home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!elroy.jpl.nasa.gov!swrinde!mips!darwin.sura.net!haven.umd.edu!ni.umd.edu!sayshell.umd.edu!louie
- From: louie@sayshell.umd.edu (Louis A. Mamakos)
- Newsgroups: comp.sys.next.sysadmin
- Subject: Re: stop UDP from bringing up SLIP?
- Keywords: SLIP, 920802, UDP, access
- Message-ID: <1992Aug13.155833.18220@ni.umd.edu>
- Date: 13 Aug 92 15:58:33 GMT
- References: <1992Aug10.134954.266402@cs.cmu.edu>
- Sender: usenet@ni.umd.edu (USENET News System)
- Organization: University of Maryland College Park
- Lines: 29
- Nntp-Posting-Host: sayshell.umd.edu
-
- In article <1992Aug10.134954.266402@cs.cmu.edu> beatty@sqhilton.pc.cs.cmu.edu (Derek Beatty) writes:
- >I'm using Mamakos' 920802 version of Dialupip to connect to campus through
- >a Cisco ASM/3. (So my default route points to this link.) I'd like to
- >stop SBook from bringing up the link, so I tried removing the word "udp"
- >from the "protocols" line of my ".access" file. I thought this would
- >suffice. But now authough when I launch SBook a message like:
- >Aug 7 18:23:09 sqhilton tcldiald[115]: Unauthorized protocol UDP (17)
- >is logged via the syslog facility, as I expect, subsequent attempts to
- >open TCP connections seem not to bring up the link automatically; instead,
- >I have to use the duioctl to bring it up by hand.
-
- I think that I understand what is going one. You have stumbled onto a
- bug that goes something like this:
-
- When an attempt is made to transmit a packet, a message is sent from
- the kernel driver to the tcldiald process. The SLIP interface is
- marked as "bring-up in progress". This prevents further messages from
- being sent to the tcldiald process. A timer is also started, by
- default it is 5 minutes. If nothing happens by then, then the
- interface is cleared (actually, it may get disabled.. I'll have to
- check).
-
- In any case, since tcldiald refuses to bring up the link because its
- not the right packet type, it ignores it and moves on. It really
- ought to reset the state of the interface and wait for further
- attempts. I will look at how difficult it might be to fix this to do
- the right thing.
-
- louie
-