Today , I had an interesting issue with the TCP UDP stats and Retrans problem with one of the Active MQ brokers.
The context of the problem is : Whenever external monitoring tools to find out the status of brokers , they are returning with the no data as a value for ceratin data points. The data point is Retrans Percent and it is 23% which is super high.
Still investigating the issue of high in Retrans percent but in the meantime what is causing this issue ?
TCP retransmissions are usually due to network congestion. Look for a large number of broadcast packets at the time the issue occurs. If the percentage of broadcast traffic in your capture is above about 3% of the total traffic captured, then you definitely have congestion. Look for both physical layer (ARP) and network layer (name resolution) broadcasts on the network. If you find a high volume of broadcast traffic you can trace it to the source from the capture data.
The above paragraph make more sense in case of Network congestion so the solution is fixing the network traffic or making the right inetrval from your external monitor tool to track on the brokers.
No comments:
Post a Comment