That's why I was asking, because I thought that a static tunnel would be a risk (because of high cost) and in turn thought that a dynamic tunnel (using AYIYA) would be a better solution in general (for dynamic as well as static IPv4 addresses), but your explanation makes a lot of sense. Thank you!

Firstly, Thank You for 3 years of great service. I have had some minor glitches along the way but you have always been very helpful in those times. Jeroen Massar wrote: Jeroen, respectfully, while your statement makes perfect sense in a perfect world there are some real world issues that need to be considered. I have a fully static IPv4 network and for quite some time ran static tunnels. About a year ago however my ISP (AT&T) pushed out a firmware update to my router (Required, I cannot have my static addresses without their router) which included software that intercepeted all the ICMP packets and handled them internally. Hence my hosts on static IPv4 addresses were effectively cut off from the outside world of IPv6. I went round and round with AT&T but there was no way to downgrade the firmware or route the tunnel maintenance packets to the proper hosts. I was thus forced to change all my tunnels to dynamic so that the maintenance packets were tunneled past the router. To this date AT&T still has no native IPv6 service in my area but they have effectively killed my ability to maintain a static tunnel. I'm just pointing out that with all the parties involved in the migration to IPv6 there often exist situations that do not neatly fit into the best case scenario. Like AT&T rolling out their support in bite sized chuncks that effectively chop existing solutions such as your fantastic tunnel service off at the knees.