[net.unix-wizards] 4.2bsd uucico LOST LINE bug

jonab@sdcrdcf.UUCP (Jonathan Biggar) (10/04/84)

I am using cron to schedule a "uucico -r1" every hour to call all of
the sites that we have queued work for.  The first site called completes
successfully, but every call after that gets a "LOST LINE" error message.
I am using a gandalf terminal switch to connect to vadic 1200 baud dialers,
with home grown code to drive the gandalf in condevs.c.

I seem to remember someone having a bug fix for this about 6 months ago,
but I can't find it.  Does anyone have a bug fix for me or at least
some suggestions?

Jon Biggar
{allegra,burdvax,cbosgd,hplabs,ihnp4,sdccsu3}!sdcrdcf!jonab

chris@umcp-cs.UUCP (Chris Torek) (10/05/84)

I took the easy way out, and modified our ``uucp.hourly'' shell script to
just invoke uucico multiple times, instead of trying to get uucico -r1 to
run for each system that had work.

Here are the relevant parts of the script:

	AWK=/bin/awk
	CICO=/usr/lib/uucp/uucico
	GREP=/bin/grep
	SNAP=/usr/bin/uusnap

	# poll any sites for which work is spooled
	for system in `$SNAP | $GREP Cmd | $AWK '{print $1}'`
	do
		$CICO -r1 -s$system
	done
-- 
(This page accidently left blank.)

In-Real-Life: Chris Torek, Univ of MD Comp Sci (301) 454-7690
UUCP:	{seismo,allegra,brl-bmd}!umcp-cs!chris
CSNet:	chris@umcp-cs		ARPA:	chris@maryland

rick@seismo.UUCP (Rick Adams) (10/05/84)

The following fragment of code from the beginning of cico.c will
fix the problem.

---rick

+ #ifdef BSD4_2
+ 	if (getpgrp(0) == 0) { /*We have no controlling terminal */
+ 		setpgrp(0, getpid());
+ 	}
+ #endif

	ret = subchdir(Spool);
	ASSERT(ret >= 0, "CHDIR FAILED", Spool, ret);
	strcpy(Wrkdir, Spool);