datadog monitoring for log files alerting

i have a log file which contains patterns . In that i need to grep only few patterns and sent those to datadog as an event in real time.

the approach taken to be is follows;

write a parsing function as described in this article :[url removed, login to view] : once your parsing function established, dogstream will collect the logs from the indicated file using the indicated parsing function every time the agent does a check, that is every 15 seconds.

Using this solution should not require you to set up a cron job, you just need the [url removed, login to view] to be populated and the parsing function to be written. Dogstream will then populate the UI from there.

Note that this function will need to be written in python

- write a custom Agent Check ( in python as well ) that will collect your logs from a designated file. You can grep any key word, but make sure you adapt the timestamp not to collect the metrics several times.

Indications on how to create a custom agent check can be found here :[url removed, login to view]

- use Dogstatsd using any library available here : [url removed, login to view]

The way you could leverage Dogstatsd would be to set up a loop that would get the metrics you want to gather from your logs, then the bucket would be sent every 15 seconds

A guide to dogstatsd is available here : [url removed, login to view] and here : [url removed, login to view]

and the log is : [url removed, login to view]

and the pattern sto be grepped are :

PEGA0028

PEGA0015

PEGA0037

LOG FILES IS UPLOADED

Python

ID Proyek: #10350158

Tentang proyek