Bug 114786 (TBa7981396)

Created attachment 97416 [details]
Traceback "traceback-ifup.txt"

Created attachment 97417 [details]
Traceback "traceback-start.txt"

Could you please try:
ftp://people.redhat.com/harald/redhat-config-network/RHEL3/1.2.60/

Result of test redhat-config-network-1.2.60-1

The program does not longer crash with traceback at start, if an
ISDN-interface is activ or you want activate it.

It is possible to configure ISDN. You can activate it but it does not
goes in the o.k status. The interface ippp0 comes up. But there is an
message in the windows:

usage: userisdnctl <interface-config> <dial|hangup|status|report

I see this message with providername "brickxmp" and also "ippp0".

It is possible to use /usr/bin/redhat-control-network.

I think the fix for the traceback is o.k.. The other mistaces are
listed in Bug #114876.

Sorry, the other mistaces are listed in Bug #114867.

I tested this with RHEL 3 and it worked with the RHEL 3 userisdnctl.
which version of isdn4k-utils do you have installed? The RHEL 3 original?

You are right. The isdn4k-utils-3.2-8.p1 with I have installed to find
the reason for the dynamic channel bundling problem is not ready for
RHEL 3 and Than was tell me that.

With the original isdn4k-utils you can activate the ISDN with dial if
you have configure it. All time later you are start
redhat-config-network and the ippp0 is already up I see that the ISDN
status is "deacativ". It is not possible to deactivate and if you use
the activate buttom the messange is "ippp0: Resource temporarily
unavailable".

The message "Dialing of ippp0 triggered" I see only one time if I
configure ISDN new.
The same problem I see in redhat-control-network.
I think, the problem is, that the status of interface will not refresh
 . Refresh the status was the problem for traceback.


This mistace is not relevant for me because I use Dial on Demand.

the status is refreshed by:
# isdnctrl status ippp0
return code 0    -> activated
return code != 0 -> deactivated

so:
# isdnctrl status ippp0 && echo Activated || echo Not Activated

btw, a manual:

# ifdown ippp0
# ifup ippp0

solved the "Resource temporarily unavailable" problem here...

This is a Dial on Demand problem I found out now.

# ifup ippp0
# isdnctrl status ippp0 && echo "Activated" || echo "Not Activated"
ippp0 is not connected
Not Activated
# ifdown ippp0
# isdnctrl status ippp0 && echo "Activated" || echo "Not Activated"
ippp0: No such device
Not Activated
# ifup ippp0
# ping 172.16.39.81
# isdnctrl status ippp0 && echo "Activated" || echo "Not Activated"
ippp0 connected from 7119035642
Activated

Now I use a configuration without Dial on Demand. The "Activate" works
correct. The connection close after HUPTIMEOUT=300. I am not able to
cancel the connection with "Deactivate" before hanguptimeout it hangs.

I use a callback connection. Normaly there is Dail on Demand relevant
for me.
In the Dial on Demand mode the "Activate" and "Deactivat" functions
are not needed.

If the "Activate" and "Deaktivate" works correct in your own tests, I
think, then the tool is o.k..

I see it again, the problem with default route, if the ifcfg-xxx is
not named ifcfg-ippp0. ifcfg-providername does not work correct, see
other reports.

This bug is apparently fixed.

An errata has been issued which should help the problem described in this bug report. 
This report is therefore being closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files, please follow the link below. You may reopen 
this bug report if the solution does not work for you.

http://rhn.redhat.com/errata/RHBA-2004-076.html