[quagga-dev 3067] Re: interface rename problems

Matt Brown matt at mattb.net.nz
Fri Apr 1 04:52:51 BST 2005


On Thu, 2005-03-31 at 17:36 -0500, Andrew J. Schorr wrote:
> On Thu, Mar 31, 2005 at 11:28:42PM +0100, Paul Jakma wrote:
> > On Thu, 31 Mar 2005, Andrew J. Schorr wrote:
> > 
> > >I don't suppose anybody had a chance to look at or try this patch 
> > >yet?
> > 
> > I dont have any objections if rename can only occur if ~IFF_UP.
> > 
> 
> Thanks for the feedback.  But I imagine the prudent thing is to wait for
> Matt to confirm that it solves his problems before committing the patch...

Pleased to report that it *mostly* works. All the critical functionality
works. The renamed interface has OSPF enabled and can successfully talk
to it's neighbour

Neighbor ID     Pri   State           Dead Time   Address         Interface       RXmtL RqstL DBsmL
10.20.20.10       1   Full/Backup     00:00:31    10.20.20.10     matt:10.20.20.5     0     0     0

However, the interface list still has the "original" interface it in
(eth2 below), this doesn't seem to break anything, but it's ugly and
inaccurate. 

rsc-soekris-1# show int
Interface eth0 is up, line protocol detection is disabled
  index 3 metric 1 mtu 1500 <UP,BROADCAST,RUNNING,MULTICAST>
  HWaddr: 00:00:24:c1:1e:b0
  inet 10.1.23.140/24 broadcast 10.1.23.255
    4352 input packets (0 multicast), 2419172 bytes, 0 dropped
    0 input errors, 0 length, 0 overrun, 0 CRC, 0 frame
    0 fifo, 0 missed
    2856 output packets, 333845 bytes, 0 dropped
    4 output errors, 0 aborted, 4 carrier, 4 fifo, 0 heartbeat
    0 window, 0 collisions
Interface eth1 is down
  index 4 metric 1 mtu 1500 <BROADCAST,MULTICAST>
  HWaddr: 00:00:24:c1:1e:b1
    0 input packets (0 multicast), 0 bytes, 0 dropped
    0 input errors, 0 length, 0 overrun, 0 CRC, 0 frame
    0 fifo, 0 missed
    0 output packets, 0 bytes, 0 dropped
    0 output errors, 0 aborted, 0 carrier, 0 fifo, 0 heartbeat
    0 window, 0 collisions
Interface eth2 is down
  pseudo interface
Interface lo is up, line protocol detection is disabled
  index 1 metric 1 mtu 16436 <UP,LOOPBACK,RUNNING>
  inet 127.0.0.1/8
    2 input packets (0 multicast), 244 bytes, 0 dropped
    0 input errors, 0 length, 0 overrun, 0 CRC, 0 frame
    0 fifo, 0 missed
    2 output packets, 244 bytes, 0 dropped
    0 output errors, 0 aborted, 0 carrier, 0 fifo, 0 heartbeat
    0 window, 0 collisions
Interface matt is up, line protocol detection is disabled
  index 8 metric 1 mtu 1500 <UP,BROADCAST,MULTICAST>
  HWaddr: 00:02:2d:70:fd:4f
  inet 10.20.20.5/24 broadcast 10.20.20.255
    0 input packets (0 multicast), 0 bytes, 0 dropped
    0 input errors, 0 length, 0 overrun, 0 CRC, 0 frame
    0 fifo, 0 missed
    0 output packets, 0 bytes, 0 dropped
    3 output errors, 0 aborted, 0 carrier, 0 fifo, 0 heartbeat
    0 window, 0 collisions
Interface teql0 is down
  index 2 metric 1 mtu 1500 <NOARP>
    0 input packets (0 multicast), 0 bytes, 0 dropped
    0 input errors, 0 length, 0 overrun, 0 CRC, 0 frame
    0 fifo, 0 missed
    0 output packets, 0 bytes, 0 dropped
    0 output errors, 0 aborted, 0 carrier, 0 fifo, 0 heartbeat
    0 window, 0 collisions

Full debugging output can be found at

0.99.0-cvs:
http://www.crc.net.nz/~mglb1/quagga-0.99.0-cvs-no-rename-patch

0.99.0-cvs + patch from quagga-dev 3010:
http://www.crc.net.nz/~mglb1/quagga-0.99.0-cvs-rename-patch

The debug output above is generated using a PCMCIA card, but the results
are essentially the same renaming a permanent interface such as
eth0/eth1 using ip link set <dev> name <foo>, the only difference in
that case is that the ifindex remains constant. 

HTH

Regards

-- 
Matt Brown
matt at mattb.net.nz
Mob +64 275 611 544 www.mattb.net.nz






More information about the Quagga-dev mailing list