routed trashes routing tables

routed trashes routing tables

Post by g.. » Wed, 04 Aug 1993 23:14:41


I am using pl11 with net-2.

When routed is started in /etc/rc.d/rc.inet2 it appears to destry the
kernel routing table entries that were set up in /etc/rc.d/rc.inet1
including the router. If I do not start routed then things go along quite
happily, but I believe routed is used to manage the routing tables when
entries need to be added to deleted, so it should really be running.

The relevant files follow:

#! /bin/sh
# rc.inet1      This shell script boots up the base INET system.





  # Set up the Ethernet connection(s).
  /etc/ifconfig eth0 ${IPADDR} netmask ${NETMASK}
 #broadcast ${BROADCAST}

  # Attach the loopback device.
  /etc/ifconfig lo
  /etc/route add

  # Set up the primary (static) routes.
  #/etc/route add microwalt
  /etc/route add ${IPADDR}
  /etc/route add ${NETWORK}
  /etc/route add default gw metric 1
  /bin/hostname -S morph
  # All done.


Note that I previously had the 'ifconfig lo' before the 'ifconfig eth0' with
no change in behaviour.


#! /bin/sh
# rc.inet2      This shell script boots up the entire INET system.
#               Note, that when this script is used to also fire
#               up any important remote NFS disks (like the /usr
#               distribution), care must be taken to actually
#               have all the needed binaries online _now_ ...



  # Constants.

  # At this point, we are ready to talk to The World...
  #echo "Mounting remote file systems..."
  #/etc/mount -a -t nfs         # This may be our /usr runtime!!!

  # Start the SYSLOG daemon.  This has to be the first server.
  echo -n "INET: "
  if [ -f ${NET}/syslogd ]
        echo -n "syslogd "

   Start the SUN RPC Portmapper.
  if [ -f ${NET}/rpc.portmap ]
        echo -n "portmap "

  # Start the INET SuperServer
  if [ -f ${NET}/inetd ]
        echo -n "inetd "
        echo "no INETD found.  INET cancelled!"
        exit 1

  # Start the NAMED/BIND name server.
  #if [ -f ${NET}/named ]
#       echo -n "named "
#       ${NET}/named

  # Start the ROUTEd server.
  if [ -f ${NET}/routed ]
        echo -n "routed "

  # Start the RWHO server.
  #if [ -f ${NET}/rwhod ]
#       echo -n "rwhod "
#       ${NET}/rwhod -t -s

  # Start the U-MAIL SMTP server.
  if [ -f /usr/local/lib/smail/smail ]
        echo -n "smail "
        /usr/bin/smail -d0 -bd </dev/null >/dev/null 2>&1 &

  # Start the various INET servers.
  for server in ${IN_SERV}
        if [ -f ${NET}/${server} ]
                echo -n "${server} "

  # Start the various SUN RPC servers.
 # if [ -f ${NET}/rpc.portmap ]
 # then
#       if [ -f ${NET}/rpc.ugidd ]
#       then
#               echo -n "ugidd "
#               ${NET}/rpc.ugidd -d
#       fi
#       if [ -f ${NET}/rpc.mountd ]
#       then
#               echo -n "mountd "
#               ${NET}/rpc.mountd
#       fi
#       if [ -f ${NET}/rpc.nfsd ]
#       then
#               echo -n "nfsd "
#               ${NET}/rpc.nfsd
#       fi
#       # Fire up the PC-NFS daemon(s).
#       if [ -f ${NET}/rpc.pcnfsd ]
#       then
#               echo -n "pcnfsd "
#               ${NET}/rpc.pcnfsd ${LPSPOOL}
#       fi
#       #if [ -f ${NET}/rpc.bwnfsd ]
#       #then
#       #       echo -n "bwnfsd "
#       #       ${NET}/rpc.bwnfsd ${LPSPOOL}
#       #fi
 # fi
 ## # Done!


I am also having a problem with lpd (along with everyone else it seems).
It used to work fine before I upgraded the kernel! So I know that all the
permissions and files are correct. I am printing all files to a remote printer
attached to a Sun.

When a print job is queued, lpr complains that there is no daemon running, even
though lpd is running fine. If I kill lpd and restart it, the job in the queue
will be printed, but when another print job is queued, lpr complains again.

Any suggestions?

|        Tony Gale           |   __/__/__/  __/__/__/  __/__/__/  __/  __/   |
|                            |     __/     __/   __/  __/   __/    __/       |

                 |  All opinions expressed are my own  |


1. How Do Routing Table Entries Get Added to Routing Table at Bootup?

I have, what I think, is a simple question for the TCP/IP guys out there
who are familar with routing tables on AIX 4.1.5/4.2.1 in an IBM SP
environment. Or maybe not.

I have recently noticed that when I execute a 'netstat -r' on a
particular node in the SP environment, there are routing table entries
that I know I didn't put in there. I have to delete these "unwelcomed"
entries and re-add the correct ones. This is undesireable and maybe a
quick crash course in routing tables is justified. These unwelcomed
routing table entries seem to be automatically added right after I have
IPL'ed the node in question.

How did these routing table entries get added in the routing table? Does
anybody know the process?

Thank you in advance for your responses.

< 1K Download

2. adding ftp support to Ip Masquerading

3. routed is trashing my 'default' route

4. when masq, why need 'named'?

5. SIGHUP on named causes trashed routing table

6. New Lilo

7. Named is trashing routing table + smail configuration problem

8. NASTAInternet (what is going on ?)

9. smail conf problem + named trashing routing table

10. building routing tables or listening to OSPF route information?

11. Routing question, how to duplicate route table in new distro?

12. Dropping route from route table

13. Removing a network route from the route table