question about Dyntrace vs application servers
Production ED. 5.0 - question –
on Monday (27/05 around 2AM SGT) the physical machine where dynaTrace server/collector app was residing crashed and went offline, consequently dynatrace server and collector service also went offline. Around the same time, we received reports that the http requests to website in application servers (where the agents are installed) were slow and intermittently timed out.
So the Customer question is:
If dynatrace server is having a problem and dynaTrace agents couldn't connect to it, will it cause the agents to keep trying to connect and crash the application server? In another word, if dynaTrace server is down, will our application servers be impacted in any way?
What do you think about this?
Thanks
GIanLuca
Hello Roman,
thanks for the clarification. I will share this with the customer.
Ragards, Ciao
GianLuca
Hi,
If the dynaTrace collector is down after the application has successfully started up you should not see any impact. The agent will periodically try to reconnect but until then the buffer will be overwritten with newer data until the connection can be re-established.
If the collector is unavailable during the startup then the agent will try to connnect to it for up to 60s (default), which means you will see a 1 minute delay in the startup times. This timeout can be changed by agent options (wait, DT_WAIT).
Best, Roman
