In version 4.18.0.0, it is common for virtual machines to be unable to access networks outside of the cloudstack, but internal communication is normal #11027

In version 4.18.0.0, it is common for virtual machines to be unable to access networks outside of the cloudstack, but internal communication is normal. Moreover, the timing of each network failure is uncertain, and the machines are also uncertain. Virtual machines in the entire cloudstack may appear, but only a few virtual machines will appear each time instead of all virtual machines. Generally, it takes about 10 minutes for the timeout to return to normal. Is this a unique fault of version 4.18.0.0? Or is it that there are some configurations in my VR that need to be adjusted, and I cannot find the logs in the management server. log.