge privileges are set to yes but still the Hammer icon is not clickable on the user panel. Only the admin (me) can do these tasks at this stage. Can you also advise the options to make these roles active for that user along with the above query?

Any help would be much appreciated! :innocent:

Did you give WATO permission?

Hi Daniel,

Yes, that is yes already:

So far (those looks to be related to comment/host in my review) are the following:

WATO - CHECK_MK’S WEB ADMINISTRATION TOOL

Use WATO
Host management
Modify existing hosts
Add & remove hosts
Rename existing hosts
Host Diagnostic

COMMANDS ON HOST AND SERVICES

Acknowledge
Add comments

The above are checked as yes already, I guess some other perms required which might have marked as no by me but not sure which are those.

Any help would be appreciated! :innocent:

I was able to edit hosts via WATO with the default normal monitoring user profile.

Make sure you also enable the “See all host and services” option under “General Permissions”

Hi @fasal,
the easiest solution is to enable the context help. You can do that in the upper right corner. (book icon image )
Then you find an explaination for the most things in checkmk :wink:

Is your user privilege problem solved? :slightly_smiling_face:

2 Likes

Hello @Norm @DanielNunesFerreira

Sorry for the delay caused, I was away for two days. This isn’t fixed yet, I’ve enabled all the necessary perms but the feature is not shows on the staff login panel.

Have you allowed WATO permissions ? You have to think of that as a tree, you can’t reach the leave if you don’t allow access to the branch.

Hope it helped. In the beginning I found more useful to give everything and then removing some permissions until its fine tuned