This problem has always existed for any connection type. It shows up in a lot of different locations on all Bering versions. I saw this on ppp connections as well as pcmcia based ethernet connections. The common denominator of all these is, that you cannot predict reliably how long they take to come up, but the init script may terminate _before_ they are up completely.
Agreed. Shorewall by default has really awful failure modes if the upstream ppp interface isn't up yet.
I'd love to have an "is up?" semaphore, but perhaps in some cases, we should instead be triggering the apps by the fact that the interface is up. Both /etc/network/interfaces and ppp have trigger scripts they can call for interface up. Then it comes down to what is "up?" -- link up? address configured and able to pass data? routing up?
I don't want to confuse things with those last questions, there probably is no universal good way to do these things. Frankly, I wish shorewall was just a little smarter when it came to ephemeral interfaces.
Paul ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ------------------------------------------------------------------------ leaf-user mailing list: leaf-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/leaf-user Support Request -- http://leaf-project.org/
