NM: the usual rant

Sunday, 1 March 2009

I'm having problems with NM again.
I often have to re-boot once or twice before it connects.

But that is not the subject of this rant.
To try to make sense of what is happening
I have been looking through NM entries in /var/log/messages
and I find these less than helpful.

Was ever so much written to so little purpose?

I see for example:

Feb 23 14:49:41 mary NetworkManager: <WARN>  check_one_route():
   (eth1) error -34 returned from rtnl_route_del(): Sucess#012
Feb 23 14:49:41 mary avahi-daemon[2272]:
  Withdrawing address record for 192.168.2.19 on eth1.
Feb 23 14:49:41 mary avahi-daemon[2272]:
  Leaving mDNS multicast group on interface eth1.IPv4 with address 
192.168.2.19.
Feb 23 14:49:41 mary avahi-daemon[2272]:
  Interface eth1.IPv4 no longer relevant for mDNS.
Feb 23 14:49:47 mary NetworkManager:
  <info>  (eth1): now unmanaged

What does this mean?
What is error -34?
How can one find out?

Or:

Feb 24 20:47:58 mary NetworkManager: <info>  (
  eth1): supplicant interface state:  starting -> ready
Feb 24 20:50:29 mary kernel: eth1:
  Error -19 disabling MAC port
Feb 24 20:50:30 mary NetworkManager: <info>
  (eth1): now unmanaged

What is error -19?

Surely Linux error messages should be intelligible
by the common user?
Or are they intended solely for gurus?

And what are all these states NM keeps going into
and coming out of?

Feb 24 20:50:38 mary NetworkManager: <info>
  (eth1): preparing device.
Feb 24 20:50:38 mary NetworkManager: <info>
  (eth1): deactivating device (reason: 2).
Feb 24 20:50:38 mary NetworkManager: <info>
  (eth1): device state change: 2 -> 3

What is "reason: 2"?
Where are the states documented?

NM = nightmare ...

-- 
Timothy Murphy  
e-mail: gayleard /at/ eircom.net
tel: +353-86-2336090, +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland