Hi,
Could someone help us understand the 'Instrumentor Response Time' measure in dynatrace self monitoring profile. Does this reflect the latency between the agents and collectors.
Also is there any metric that measures the latency between dynatrace components (server, collector,agents)
-Sreerag
Answer by Andreas G. ·
Latency and slow bandwidth will just impact that data takes longer to be sent from the Agent via Collector to Server. So - you may see data longer in the buffers and with that the actual PurePaths will just be "delayed" and later ready for processing on the dT Server.
if you do not have proper buffers you "may" see some skipped events or purepaths. Thats all I can think of in the moment
Answer by Andreas G. ·
Hi Sreerag
This measure tells you the average time it takes for the collector to instrument a class it receives from the agent. It is not latency related but rather how long it takes to actually do the byte code instrumentation of each class tha the agent sends to the collector
As for latency measures: I am not aware of any latency measures that we provide between the dT Components
Hi Andi,
Our current dynatrace architecture involves firewalls in between the agents and collectors. not sure if that is causing any latency or not. If there is a significant latency then we may have to find a solution for that and it will be a big change so it need to be significant enough to push for a change.
Also if there is no measure for latency, if we assume that there is latency, what kind of an impact will it have in the dynatrace system? May be we can relate it to some symptoms or after effects which may be caused by latency.
-Sreerag
JANUARY 15, 3:00 PM GMT / 10:00 AM ET