home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.bsd
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!decwrl!elroy.jpl.nasa.gov!usc!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!csn!teal.csn.org!dfishman
- From: dfishman@teal.csn.org (Dan Fishman)
- Subject: Re: remote lpr/lpd problems...
- Message-ID: <BxnsKC.C27@csn.org>
- Sender: news@csn.org (news)
- Nntp-Posting-Host: teal.csn.org
- Organization: Colorado SuperNet, Inc.
- X-Newsreader: Tin 1.1 PL4
- References: <BxM76t.9Dw@unx.sas.com>
- Date: Fri, 13 Nov 1992 14:36:10 GMT
- Lines: 46
-
- sastdr@torpid.unx.sas.com (Thomas David Rivers) writes:
- :
- : I'm having an interesting problem with the lpd on 386bsd.
- :
- : Namely, if I (from a remote lpr) print a small file, things work
- : fine. However, if I remotely print a large file the lpd on 386bsd
- : "looses the connection."
- :
- : Looking at the lpd sources, I see such a message emitted if a read
- : from the incoming socket returns <= 0; which is reasonable. (i.e. it
- : either got EOF sooner than expected or there was a problem.)
- :
- : The number of bytes read before this happens is *always* 5120; which is
- : a rather suspicious number.
- :
- : So, I was wondering if there is a limit somewhere I'm running into. I
- : couldn't find anything, but thought I would post and ask..
- :
- :
- : - Thanks -
- : - Dave Rivers -
- : (rivers@ponds.uucp (home))
- : (sastdr@unx.sas.com (work))
- :
- :
- : --
- : UPDATE ALL INFORMATION AND POD INTO COSMOS - Federal Express
-
- I'm not currently a 386BSD user but this could be a basic datacomm issue
- such as I have seen in a BSD4.2 port. If the printer is serial attached
- (RS232) then it may be doing BOTH a XON/XOFF flow control handshake and
- an DTR HI/LOW flow control handshake. DTR is the norm for DOS, XON/XOFF
- is the norm for most UNIX. The DTR handshake however may appear to the
- UNIX serial line driver as a lost line (lost modem control).
-
- If this is the issue, many (I don't know about 386BSD) drivers have a control
- (ioctl) to enable/disable modem control protocols (or possibly even use DTR as
- a standard handshake). Another solution is to modify a cable so as to
- preclude the DTR handshake from reaching the CPU side of the cable as any
- modem control signal. This would mean tying required modem control signals
- high or low as required for your serial IF/driver.
-
- OR NOT! It could be a more subtle networking problem since you specifically
- mentioned remote printing? Hope this helps.
-
- Dan Fishman (dfishman@csn.com)
-