How often do you monitor your applications and servers?
Monitoring Applications and Servers is critical for a company to sustain and serve its customers. Joseph added his favorite item on sale to the cart and suddenly the webpage refreshed and boom...., his item was not in the cart. He tries to add the item to the cart again and checkout, but this time he sees weird errors on the page. Debating whether to continue his battle to buy his favorite item now or stop and try in the evening again, he logs out from the website and in the next half hour, he receives an email saying that the sale has ended on that item. Frustrated, Joseph takes an oath that he will never shop on that website, and that results in the loss of one loyal and regular customer. This might have happened to hundreds, if not to thousands of customers at that moment.
What do you think has happened above? How often do you encounter these scenarios? What do you do when one of the servers goes down and you don't have proper monitoring in place?
There could be many reasons that might have contributed to the issue above. One reason could be the server might not have responded and a server reboot may have taken place. The other reason could be insufficient monitoring of the activity on the servers or applications which led the team to react to the issue at a later time. These issues may happen regularly if the underlying issues are not resolved on time. For this, you must have proper monitoring in place.
Example:
https://www.shashidiaries.com is the front-end URL serving the user base 24/7, 365 days a year. Let's say that the above URL is serving as a proxy deployed on Akamai and the reverse proxies are deployed on an external Apache server in a demilitarized zone of the firewall. These reverse proxies are talking to an internal F5 URL in the backend which is not visible to the world. The F5 is talking to an Internal Apache server which is routing the requests to an application server deployed on an OCP container. This OCP container processes the requests and performs the user actions by adding/deleting/modifying the related data to the database tables and returns the data to the user.
Here when Joseph added an item to the cart, all that data would sit in his buffer cache and then commit only when he submits the request successfully. When an error pops up, that error could be anywhere starting from the Browser -> Akamai -> External Firewall -> External Apache Server -> Internal Firewall -> Internal F5 URL -> Internal Apache server -> Firewall -> Application Server on OCP container -> Firewall -> Database server and vice versa.
If you can put monitoring starting from the customer front end URL all the way to the Database, may help to proactively resolve any issues and help your business or your customers not get impacted.
Dynatrace is one monitoring tool that can be used to monitor the URLs and servers actively and cut incidents when an issue occurs.
Conclusion
It is crucial to monitor your apps and servers to prevent damage to the company's reputation and finances. Investing in proven monitoring tools is one way to prevent these issues from happening in the future. Monitoring also helps a company from hackers and other malicious members of the cult society.
Comments