I wrote a Python program that runs iperf3
in a for loop with different datagram sizes. I am using Python's subprocess.call
function to run iperf3
like this:
iperf3 -u -c 185.82.21.60 -b 100M -t 5 -l <datagram size>
My datagram size starts with 1000 and increases by 100 after each iteration and should run until it reaches a size of 2000. Every time I am running my program, iperf3 keeps hanging after the 6th time and starts using 100% of one of my CPUs. After this I cannot use iperf3 anymore until I kill the running client process and restart the iper3 server.
So, it turned out that my problem is caused by this bug. The solution seems to be to update the ESX hosts's drivers.