draper@xxx.dec.com (11-Dec-1987 1321) (12/12/87)
as the subject said, i've got the vanilla uucp system installed (obligatory wish for hdb . . . ) i'm able to contact systems ok thru ph0 and the OBM. a direct connection thru tty000 across a null-modem cable is another story, tho. i've got a cable running from the 3b1 to a dec pro380 (pdp11/70) running venix 5.0 (sys5.2.0). the pro has getty running on the serial port, and i'm able to connect and login in fine from the 3b1 using cu. the 3b1 L.sys entry is thepro Any tty000 9600 tty000 ogin:--ogin:-- nuucp similarly, on the pro, the USERFILE entry is nuucp,giff / (the 3b1 is named giff, the pro380 is thepro.) i can queue jobs on the 3b1 for the pro380 w/o a problem. unfortunately, when uucico starts up, it immediately fails on the connection. the message in /usr/spool/uucp/LOGFILE is something like . . . . DIRECT WRITE TO TTY000 FAILED (TIMEOUT) i really don't know where to go from here. i've tried manually starting uucico up with a -x4 debug level, but i don't understand all the messages and i still don't understand why it's failing. i'd appreciate any and all suggestions; i'll summarize email responces. btw: am i being incompetant, or were the man entries for uucp/uux/uupick/uusend/uustat/uuclean/uucico and the one "high-level" chapter in the programmer's manual the only information about uucp in the 3.51 doc set??? if so, <FLAME TIME> i see no excuse for that. i realize this is a "user-friendly" system with menu's for editting the L[.-]* files, but a few pages on the file formats should be in the docs. i had to go the the docs on another system to understand them. <FLAMES EXTINGUISHED> bruce draper draper@tallis.dec.com
allbery@ncoast.UUCP (Brandon Allbery) (12/17/87)
As quoted from <8712111822.AA04054@decwrl.dec.com> by draper@xxx.dec.com (11-Dec-1987 1321): +--------------- | am i being incompetant, or were the man entries for | uucp/uux/uupick/uusend/uustat/uuclean/uucico and the one | "high-level" chapter in the programmer's manual the only | information about uucp in the 3.51 doc set??? if so, | <FLAME TIME> | i see no excuse for that. i realize this is a "user-friendly" | system with menu's for editting the L[.-]* files, but a few | pages on the file formats should be in the docs. i had to | go the the docs on another system to understand them. | <FLAMES EXTINGUISHED> +--------------- How true. I was pleasantly surprised to find that the System V/386 docs on a local system I was configuring for UUCP documented not only these files, but also gave a fairly comprehensive description of C. and X. files. *That* is what *ought* be shipped with UUCP! -- Brandon S. Allbery necntc!ncoast!allbery@harvard.harvard.edu {hoptoad,harvard!necntc,cbosgd,sun!mandrill!hal,uunet!hnsurg3}!ncoast!allbery Moderator of comp.sources.misc
wtm@neoucom.UUCP (Bill Mayhew) (12/18/87)
The DIR uucp connection only needs to have pins 2, 3, and 7. If everything else is left unconnected, the unused signals are pulled high at the 3B1's connector. Release 3.51 does support xon/xoff flow control. If the DEC machine isn't pulling the DSR and DCD leads (pins 4 and 6) high, the 3B1 will probably get mad and not connect. The incorrect signals might not affect the operation of cu. I have a Telebit modem on tty000. I lied to L.sys and inittab that it was a direct to host connection. Of course, the telebit does provide athe full set of signals. It is wired by a straight cable. I use the send/expect tokens in L.sys to program the Telebit, as it is much easier than using modemcap. <<insert obligatory HDB request>> I also agree that the included uucp docs are utter cr*p. Rather surprising, since AT&T is a telecommunications company. The programmers's reference has a reasonable description of what a uucp network is, but does not inform on how to implement it. AT&T suggested ordering the "Basic Networking Utilities" manual for the 3B2 to use with the 3B1's uucp. I'm not sure it applies, though. There is an "electronic mail utility" available for the 3B1, but nobody has been able to give an adequate description of what it is. I have to assume the worst that it is just some awful user agent monstrosity. I don't want to ante the $250 unless I can find out what it is. Were it HDB, I'd be willing to shell out the bucks. --Bill