Date: Sun, 7 Jun 1998 19:25:03 +0200 (CEST) From: hm@kts.org (Hellmuth Michaelis) To: freebsd-isdn@FreeBSD.ORG (ISDN for BSD) Subject: killing and restarting isdnd (FYI) Message-ID: <m0yijBn-00000XC@bert.kts.org>
next in thread | raw e-mail | index | archive | help
After some bug reports which i didn't understood, i finally got it ..
Since some time, there must have been an open() on /dev/i4b taking place to
enable the transmission of received frames upwards in the ISDN stack. When
the daemon exits, the close() on /dev/i4b disables the sending of frames
upwards in the ISDN stack.
Now when the system is booted, the isdn kernel part is properly initialized
and reset and when the daemon is started, the kernel part of i4b starts from
clean data structures. When the daemon is killed, no further received frames
from the S0 bus are sent to the upper layers, effectively freezing the kernel
part of i4b. What is currently _not_ done is, resetting the data structures
describing the states of the layers to their defaults, instead they remain
as they were left.
When the daemon then is started again in this situation, the kernel parts
of i4b un-freeze in the state they had when the daemon was killed and this
can lead to all sorts of bad behaviour.
The effect is, that it is currently _NOT_ possible to cleanly kill isdnd and
then restart it some time later, bad things can happen. You have to reboot
the whole machine in this case.
Reconfiguration of isdnd should work; this is possible by sending HUP to the
daemon or from the fullscreen menu interface.
I'll try to fix it ASAP, but i'm currently a bit short of time ....
hellmuth
--
Hellmuth Michaelis hm@kts.org Hamburg, Europe
A duck is like a bicycle because they both have two wheels except the duck
(terry@cs.weber.edu)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-isdn" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?m0yijBn-00000XC>