Hi,
We have a custom performance counter setup that is giving us daily false alerts from it's speed trigger.
The sensor is set up to scan the performance counter every 30 seconds. We just want alerts if it sits at 0 for a certain amount of time (we have it set to 30 minutes at the moment). This indicates to us that the service it is attached to isn't working properly.
When we receive a notification that the speed trigger conditions have been met, we log in to the PRTG server and check the live data. From the live data we can see that the counter is not at 0, and isn't at 0#/s and hasn't been during the last 30 minutes. This happens daily but we haven't been able to determine the cause yet.
Sensor Settings
Notification Settings
Log
Historic Data
Article Comments
I've attempted to edit the post with correct links, but it fails.
Here are corrected links:
Sensor Settings
Notification Settings
Log
Historic Data
Aug, 2018 - Permalink
Dear elliott-pd,
thank you; however these screenshots are cut-outs. This removed important context. Please capture the full interface of PRTG in the screenshot. In addition, the "Overview" tab of the sensor would be helpful as well.
Aug, 2018 - Permalink
Hi Arne,
Here are full page screenshots as per your request. I've also attached direct links to the images, as they seem to be displaying weirdly.
For additional information about what we are trying to monitor. We have a service setup that processes incoming messages, each time a message is successfully processed we increment a performance counter by 1. Each morning we reset all of the performance counters to 0. This service does run 24 hours a day, but we want to monitor it during certain business hours (9am-8pm) with a 2 hour window for lunch which is ignored. Outside those hours the sensor is turned off, as there won't be any data for the service to process.
We expect a constant flow of messages, but it is a low amount. Only about 48K over the course of a business day. The average per second will be low, usually under 1 depending on how busy the service is. If there is a period of 10+ minutes where there has been 0/s, we want the sensor to go down so that we can investigate this. As it indicates the service is not processing any messages.
Log
Notification Settings
Overview
Settings
Channel Settings
Here are the direct image links, as once posted they seem to show weird dimensions (likely because the images are so large).
https://image.ibb.co/ir9KHp/Live_Scripts_Log.png https://image.ibb.co/hgwMV9/Live_Scripts_Notification_Settings.png https://image.ibb.co/cTrAOU/Live_Scripts_Overview.png https://image.ibb.co/n0cjiU/Live_Scripts_Settings.png https://image.ibb.co/cEyqOU/Live_Scripts_Channel_Settings.png
Aug, 2018 - Permalink
Dear elliott-pd,
thank you. Please also send the historic data for that time frame, so that I can check it. The latest trigger according to the screenshot was on August the 28th, 10:44:22.
Aug, 2018 - Permalink
Dear elliott-pd,
I saw that you are already in contact with one of my colleagues. Please follow-up that case, as it is easier via email than through the knowledge base.
Sep, 2018 - Permalink
Dear elliott-pd,
as the images are not visible, please contact us at support@paessler.com. Please include the description of the issue, and the screenshots as file attachment (up to 10 MB combined attachment size.)
Aug, 2018 - Permalink