Thread View: comp.protocols.kermit.misc
2 messages
2 total messages
Started by Bjorn Eng
Mon, 02 Oct 1995 21:51
Real Fix for Kermit on Solaris?
Author: Bjorn Eng
Date: Mon, 02 Oct 1995 21:51
Date: Mon, 02 Oct 1995 21:51
23 lines
1473 bytes
1473 bytes
Hi, I have built C-Kermit 5A(190) on a Sun Sparc-20 Solaris 2.4 machine. It runs fine but if I use it for dial-out it messes up the port monitor (Solaris getty-like process). Looking through the '.bwr' files I saw this problem mentioned with a fix that didn't appeal to me too much. The fix involves killing off the port monitor and starting a new one every time you run Kermit. Aliasing kermit to a script that runs setuid root to reset the port... Does anyone have a fix for whatever Kermit is doing to the port monitor? The symptom is this: After completing a dial-out call with kermit the port monitor will not respond to incoming calls. The modem will still answer but the port monitor will not prompt for a login. This does not happen when you use tip for dialout. But tip is a little too hard core for some of our users :-) so I'm hoping for a kermit solution... Thanks, Bjorn Eng Jet Propulsion Laboratory
Re: Real Fix for Kermit on Solaris?
Author: Paul Slootman
Date: Thu, 05 Oct 1995 15:15
Date: Thu, 05 Oct 1995 15:15
55 lines
2842 bytes
2842 bytes
Frank da Cruz <fdc@watsun.cc.columbia.edu> wrote: >Bjorn Eng <Bjorn.T.Eng@jpl.nasa.gov> wrote: >: I have built C-Kermit 5A(190) on a Sun Sparc-20 Solaris 2.4 machine. >: It runs fine but if I use it for dial-out it messes up the port monitor How does it mess it up? What are the symptoms? >: (Solaris getty-like process). Looking through the '.bwr' files I saw >: >: Does anyone have a fix for whatever Kermit is doing to the port >: monitor? > >Kermit isn't doing anything to the port monitor. Kermit is using the same >technique that it has used successfully for over a decade to access dialout >devices on more than 400 different UNIX platforms. [...] Kermit used, and is using, the same technique on more than 400 different UNIX platforms? The same one? On all of them? Magic! :-) Pray tell, what is that technique? >If Sun changes the rules >and requires some proprietary scheme for obtaining indulgences from their >"Port Monitor", then we'll be glad to include the code for this if it is not >proprietary and need not be licensed and can be redistributed and somebody >tells us what it is. I would expect that Sun is using the "normal" SVR4 thang for Solaris. If the port is properly configured (with sacadm/pmadm/ttymon-specific or whatever; use admintool if you can!), an appropriate lock in /var/spool/locks should do the trick. >: This does not happen when you use tip for dialout. >: >Of course it doesn't -- Solaris Tip is specially coded by Sun for Solaris, >and you don't get source code to see how they handle the port monitor. The command "truss" can do wonders in cases you don't have source code, and are wondering what magic some command is doing. Many, if not most, unknown interactions can be, eh, reverse-engineered using truss. >Kermit is coded to be portable. It's not easy being portable, especially >now that every company has its own unique "open standard" for doing each >thing :-) Amen to that :-| comp.unix.solaris added to newsgroups, and followups adjusted. Paul Slootman -- key to UNIX: echo '16i[q]sa[ln0=aln100%Pln100/snlbx]sbA0D4D465452snlbx'|dc Murphy Software, Enschede, The Netherlands paul@murphy.nl work: paul@ahwau.ahold.nl +31 756 593946 fax +31 756 351240 home: paul@wurtel.hobby.nl +31 546 851514
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads