Loading ...
Sorry, an error occurred while loading the content.

3016Re: NSLU2 sometimes calls itself "noname" (or gives itself no name!?)

Expand Messages
  • Sebastian Meiss
    Nov 8, 2005
    • 0 Attachment
      Found a the problem and a workaround, but need a hint to find a real
      solution.

      the problem:

      udhcpcist started only with the option for Hostname
      >> -H, --hostname=HOSTNAME Client hostname <<

      (in my case I can see after ps'ing that it runs as:
      >> /sbin/udhcpc -H gillian << #gillian being my NSLU2's name

      but without the option for "identifier":
      >> -c, --clientid=CLIENTID Client identifier <<

      which seems to be needed for some things, cause after killing and
      restarting the udhcpc with "-c gillian" added, it no longer gets
      identified as "noname" but as "gillian" - as it should.

      Since I could not find where (which start script or such) the udhcpc
      is being started during booting (ANY HINT VERY WELCOME!), I wrote a
      script taht kills the udhcpc and then starts it again with both
      parameters:
      /opt/etc/init.d/S60udhcpc
      -------
      #!/bin/sh

      if [ -n "`pidof udhcpc`" ]; then
      /bin/killall udhcpc 2>/dev/null
      fi


      /sbin/udhcpc -c gillian -H gillian
      --------

      (replace "gillian" by hostname of your choice).

      For now this works, but it is not satisfying, being a workround rather
      than a solution. Better than killing and restarting would of course
      be, to start it with the correct parameters right from the beginning.
      Can anybody tell me where to do this editing?
    • Show all 2 messages in this topic