How can I run PRTG on a virtual box linux machine?

Hello,

Thank you for your message.

As of now, I'm afraid that PRTG (Core and Probe services) can be installed only on Windows devices. However, you can use the Linux sensors to monitor your devices.

Here is our dedicated page regarding Linux monitoring: https://www.paessler.com/linux_network_monitoring

Kind regards.

Hi,

Thanks for your reply.

I've one more question, Two different user can use same PRTG package, I mean if user-A has downloaded the package and using it, and also given to USER-B to use can USER-B access with same key? or each user have to download and generate license key and then install?

Hey,

When USER-A installs PRTG, he will by default use a trial key valid for 30 days which will allow him to use an unlimited number of sensors. (Limited to 100 sensors after the trial period)

Then, USER-B can have access to the PRTG server via the Web interface or the PRTG Desktop client, which is available for Windows, Linux and Mac.

Please, have a look to our page regarding PRTG Network Monitor for more information: https://www.paessler.com/prtg

You will also find many videos regarding PRTG on our Youtube channel right here: https://www.youtube.com/c/PRTGNetworkMonitorByPAESSLER/videos

Let me know if you have further questions.
Regards.

Thanks for your reply,

Can I run PRTG as a docker? IF yes, then what would be the recommended size or resources ? and running PRTG as docker is feasible or good idea?

Hello,

As of now, I'm afraid that PRTG only runs on Windows. Therefore, I can't provide you more information regarding Docker for now.

Have a nice day.

thanks, I was going through prtg internal and have few quesiton regarding it.

1) What is local probe ?? what is it mean monitoring the probe system? Do it means it will monitor PRTG internal components?

2) As remote probe is used to monitor machines on different network, can we install remote probe on linux machines also? or we do really need linux sensors?

Hello,

1) The local probe is installed along with the PRTG core server and is running on the same system. With the local probe you can monitor devices that are reachable from it, and all monitoring is done by it (until you install remote probes). Monitoring the probe system means monitoring the OS on which the probe is installed.
2) Remote probes can be installed on other systems in order to monitor their network, or to reduce the load on the local probe (to distribute the sensors). Currently you can install remote probes only on Windows systems.

For more information about probes, please see the following article.

Kind regards,
Sasa Ignjatovic, Tech Support Team

With Free trial, 100 sensors will be available.

can I get the list of all that 100 sensors?

As I was monitoring my linux machines, I couldn't only ping, disk and mem sensor. How can I get CPU usage ? Is there any sensor for it?

With the free trial license you are limited only by the number of sensors that you can use, but you can use any sensors that you want. So it does not matter which 100 sensors you add.

To monitor the CPU load of your Linux machine you can use the SNMP CPU Load sensor.

Kind regards,
Sasa Ignjatovic, Tech Support Team

Thanks for your reply,

One more query,

suppose i've 100linux machines running and I want to monitor all of them. So as in free trial we've 100 sensors available , so we can use 100 sensors for each machine or it's like if we use 99 sensors of machine 1 and then only 1 available for 99machines?

With the free license you have 100 sensors total, and you can distribute them between your devices however best suits you. For example you can monitor one device with 100 sensors, but that would leave you with 0 sensors for other devices. Or you can have 10 devices with each 10 sensors.

Kind regards,
Sasa Ignjatovic, Tech Support Team

But I can use same sensors for two different devices right? ex: ssh mem sensors I can use for all 100 devices

if yes, then it means 100 sensors I can use for different 100 devices.

or if we cannot use same sensors, then is there any way to achieve it?

Suppose i've added 5 linux machine and did auto discovery, it will show ssh meminfo in all of them . so it means i've already used 5 sensors? or it just will be one sensor??

Each sensor can monitor only one device, and each sensor is counted towards the 100 sensor limit. So for example if you have 100 devices and on each a Ping sensor, then you have used 100 sensors not 1. Also, if you create 1 device but add 100 Ping sensors to it (no reason to do this, but you still can), you are still using 100 sensors. PRTG also displays the sensor count in multiple locations so that you can check how many sensors are still available to you.

Kind regards,
Sasa Ignjatovic, Tech Support Team

Thanks for your reply,

Can we get the data/metrics from the rest api from remote machine and raise alert accordingly? For http sensor do we need XML format?

Hello,

The HTTP sensor just monitors if the web site specified in the given URL is reachable, so no XML is required here.

Can you please describe your first question more in detail (regarding the rest API)?

Kind regards,
Sasa Ignjatovic, Tech Support Team

Hi,

Yes sure,

Ex: suppose I've a Linux machine running and exposing some metrics using rest-api like curl:xx.xxx.xx.xx:port/metrics and I want to monitor those metrics and raise alert according to threshold values. so how can we achieve this scenario? and if yes, then suppose if metrics endpoints exposes some 20 metrics, do we have any provision to read particular metric?

We have the Rest Custom sensor that can read information from the API, once you have setup the sensor and the information is being displayed you can set limits in the channels to receive notifications when these are reached.

Kind regards,
Sasa Ignjatovic, Tech Support Team

Hi,

thanks for reply, I couldn't find any rest custom sensor except beta rest custom sensor. So I'm using Http advanced sensor, but couldn't find relevant information how to use that sensor

for example :- First I need to POST to get token for further methods like POST GET. Post information curl -X POST -d '{"username":"test", "password":"123"}' http://xxx.xx.xxx.xxx:<port>/login this will return token and then need to use for get information

Get information curl -X GET -H "Authorization":"Bearer eyJhbGciOiJIUzM4NCIsInR5cCI6IkpXVCJ9.eyJ1c2VybmFtZSI6ImFkbWluIiwidXNlcklEIjoiOTQ0YjcyOWEtOTQxMy00MmE4LWFkMTItNjY3ZWVkMGY1NmU4IiwiZXhwIjoxNjA0OTQyMDcyLCJpYXQiOjE2MDQ5Mzg0NzJ9.497wRxkH-G-JBe0PG6i8vqbW0OOAxZEeViNSlfbmE-z63Czt0aG0OB34hJ37jd5N" -H "Content-type: application/json" http://xxx.xx.xxx.xxx:<port>/version

How to can simulate above scenario and which the best sensor to use?

In HTTP advance sensor where to mention token?

POST AND GET data stores to some file , how to use that information for next method?

I can see data files created at this path ProgramData\Paessler\PRTG Network Monitor\Logs\sensors

and files are created, and I can see return information in one of the file i.e result of sensor 2113-A. ex {"access_token":"eyJhbGciOiJIUzM4NCIsInR5cCI6IkpXVCJ9.eyJ1c2VybmFtZSI6ImFkbWluIiwidXNlcklEIjoiYTMzNTAxMzctYjJlMi00MzhiLTg5YzMtOTJkNTA1NGVhMjg4IiwiZXhwIjoxNjA0OTQ1MTA1LCJpYXQiOjE2MDQ5NDE1MDV9.VlxF1tGJnuQ61zwRdmmNm6_svlXaKzRJQOBfrGdUWd5lznbuEJS09i9zRuyE1jkg","token_type":"Bearer"}}

But still seeing error in sensor details The returned JSON does not match the expected structure (Prtg is missing). (code: PE231) why this error is coming?

how can we use above token to for other method like get and post?

The Beta Rest Custom sensor is the "Rest Custom" sensor that I have mentioned. Regarding the file that you have found in ProgramData\Paessler\PRTG Network Monitor\Logs\sensors, that is a debug file that is used for troubleshooting, it is not meant to be further used in the sensor.

If I understand your use case correctly you want to perform one query to retrieve a token and then use the token in a second query which would give you the needed results. If that is the case, then this is not possible with our current native sensors. The only solution would be that you create a custom script which would perform the necessary tasks and at the end return the result. You could then run the script with a EXE/Script sensor or a EXE/Script Advanced sensor.

But still seeing error in sensor details The returned JSON does not match the expected structure (Prtg is missing). (code: PE231) why this error is coming?

while debugging i can see sensors director has the data file where i'm getting the data but why it is not showing in sensor details in prtg gui? why i'm seeing the above error?

The error means that the output that the sensor receives is not in the expected format. Please see our manual on Custom sensors for more information on how the return format should look like for each sensor.

Kind regards,
Sasa Ignjatovic, Tech Support Team

Hi,

I'm working on alerting with PRTG. I've configured sensor setting -> notification trigger based on threshold value for some particular channel. ex: When Response Time (msec) channel is Above 144 for at least 1 seconds, perform Email and push notification to admin When condition clears after a notification was triggered, perform no notification

1) I'm receiving alerts through mail but only one time. With my understanding above configuration tells if response time is above 144 for alteast 1 seconds , notification will be triggered right?? But i'm not receiving it every 1sec or any time configured. How this value is compared like every 1 seconds or every time sensor scan will run?

2) What is unit #? if i'm getting channel value in bytes using rest then by default it will take bytes or how it is?

3) And I want to configure telegram alerting also, how can we achieve this?

Hello,

Threshold notifications are only able to send notification twice :

• When the condition is fulfilled
• When the condition clears (not fulfilled anymore)

Please notice that you can't use a delay ("for at least x seconds") lower than the sensor's scanning interval, as mentioned in the manual.

Then, PRTG uses different units including "#" for display purposes. The unit can be easily changed in the Channel Settings.

Regarding Telegram notifications, please have a look to our blog post: https://blog.paessler.com/how-to-receive-prtg-notifications-by-telegram

Kind regards.

Hi Florian,

I've one query, Is PRTG is PULL or PUSH based?

As it is running probes and fetching information about remote machine , i.e. pull based approach right? Are PRTG, supports PUSH based mechanism? If yes, can you please provide some light on it.

And last time, I've added one query regarding PRTG notification issue, you replied with some support ticket, but there is no link , couldn't able to find it. can you please help?

Hello,

Most of the sensors in PRTG actively execute requests on the devices where they are configured. However, you also have the possibility to use passive sensors as examples :

• Syslog Receiver
• SNMP Trap
• HTTP Push sensors

Regarding notification, here is the link I provided for Telegram: https://blog.paessler.com/how-to-receive-prtg-notifications-by-telegram You can also change the channel units in the device/sensor settings, under "Channel Unit Configuration". Here is the corresponding manual: https://www.paessler.com/manuals/prtg/device_settings#channel

Regards.

Hi Florian,

Couldn't able to see my last question here, so got confused on which query you've posted!!.

Are you saying some of the sensors , supports push based mechanism? those who supports are called passive sensors? Most sensors in PRTG are pull based?

How push based sensors works? where it push data to?

Most of the sensors in PRTG actively execute requests on the devices where they are configured. However, you also have the possibility to use passive sensors as examples :

*Syslog Receiver
*SNMP Trap
*HTTP Push sensors

The active sensors execute requests directly on the monitored devices with different protocols including WMI, SNMP, HTTP, etc. However, the "passive" sensors listen for data and then display the received information in PRTG.

Please, have a look at the list of sensors included in PRTG for further information about them: https://www.paessler.com/manuals/prtg/available_sensor_types

Regards.

Hi Florian,

I'm using http Push advance sensor, I've gone through doc related to that. where I can see , http command is mentioned there, which ask for probe IP and port and token as well. Port and token we can get easily, but probe IP is 127.0.0.1 , that means we cannot connect or push data to sensor from remote machine right?

core and PRTG is running in windows and I'm trying to push data from some Baremetal/linux server. how can we achieve above scenario?? Please put some light .

Hi,

To send data to the HTTP Push Advanced sensor, you must the IPv4 address with which you can reach the PRTG server from the remote machine.

By default, the probe automatically select the appropriate interface to use (default option: Auto) and therefore should be reachable on the IPv4 of the PRTG server.

Regards.

Hi,

I didn't get you, are you saying to change PRTG server IP from 127.0.0.1 to reachable IP? If yes, how can i change that. or is there any other alternative?

Does your PRTG server has access to internet or any of your other devices? If that's the case then it has an IP address (other than the loopback address 127.0.0.1 which can' t be changed) that you can use to get access on the server.

To check the IP adresses available on this server, you can use the ipconfig command within a command prompt or a PowerShell console.

You are referring prtg server as my window machine?

Indeed, PRTG can only run on Windows.