Don't throw the baby out with the bathwater! Synthetic monitoring (active monitoring) helps reduce key blind spots for critical applications. We just experienced a production issue on a fully instrumented critical business application that first appeared nebulous.
During peak volume time the Service Desk was taking calls from users across random locations stating that they couldn't login, however if they were already on the system all was well. Even when they logged out they could still login again and continue working.
Other facts that came in made the issue more perplexing:
- RUM showed transaction volume and performance was normal.
- Deep dive Java monitoring agents showed the same.
- There were no glaring HTTP 500 errors and the backend database was fine.
- Infrastructure monitoring was green in all tiers and resource consumption was within baseline.
What did we use to find the issue then? It was our synthetic monitoring tool that popped an alert on two externally facing applications.
Root Cause? Our internet provider’s DNS resolution was not working properly. So any machine that needed name resolution that wasn’t already cached for the day, couldn’t get a login page.
If an event occurs and no one sees it, believes it, or takes action on it, APM's value can be severely diminished and you run the risk of owning “shelfware.” Our experience has shown that active monitoring (synthetics) are a good complement when used with passive monitoring to help provide visibility on application availability especially when monitoring outside the Data Center.
You can contact Larry on LinkedIn.
If you have more questions on APM and want to connect with thought leaders in this technology space join the Application Performance Management (APM) Strategies Group on LinkedIn.
For more information on the critical success factors in APM adoption and how this centers around the End-User-Experience (EUE), read The Anatomy of APM and the corresponding blogs APM’s DNA – Event to Incident Flow.