Hi all, PRTG has a number of http advanced sensors deployed to monitor a number of different websites. The sensor is set up to monitor these sites every 60 seconds. When the sensor fails five times in a row, a notification email gets sent out to a number of IT personnel. This process works great.

I demonstrated this monitor process to our company's business units and they say they like it. Then they ask, "Can you create a report that shows us when the sensor fails? We need a report that shows the times and durations that http sensor fails." There could be times when the sensor fails two or three times in a row and recovers before a notification email is sent out which impacts the uptime percentage.

Therefore, the question is: Can a PRTG report be generated that will show the times that a specific http sensor fails and then recovers?

Article Comments

Dear jbush

This kind of report is not directly possible with PRTG. What you can do is:

  • Get a table which contains each scan. Failed scans result in an empty row. You can extract that data also via the PRTG API and then compute the outage report yourself via a custom script.
  • A more consolidated report is possible with an API call. It contains all sensor up/down changes. If the sensor ID is 2134, the call looks like 
/tablewithstyles.htm?id=2123&content=messages&columns=datetime,sensor,status,message&sortby=date&filter_status=608&filter_status=607&filter_status=620&filter_drel=30days

This is still not exactly what you are looking for, but the closest thing you can extract from PRTG directly. To get a CSV instead, replace /tablewithstyles.htm with /api/table.csv. You can also get an XML, using /api/table.xml instead.


May, 2017 - Permalink