dynaTrace is deployed on RedHat Enterprise Linux.
Sometimes dynaTrace reports "Host Network Unhealthy" incident.
According to the following documents, it will report "Host Network Unhealthy" incident if the
Network Rate In or Out ratio exceeds 90%.
But all other monitoring applications do not find anything wrong with the network utilization issue.
So I would like to know how does dynaTrace determine the capacity of the network interface and understand why does it only report the network utilization issue from dynaTrace.
I think the answer is in that doc page. We query current throughput and total bandwidth available through the OS Native Libraries -> basically what you can also get if you open up your linux monitoring tools. If the Network Utilization (bytes sent&received) reaches 90% of the total theoretical capacity of that NIC we raise that Incident
I hope this helps
Thank you very much for your reply! What you have explained does help a lot.
As for my customer's original issue, I checked the /var/log/sa/sar* files and confirmed the network traffic was very high when the dynaTrace incident was reported. So dynaTrace did the correct thing.
Hello Andi, sorry do dig up an old thread, however I'm kind of having a similar question. If there is a link aggregation on Solaris system. How would dynatrace recognize the capacity of network interface? It would be very helpful to know which parameter is dynatrace looking at. Many thanks, Yifu