On Wed, Oct 24, 2012 at 11:54 PM, Ben Pfaff <b...@nicira.com> wrote: > On Wed, Oct 24, 2012 at 05:45:36PM +0800, Cong Wang wrote: >> Before waiting for the kernel to reject an invalid name, we >> can actually check it before going into the kernel. The code >> is stolen from linux kernel function dev_valid_name(), >> but it should apply to non-Linux arch as well, because >> IFNAMSIZ is POSIX and other errors are obvious. >> >> After this patch I got: >> >> # ovs-vsctl add-port ovsbr0 12345678901234567890 >> ovs-vsctl: cannot create a port named 12345678901234567890 because the name >> is not valid >> # ovs-vsctl add-br 12345678901234567890 >> ovs-vsctl: cannot create a bridge named 12345678901234567890 because the >> name is not valid > > I understand why this is an attractive patch, but it restricts what > ovs-vsctl can do to what Linux can handle. ovs-vsctl, and Open vSwitch, > are meant to be more portable than that. Different operating systems > have different limits on the maximum length and the allowed format of > port names. I don't have the ESX source code right here, for example, > but if I recall correctly the maximum length of a port name is much > longer in ESX, and less restricted, than in Linux.
Hi, Ben, As I mentioned above, I think it is not bad to apply the same restriction to other arch than Linux, so that people don't need to modify their openvswitch script for different arch. :) And, IFNAMSIZ is POSIX, don't know why ESX doesn't define its own. Thanks. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
