A modern day 500 mile email.
Consistently great posts from this site. Would love to know what the root cause is.
Years of low-level sysadminning on ‘nix systems going back two decades, steeped in a delightful burnout sauce?
Oh, you mean root causes of the delay thing?
I would think it’s in VMWare Fusion’s network virtualization stack, or maybe that’s obvious?
I’m just curious about the nitty gritty. What packet never gets acked or is dropped? Does vmware mangle the packet?
Nice bit of troubleshooting! Also I didn’t know about being able to use ‘tc’ to change global network tuning params on a machine.
Wait. Run this incomprehensible command to make IPv6 work? Shouldn’t IPv6 work without this? Tell me again why I want switch from a protocol that just works to this.
With respect you are way, way off the mark. The author is reporting a particular flaw in VMWare Fusion’s ipv6 support. Shall we rant and roar about how awful ipv4 is because the Magic Cap didn’t support full subnet routing?
IPV6 is rock solid. My entire home LAN uses it, every mobile phone that comes into my house and connects to my wifi uses it, and I regularly use it to connect to ipv6 only sites.
Because this can be fixed, but the address space limitation of ipv4 won’t be.
As for using ipv6 in vmware right now? Yeah, I’ll give it a pass if I don’t have to.