Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

setting default interface for multicast

Status
Not open for further replies.

jbm417

Technical User
Dec 11, 2007
15
0
0
US
I have an old system running SunOS Rel 5.6 that boots very slowly. During the boot process there is a message:
setting default interface for multicast limey: bad value.

Later in the boot process there is another message: syslogd: line 24 unexpected gatnetconfig failure

The best I can figure out is the routing table is messed up & causing a panic attack. My problem is I don't have any clue about how to fix.

I know the OS is very old, but I cannot upgrade because the machine has some special hardware attached & the driver is not available for any newer OS version.

Please help. TIA, Joe
 
Sorry, we don't have any SunOS 5.6 systems where I work. But this is how I would 'troubleshoot' the problem. Try
grep multicast /etc/init.d/*
grep limey /etc/init.d/*

I'm looking for the script that runs at boot time that causes the error. If a script was changed, it might not now work as intended. Any good Sys Admin would have kept the old version(s) (as .old or _orig or _ddMmmyy) and documented the changes. Once found, debug the script.

If nothing is obvious, how about making changes to /etc/rc2 and /etc/rc3 (keeping the originals to put them back later) to echo debug messages to the console at boot time as each script is called in the "for f in /etc/.../S*; do" loop. Then reboot.


I hope that helps. Please feel free to post further questions.

Mike
 
Thx for the reply, Mike.

Grep multicast returned:

echo "setting default interface for multicast: \c"
for the script inetsvc.

In the script this line follows the echo:

/usr/sbin/route add -interface -netmask "240.0.0.0 "224.0.0.0" "$mcastif"

I added an echo in the script (before the route) & $mcastif is set to the host name of limey which is what I expected.

I do understand the concept of route add, but I do not understand this particular one; e.g. 240.0.0.0 and 224.0.0.0
All my IPs are 192.168.0.X. This machine, limey is 192.168.0.111.

The routing table on limey is:
192.168.0.0 192.168.0.111 U 2 0 le0
127.0.0.1 127.0.0.1 UH 0 326 lo0

I would appreciate any thoughts/help you can give.

TIA, Joe
 
Hi Joe,

On one of our SunOS 5.8 (Solaris 8) systems, the equivalent line in /etc/init.d/inetsvc (near the end) is:

/usr/sbin/route -n add -interface "224.0/4" "$mcastif" >/dev/null

The routing table contains 2 extra lines compared with yours, which would look like:

224.0.0.0 192.168.0.111 U 1 0 le0
default <gateway-IP> UG 1 <nos>

The 'default' line is there because the file /etc/defaultrouter contains <gateway-IP> .

In the Solaris 8 man page for route there doesn't seem to be a -netmask 'modifier'. However on Compaq's Tru64 there is a -netmask 'modifier', so perhaps it can be used but not on Solaris. Could someone with experience of another UNIX flavour have made the change? How about trying the line from Solaris 8 (above) and seeing if that works?


I hope that helps.

Mike
 
If you are not using multicasting you can just comment out those lines if you want.

I noticed though in your typing you left of a qoute?

/usr/sbin/route add -interface -netmask "240.0.0.0[red]"[/red] "224.0.0.0" "$mcastif"
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top