sim@phri.UUCP (Kristian Simsarian) (09/03/88)
After successfully getting most of the CAP package to run, we
are trying to get lwsrv up, but there seem to be some problems. Our
Sun's are running 3.2, and we are trying to use lwsrv with some Mac+'s,
and the Laser Writer (box et al).
Certainly everything compiles alright, but when run given:
lwsrv -n Larry -p laser10 -a procset -f LWFonts -l log
it produces this:
lwsrv: Spooler starting for Larry.LaserWriter.*
lwsrv: simple: Font list from file LWFonts
lwsrv: (re)Scanning for Procsets
lwsrv: ProcSet 'appledict' file is './procset/laserprep'
lwsrv: SLInit failed: -1027
A rather cryptic error, which may be expalined in something like
Inside Appletalk, but unfortunetely I don't have one of those handy.
I have attempted to debug, but this doesn't work so well, as I
get an entirely different response in single fork mode:
lwsrv -n Larry -p laser10 -a procset -f LWFonts -d dands
gives:
debugging DDP
debugging ATP
debugging NBP
debugging DDP
debugging ASP
ddp: open socket: opened socket 129
lwsrv: Spooler starting for Larry.LaserWriter.*
lwsrv: simple: Font list from file LWFonts
lwsrv: (re)Scanning for Procsets
lwsrv: ProcSet 'appledict' file is './procset/laserprep'
ddp: open socket: opened socket 130
NBP SndNBP: sending
NBP nbp_timeout: 4 tick timeout on 251657324, 3 remain
NBP SndNBP: sending
NBP nbp_timeout: 4 tick timeout on 251657324, 2 remain
NBP SndNBP: sending
NBP nbp_timeout: 4 tick timeout on 251657324, 1 remain
NBP SndNBP: sending
NBP nbp_timeout: 4 tick timeout on 251657324, 0 remain
NBP SndNBP: sending
NBP status done: found 251657324
atp: creat_rqcb: create 2bc24
ddp: open socket: opened socket 131
........( goes into sleep loop)........
Anyway dbx here isn't much help. The second response seems
encouraging, but I say it doesn't work because, even after a while, the
name "Larry" doesn't come via atlook or The Chooser. If anyone has any
insight, I would appreciate any/all responses to this problem. atis, by
the way, seems to be runing OK.
-Kristian Simsarian
-Public Health Research Instituteroy@phri.UUCP (Roy Smith) (09/08/88)
In article <3455@phri.UUCP> sim@phri.UUCP (Kristian Simsarian)
writes about some problems we've been having getting lwsrv running. Well,
we've got most of the problems solved now.
Several people pointed out that we might be having a problem with
broadcasts. Turns out that was the case. Pretty much in desperation I
went at our atalkatab (which I consider to be largely black magic). I
changed:
55.1 N1 128.122.136.0 PHRI-AT # PHRI AppleTalk
to:
55.1 H 128.122.136.150 PHRI-AT # PHRI AppleTalk
which seems to have solved the broadcast problem.
Next we discovered a bug in lwsrv. The last thing lwsrv does is
something like:
sprintf(buf,"/usr/ucb/lpr -P%s %s %s -J 'MacUser: %s' %s\n",
unixpname,rflag ? "" : "-r",hflag ? "" : "-h", username,tname);
system(buf);
Note that the exit status of system isn't checked. We were testing
lwsrv by printing from a machine called "Lizardi's Mac-II". Of course, the
apostrophie (or single-quote, if you prefer) in the username makes the
system() fail. Lwsrv really should check for things like that or just do a
fork/exec itself instead of system. At the very least, it should check for
an error return from system().
--
Roy Smith, System Administrator
Public Health Research Institute
{allegra,philabs,cmcl2,rutgers}!phri!roy -or- phri!roy@uunet.uu.net
"The connector is the network"