Unfortunately, it is currently not possible to suppress the "down" state of a sensor.

A Sensor Indicates a Device's Current Status

A sensor always indicates the current status of a device for the aspect it is monitoring. For example, if a printer is not reachable, a sensor on this device will change to a red "error" state to indicate the down status. You cannot change this, because otherwise the sensor would show incorrect information.

Ok,

Thanks for the reply.

Would be nice to have this feater though.. I understand that a sensor indicates the curent status. I just don't want them to show up in the alarmlist....

how do you want to be notified about a real printer issue then. e.g. someone unplugged the cable or the NIC is not working anymore?

If it does not show up as an error or if you ignore it you can as well not monitor the printer at all.

Sorry for the late response...

I don't want to monitor events. I just want to monitor the page count. PRTG is perfect for this.

I don't want to know when the printer is turned of when the user is leaving for home. If there is a real problem with printing the user wil know where to find me if he can't solve it himself.

we understand the issue, but one of the intentions of PRTG is that you know of any errors before the user does.

I have the same demand as Martijn.

I just want to monitor the toner level of some printers with a snmpcustomsensor. Everytime a user switches off the printer I receive an alert. I wish there could be a configuration in PRTG to turn off the alert-behaviour for some devices or sensors and just showing an unknown-state or something like this, but not an alert! Is there any possible solution in PRTG 9 to prevent the alert-state in the case mentioned above?

Flo

Works absolutely perfect!! Thank you very much!

I have a similar requirement but not the same. I am using PRTG to monitor cellular bandwidth usage on mobile devices that are on and off all the time.

The cellular provider has blocked ICMP on their network so i'm not able to use the workaround provided here.

Is there another way to achieve this? The only sensor I have to work with is SNMP traffic sensor.

There is no other workaround at the moment, sorry.

Torsten Lindner is perfect for me as well! Thanks!

PS: This "You need 15 reputation to up vote" rule is the stupidest thing I've ever seen. Now you have the wrong answer marked as the right one. Retarded.

Anny updates to this issue?

We are using PRTG to Monitor the performance of some "problem" CAD Workstations. Works fine, but of yource when the ping sensor goes down the helpdesk gets an error. (and so they are loosing focus for the important errors) An acknowledged error isn't a good solution because itLs also read and nearly no difference to a real error. I think for a monitoring system the solution to switch off alarms for some sensors/devices/groups should be a must!!

The solution still is to use the "Auto-Acknowledge"-Option on such Ping sensors. We really do believe there should not be a switch/option to disable error states. An error is an error. Maybe the Ping Sensor with only Warning would be an alternative.

Ok, i will try the ping-sensor with only warning, that sounds much better! Thanks!

In this case the ping error is not an error, because user CAD-workstations may be tunred off at any time from the user. But when running we still wan't to monitor them. And again, the autoacknowledged error sounds fine, but in the map there is nearly any difference between an normal error and an autoacknowledged error.

ok, I have now tried the ping sensor with warning. That would be a good solution, but setting the ping sensor with warning as master object for parent does not work, because the warning state doesn't take the sensor down. So our helpdesk now gets several red errors instead of one per workstation. Any solution for this?

Sorry no. A warning sensor cannot trigger dependencies.

Torstein Lindner wrote: > We really do believe there should not be a switch/option to disable error states. An error > is an error. Maybe the Ping Sensor with only Warning would be an alternative.

I have a similar issue with Process Monitor (WMI). I need to track the behaviour and resource usage or a specific running program (not a service) on a Windows server, and it goes into an error state whenever 0 instances are running. The process is a program that churns data into reports (IBM InfoSphere) and it is not supposed to be running when there is no data to process, and should never trigger any alarms or notifications (unless I set it up threshold triggers)

I have a WORKAROUND for this which might help other users: 1. Split your PRTG tree into a primary tree and a secondary "test/troubleshooting" tree or whatever you want to call it. 2. For the devices on which you want to have monitoring without triggering an alert, create a duplicate device with the same IP address and add the sensors that you are having trouble with to the secondary tree 3. Exclude this secondary tree from your maps and disable notifications on the devices. (Remember: You can still have graphs from the secondary tree on your map, even if you, for instance, have a big Sunburst on your map with the primary tree sensors)

Then you can place sensors which go DOWN often in the secondary tree.

My own big sunburst map is displayed on big screens here, and this method prevents me from being bugged by management when they see something RED on the wall screen :)

Im still looking for a solution for this... I have an WMI process monitor running for a process that is not always running, nor does it need to be... I need to log its performance, yet i get warnings every time the process does a normal shut down... ANy way to disable this warning?

Hello, while this is not possible with the inbuilt process sensor, you can achieve it with a script running as a Custom Exe Sensor. You can use the script of the following article as base: How can I monitor the memory usage of Windows Processes without alarms?

Hi all, in case someone still looking for a solution... I was searching for disabling the alert of a probe and came to this blog entry which finally brought me to the final idea with even different detailed options you have.

The key point of see / hiding the alerts of things you are not intersted in / responsible for is the root ID you start from. On the setup of the PRTG connection on the application you can enter the Root ID your view is starting from. So everything outside the subtree of your starting point is not considered, also no errors.

Option 1) You add an additional probe which monitors the time / case depending devices and you select the root ID on your main probe only. (In case you already have several probes, this will causing a connection for each of the other probes)

Option 2) You create a group element on top level of your probe and select the sibling node ID as your root ID on the configuration.

Finally a total different option could be to handle it with a user permission. So your standard users do not have the rights to see these sensors temporary off and you change to a second user / connection setting (on GUI application) to see those sensors.

Finally if all these above do not work. Run the query you want to do for a printer by SNMP or network bandwidth by WMI other protocols by script (Powershell, VB, ...) and handle the error in case of no response with "Sucess" return value. There you have it in your hands to outline the value just as warning and not as error. The device element you define above can be anything then which does not change to error in case it is not reachable.

Hope this helps. I will at least set it up now like this on my side to get rid of the error.

Finally, in case devices are powered off in predefined time frames or are powered on on fixed moments (like a printer always at core office hours) you can query the numbers / toner status once a day for 10 min by the "Schedule" in the settings of each Group, Device and Sensor (inherited).

I don't agree that the only option for a sensor should be to always provide "visual" alerts or notifications to the status of a sensor. I agree that PRTG needs to monitor the current status and record/log all activity.

For example, I want to monitor a service that I do not manage. What I care about is the long-term uptime and if the uptime SLA is real. i.e. my supplier is telling me the service will be available 99.9% of the time. To that end, I only want PRTG to monitor and log all the status changes, but I do not want to have any visual alerts, notifications, pop-ups, etc.

I simply want to go to that sensor once a month and look at the report to get the uptime, look for patterns, etc.

So, having an option to suppress all visual notifications of a sensor would be amazing. If PRTG could allow multi-level sensor alerting that would perhaps make sense. For example, Group A (or Primary) sensors are what we need to be alerted about all the time. Group B (or Secondary) sensors are what we want to monitor for reference only, and have no real-time alerting provided.

Is it possible to please request this as a feature?