Hello,
Actually, both 1.6-2 and 1.5-16 still disable IPv6 on the parent interface for
me, in a config like this:
iface br-test inet static
address 192.168.9.101
netmask 255.255.255.0
bridge-ports eth1.1008
After "ifup br-test", /proc/sys/net/ipv6/conf/eth1/disable_ipv6 is set to 1.
--
Hello,
I just hit this bug as well. From what I see this has been fixed in 1.5-16,
but Stretch currently contains 1.5-13+deb9u1. And 1.5-16 is not available in
any archive. The Buster version is 1.6-2. Is it possible to push the fixed
version to Stretch?
Thanks
--
With respect,
Roman
Package: bridge-utils
Version: 1.5-13+deb9u1
Followup-For: Bug #873086
I've just ran into this same bug.
I believe the proble comes from line 63 of /lib/bridge-utils/bridge-utils.sh
63 dev="${port%.*}"
because this runs in a function within the containing script, it alters
the value of
Hi
Same issue here (Stretch, bridge-utils 1.5-13+deb9u1 ). Thanks to the
previous comment,
I'm now using the following workaround : remove the disable_ipv6 in
create_vlan_port()
in /lib/bridge-utils/bridge-utils.sh (see attached patch). I really
don't see either
why setting up a bridge on
Hello,
we can reproduce this bug.
We have configuration like this:
auto vlan992
iface vlan992 inet manual
bridge_portsbond0.992
bridge_stp off
bridge_maxwait 0
bridge_fd 0
bond0 is the management interface (with IPv4 and IPv6 addresses).
Package: bridge-utils
Version: 1.5-14
Well, what the subject says, this just killed my IPv6 connectivity on a
machine that's untagged on the same interface on which a VLAN is supposed
to be attached to a bridge. I guess at the very least, bride-utils should
avoid touching the configuration of
6 matches
Mail list logo
