Route nodes in Unix problem

[_dan_]

Hi all, we are running a Sun V880 with Solaris 8 on it.

A while back I added some manual routes (using route add x.x.x.x x.x.x.x), then we found we didn't need them so I deleted them (using route delete). Looking in the routing table (using netstat -rn) I can see the routes have gone, and everything works ok.

Thing is, when the Sun is rebooted, the old routes are reinstated and we have to go delete them again, I've seached through inetd files and cannot for the life of me find where the hell these unwanted routes are being stored!

If anyone can help, it'd be greatly appreciated.

[Bodo]

I have Linux and the routes are stored in:

/proc/net/ipv6_route
/proc/net/route
/proc/net/rt_cache

[paul99]

Have a look in /etc/inet/networks and /etc/defaultroute.

[chilled]

Some people put scripts into /etc/rc2.d/ and /etc/rc3.d/ without linking them to /etc/init.d/ so that's the main place to look.

Otherwise trawl /etc with a find command

[zumbruk]

Bodo said:

I have Linux and the routes are stored in:

/proc/net/ipv6_route
/proc/net/route
/proc/net/rt_cache

/proc is not writable under Solaris.

[zumbruk]

Check the machine isn't running routed (ps -eaf | grep routed - does Sol8 have pgrep? I forget)

If it is, also check /etc/gateways

And like 'chilled' says, check /etc/rc{n}.d for scripts that aren't linked to /etc/init.d

find /etc/rc?.d -name "*" -exec grep route {} /dev/null ;

Will find all the rc files that mention "route" (NOTE: PH has removed the backslash immediately before the semicolon!)

Off to watch "Life on Mars".

Edited by zumbruk on Tuesday 20th February 20:47

[_dan_]

I have run a find/grep through the whole etc directory and can find nothing mentioning these IPs. Also no mention in etc/inet/networks or defaultrouter

The rc2.d (S69inet) does indeed have a script in it mentioning routes. I'll post it if I may.....



#!/sbin/sh
#
# Copyright (c) 1995, 1997-1999 by Sun Microsystems, Inc.
# All rights reserved.
#
#pragma ident "@(#)inetinit 1.44 99/10/04 SMI"

#
# This is the second phase of TCP/IP configuration. The first part,
# run in the "/etc/rcS.d/S30rootusr.sh" script, does all configuration
# necessary to mount the "/usr" filesystem via NFS. This includes configuring
# the interfaces and setting the machine's hostname. The second part,
# run in this script, does all configuration that can be done before
# NIS or NIS+ is started. This includes configuring IP routing,
# setting the NIS domainname and setting any tunable parameters. The
# third part, run in a subsequent startup script, does all
# configuration that may be dependent on NIS/NIS+ maps. This includes
# a final re-configuration of the interfaces and starting all internet
# services.
#

case "$1" in
'start')
;; # Fall through -- rest of script is the initialization code

'stop')
#
# If we were routing dynamically, we will note this with
# the .dynamic_routing file, so that we can leave the routes
# in place without thinking they're static route entries
# when we come back into states 2 or 3.
#
if /usr/bin/pgrep -x -u 0 'in.routed|in.rdisc' >/dev/null 2>&1; then
/usr/bin/pkill -x -u 0 'in.routed|in.rdisc'
> /etc/.dynamic_routing
fi
exit 0
;;

*)
echo "Usage: $0 { start | stop }"
exit 1
;;
esac

#
# Initialize IPsec only if ipsecinit.conf exists. Otherwise, save the
# kernel memory that'll be chomped if IPsec is loaded.
#
if [ -f /etc/inet/ipsecinit.conf ] ; then
/usr/sbin/ipsecconf -qa /etc/inet/ipsecinit.conf
fi

#
# Set the RFC 1948 entropy, regardless of if I'm using it or not. If present,
# use the encrypted root password as a source of entropy. Otherwise,
# just use the pre-set (and hopefully difficult to guess) entropy that
# tcp used when it loaded.
#
encr=`/usr/bin/awk -F: '/^root:/ {print $2}' /etc/shadow`
[ -z "$encr" ] || /usr/sbin/ndd -set /dev/tcp tcp_1948_phrase $encr
unset encr

#
# Set TCP ISS generation. By default the ISS generation is
# time + random()-delta. This might not be strong enough for some users.
# See /etc/default/inetinit for settings and further info on TCP_STRONG_ISS.
# If not set, use TCP's internal default setting.
#

# Get value of TCP_STRONG_ISS
[ -f /etc/default/inetinit ] && . /etc/default/inetinit
if [ $TCP_STRONG_ISS ]; then
/usr/sbin/ndd -set /dev/tcp tcp_strong_iss $TCP_STRONG_ISS
fi

#
# Configure default IPv4 routers using the local "/etc/defaultrouter"
# configuration file. The file can contain the hostnames or IP
# addresses of one or more default routers. If hostnames are used,
# each hostname must also be listed in the local "/etc/hosts" file
# because NIS and NIS+ are not running at the time that this script is
# run. Each router name or address is listed on a single line by
# itself in the file. Anything else on that line after the router's
# name or address is ignored. Lines that begin with "#" are
# considered comments and ignored.
#
# The default routes listed in the "/etc/defaultrouter" file will
# replace those added by the kernel during diskless booting. An
# empty "/etc/defaultrouter" file will cause the default route
# added by the kernel to be deleted.
#
# Note that the default router file is ignored if we received routes
# from a DHCP server. Our policy is to always trust DHCP over local
# administration.
#
if [ "$_INIT_NET_STRATEGY" = "dhcp" ] && [ -n "`/sbin/dhcpinfo Router`" ]; then
defrouters=`/sbin/dhcpinfo Router`
elif [ -f /etc/defaultrouter ]; then
defrouters=`/usr/bin/grep -v ^# /etc/defaultrouter |
/usr/bin/awk '{print $1}'`
if [ -n "$defrouters" ]; then
#
# We want the default router(s) listed in /etc/defaultrouter
# to replace the one added from the BOOTPARAMS WHOAMI response
# but we must avoid flushing the last route between the running
# system and its /usr file system.
#

# First, remember the original route.
shift $#
set -- `/usr/bin/netstat -rn -f inet | /usr/bin/grep '^default'`
route_IP="$2"

#
# Next, add those from /etc/defaultrouter. While doing this,
# if one of the routes we add is for the route previously
# added as a result of the BOOTPARAMS response, we will see
# a message of the form:
# "add net default: gateway a.b.c.d: entry exists"
#
do_delete=yes
for router in $defrouters; do
set -- `/usr/sbin/route -n add default $router`
[ $? -eq 0 -a "x$5" = "x$route_IP:" ] && do_delete=no
done

#
# Finally, delete the original default route unless it was
# also listed in the defaultrouter file.
#
if [ -n "$route_IP" -a $do_delete = yes ]; then
/usr/sbin/route -n delete default $route_IP > /dev/null
fi
else
/usr/sbin/route -fn > /dev/null
fi
else
defrouters=
fi

#
# Set NIS domainname if locally configured.
#
if [ -f /etc/defaultdomain ]; then
/usr/bin/domainname `cat /etc/defaultdomain`
echo "NIS domainname is `/usr/bin/domainname`"
fi

#
# Run in.routed/router discovery if we don't already have a default
# route installed or if we had been running them in a previous
# multiuser state.
#
if [ -z "$defrouters" -a ! -f /etc/.dynamic_routing ]; then
#
# No default routes were setup by "route" command above - check the
# kernel routing table for any other default routes.
#
/usr/bin/netstat -rn -f inet |
/usr/bin/grep default >/dev/null 2>&1 && defrouters=yes
fi
[ -f /etc/.dynamic_routing ] && /usr/bin/rm -f /etc/.dynamic_routing

if [ -z "$defrouters" ]; then
#
# Determine how many active interfaces there are and how many pt-pt
# interfaces. Act as an IPv4 router if there are more than 2 interfaces
# (including the loopback interface) or one or more point-point
# interface. Also act as an IPv4 router if /etc/gateways exists.
#
# Do NOT act as an IPv4 router if /etc/notrouter exists.
# Do NOT act as an IPv4 router if DHCP was used to configure
# interface(s)
#
inetifaddrs="`/usr/sbin/ifconfig -a4u | /usr/bin/grep inet`"
numifs=`echo "$inetifaddrs" | /usr/bin/wc -l`
numptptifs=`echo "$inetifaddrs" | /usr/bin/egrep -c -e '-->'`

if [ "$_INIT_NET_STRATEGY" = "dhcp" ]; then
numdhcp=`/usr/sbin/ifconfig -a4 | /usr/bin/grep -c DHCP`
else
numdhcp=0
fi

if [ ! -f /etc/notrouter -a $numdhcp -eq 0 -a
( $numifs -gt 2 -o $numptptifs -gt 0 -o -f /etc/gateways ) ]; then
#
# Machine is an IPv4 router: turn on ip_forwarding, run
# in.routed, and advertise ourselves as a router using router
# discovery.
#
echo 'Machine is an IPv4 router.'
/usr/sbin/ndd -set /dev/ip ip_forwarding 1

[ -f /usr/sbin/in.routed ] && /usr/sbin/in.routed -s
[ -f /usr/sbin/in.rdisc ] && /usr/sbin/in.rdisc -r

else
#
# Machine is an IPv4 host: if router discovery finds a router
# then we rely on router discovery. If there are no routers
# advertising themselves through router discovery
# run in.routed in quiet mode. In both cases, turn off
# ip_forwarding.
#
/usr/sbin/ndd -set /dev/ip ip_forwarding 0

if [ -f /usr/sbin/in.rdisc ] && /usr/sbin/in.rdisc -s; then
echo 'Starting IPv4 router discovery.'
elif [ -f /usr/sbin/in.routed ]; then
/usr/sbin/in.routed -q
echo 'Starting IPv4 routing daemon.'
fi
fi
else
/usr/sbin/ndd -set /dev/ip ip_forwarding 0
fi

#
# Run IPv6 if more than lo0 plumbed for IPv6.
#
/usr/sbin/ifconfig -a6u >/tmp/ifconfig.$$
numv6ifs=`/usr/bin/grep -c inet6 /tmp/ifconfig.$$`
if [ $numv6ifs -gt 1 ]; then
#
# Run IPv6 routing only if /etc/inet/ndpd.conf exists, otherwise just
# run the host portion.
#
if [ -f /etc/inet/ndpd.conf ]; then
#
# Machine is an IPv6 router: turn on ip6_forwarding,
# ip6_send_redirects and ip6_ignore_redirect, run in.ripngd, and
# advertise ourselves as a router using neighbor discovery.
#
echo "Machine is an IPv6 router."
/usr/sbin/ndd -set /dev/ip ip6_forwarding 1
/usr/sbin/ndd -set /dev/ip ip6_send_redirects 1
/usr/sbin/ndd -set /dev/ip ip6_ignore_redirect 1
if [ -f /usr/lib/inet/in.ndpd ]; then
/usr/lib/inet/in.ndpd
fi
if [ -f /usr/lib/inet/in.ripngd ]; then
/usr/lib/inet/in.ripngd -s
fi
else
# Machine is an IPv6 host - just run neighbor discovery.
/usr/sbin/ndd -set /dev/ip ip6_forwarding 0
/usr/sbin/ndd -set /dev/ip ip6_send_redirects 0
/usr/sbin/ndd -set /dev/ip ip6_ignore_redirect 0
if [ -f /usr/lib/inet/in.ndpd ] ; then
echo "Starting IPv6 neighbor discovery."
/usr/lib/inet/in.ndpd
fi
fi
#
# Add a static route for multicast packets out of a link-local
# interface, although would like to specify multicast interface using
# an interface name!
#
set -- `/usr/bin/awk '
/inet6 fe80:/ {
print substr($2, 1, index($2, "/" - 1)
}' /tmp/ifconfig.$$`

if [ -n "$1" ]; then
echo "Setting default IPv6 interface for multicast: c"
/usr/sbin/route -n add -interface -inet6 "ff00::/8" "$1"
fi
else
/usr/sbin/ndd -set /dev/ip ip6_forwarding 0
/usr/sbin/ndd -set /dev/ip ip6_send_redirects 0
/usr/sbin/ndd -set /dev/ip ip6_ignore_redirect 0
fi
/usr/bin/rm -f /tmp/ifconfig.$$

#
# Configure tunnels which was deferred by /etc/rcS.d/S30network.sh since
# it depends on the tunnel endpoints being reachable i.e. routing must be
# running.
#
# WARNING: you may wish to turn OFF forwarding if you haven't already, because
# of various possible security vulnerabilities when configuring tunnels for
# Virtual Private Network (VPN) construction.
#
# Also, if names are used in the /etc/hostname.ip.tun* file, those names
# have to be in either DNS (and DNS is used) or in /etc/hosts, because this
# file is executed before NIS or NIS+ is started.
#

#
# IPv4 tunnels
# The second component of the name must be either "ip" or "ip6".
#
interface_names="`/usr/bin/ls /etc/hostname.ip*.*[0-9] 2>/dev/null |
/usr/bin/grep '/etc/hostname.ip6{0,1}.'`"
if [ -n "$interface_names" ]; then
(
echo "configuring IPv4 tunnels:c"
# Extract the part after the first '.'
set -- `for intr in $interface_names; do
/usr/bin/expr //$intr : '[^.]*.(.*)$'; done`
while [ $# -ge 1 ]; do
# Skip empty files
if [ ! -s /etc/hostname.$1 ]; then
shift
continue
fi
/usr/sbin/ifconfig $1 plumb 2>&1 >/dev/null
while read ifcmds; do
if [ -n "$ifcmds" ]; then
/usr/sbin/ifconfig $1 inet $ifcmds
2>&1 >/dev/null
fi
done </etc/hostname.$1
echo " $1c"
shift
done
echo "."
)
fi

#
# IPv6 Tunnels
# The second component of the name must be either "ip" or "ip6".
#
interface_names="`/usr/bin/ls /etc/hostname6.ip*.*[0-9] 2>/dev/null |
/usr/bin/grep '/etc/hostname6.ip6{0,1}.'`"
if [ -n "$interface_names" ]; then
(
echo "configuring IPv6 tunnels:c"
# Extract the part after the first '.'
set -- `for intr in $interface_names; do
/usr/bin/expr //$intr : '[^.]*.(.*)$'; done`
while [ $# -ge 1 ]; do
# Skip empty files
if [ ! -s /etc/hostname6.$1 ]; then
shift
continue
fi
/usr/sbin/ifconfig $1 inet6 plumb 2>&1 >/dev/null
while read ifcmds; do
if [ -n "$ifcmds" ]; then
/usr/sbin/ifconfig $1 inet6 $ifcmds
2>&1 >/dev/null
fi
done </etc/hostname6.$1
echo " $1c"
shift
done
echo "."
)
fi



edit: replace the wink smilie with ; )

Edited by _dan_ on Wednesday 21st February 11:34