the snippet from /etc/init.d/network, which is owned by openrc, causes one route unreachable. if [ "$RC_UNAME" = "Linux" ]; then ifconfig lo 127.0.0.1 netmask 255.0.0.0 || cr=1 route add -net 127.0.0.0 netmask 255.0.0.0 \ gw 127.0.0.1 2>/dev/null route add -net 127.0.0.0 netmask 255.0.0.0 \ gw 127.0.0.1 reject 2>/dev/null Here it goes: localhost linux # ip route 192.168.1.0/24 dev br0 scope link metric 5 unreachable 127.0.0.0/8 scope host 127.0.0.0/8 via 127.0.0.1 dev lo scope link default via 192.168.1.1 dev br0 metric 5 Once I comment out the line with "reject" directive I get this after a reboot: 192.168.1.0/24 dev br0 scope link metric 5 127.0.0.0/8 via 127.0.0.1 dev lo scope link 127.0.0.0/8 via 127.0.0.1 dev lo default via 192.168.1.1 dev br0 metric 5 which is not good either. Reproducible: Always
Could you please give us some more info about your network configuration?!
Fixed here http://roy.marples.name/projects/openrc/changeset/23f0d30e6e545d4059ad07baff6ec437d82510dc
This is in upstream git, so it will be included in the next release.
(In reply to comment #2) > Fixed here > http://roy.marples.name/projects/openrc/changeset/23f0d30e6e545d4059ad07baff6ec437d82510dc > Hi, the above fix doesn't work with my version of openrc-0.6.1.-r1. Although the duplicate route to localhost dissapears other interfaces stop working. Running the script results in: SIOCADDRT: no such device It might work with the git version because the line numbers of the two versions do not match. I infer there is more changes int the git version which I do not have installed than in the 0.6.1-r1 version.
(In reply to comment #4) > (In reply to comment #2) > > Fixed here > > http://roy.marples.name/projects/openrc/changeset/23f0d30e6e545d4059ad07baff6ec437d82510dc > > > Hi, the above fix doesn't work with my version of openrc-0.6.1.-r1. > Although the duplicate route to localhost dissapears other interfaces stop > working. > Running the script results in: > SIOCADDRT: no such device This message would not come from deleting a route command, so it probably does have to do with other changes in the script. However, since this is fixed in git and will be installed with the next release of openrc, I am closing this bug.
*QA hat on*: since when a bug is considered “FIXED” without being committed to the Gentoo Tree? I can undersand a “RESO UPSTREAM” resolution, but this is definitely not “RESO FIXED”.