This python script sends alertsite monitoring results to wavefront. This allows you to monitor availability and performance of the Alertsite checks in Wavefront (and combine alertsite data with other data monitored by wavefront).
Every 30 seconds it:
- Gets its configuration from config.json
- Connects to Alertsite report API which retrieves the latest status of all checks from all locations they are checked from:
- Parses the XML response for each check at each location. Fields:
- display_descrip - alertsite check name
- last_status - alertsite status (0 = success, etc Alertsite code documentation)
- Sets status to 1 if successful
- Sets status to 0 if not successful
- Ignores the entry if the last check indicates a problem at alertsite (vs a problem with our system)
- dt_last_status - timestamp for check
- resptime_last - seconds in response time for check
- Stores them in a dictionary, only storing the most recent one (as we don't want to send all data to wavefront, just the most recent)
- Loads the last set of results from a file stored from the last run
- Iterates through the dictionary and checks to see which ones are newer that what was sent to wavefront last time
- Sends each to the wavefront proxy : Wavefront data format.
- Format: alertsite.(alertsitecheckname).(status|seconds) metricvalue timestamp source=alertsite
- Saves the current dictionary in a file (to use in the next run)