> Last week I compiled ghostscript 3.33 and Monroe Williams' StyleWriter 
> driver and finally got my system to print.  Unfortunately, while doing 
> so, I ran out of drive space, and even though it printed a few times, 
> eventually I kept getting messages that said something to the tune of, 
> "could not write...drie full."  I had quit trying to print, but I would 
> get these messages even when logging off.  So, I rm'd my games and some 
> other space hogs.  I also rebooted for some reason or another, and during 
> the reboot, I got a core dump due to a segmentation error.  I rebooted 
> again into single-user and did a disk check.  Then I rebooted into 
> multi-user.  Now whenever I boot I get these messages:
> 
> starting rpc daemons:  portmapportmap: permission denied
> 
> ...then a few lines down...
> 
> quotaon: permission denied

Probably filesystem corruption...

> Also, shortly after I log in, either as myself (using bash) or as root 
> (using csh), I get spurious messages timed at about two minute intervals 
> for the first six minutes or so of the log in.  The latest set of these 
> messages read:
> 
> Mar 31 12:05:41 hussar inetd[78]:pmap_set: 100001 3 17 1024: Address 
> already in use
> Mar 31 12:05:41 hussar inetd[78]:pmap_set: 100001 3 17 1024: Address 
> already in use
> Mar 31 12:05:41 hussar inetd[78]:pmap_set: 100001 3 17 1024: Address 
> already in use
> Mar 31 12:07:42 hussar inetd[78]:pmap_set: 100002 2 17 1025: Address 
> already in use
> Mar 31 12:07:42 hussar inetd[78]:pmap_set: 100002 2 17 1025: Address 
> already in use
> Mar 31 12:07:42 hussar inetd[78]:pmap_set: 100002 2 17 1025: Address 
> already in use
> Mar 31 12:09:42 hussar inetd[78]:pmap_set: 100002 3 17 1025: Address 
> already in use
> Mar 31 12:09:42 hussar inetd[78]:pmap_set: 100002 3 17 1025: Address 
> already in use
> Mar 31 12:09:42 hussar inetd[78]:pmap_set: 100002 3 17 1025: Address 
> already in use
> 
> Now I can't print, and lpd's log doesn't even show my attempts to print, 
> even though it shows my abortive attempts from before I had gs and the 
> stylewriter stuff set up.

I don't quite know what is causing the above.  You might still be rather 
close to your disk limit, tho.

> I have dug through the "Red Book" and the man pages for portmap and 
> pmap_set, as well as an O'Reilly book "Essential System Administration".  
> I think I may be dealing with a permissions problem, but I don't know 
> why.  If I do a "ls -alF" on portmap and quotaon it shows:
> 
> br-S--S---  256 768  1792   232,   0 Mar 31 11:56 portmap
> br-S--S---  512 768  1792   176,   0 Jul 13 10:42 quotaon
> 
> (portmap's date is different because I "touch"'d it.)
> 
> I read through the "Red Book's" explanation of the setuid and sticky 
> bits, but I really don't know if that applies here.  I have also tried to 
> set the permissions on both of these files by using "chmod ug+s 
> ./portmap", but that didn't seem to work.

Ah, the problem is that these files have become corrupted.  Are you using 
the ncrscsi driver on an '030-based Mac?  If so, you might consider 
changing to the SBC driver instead.  If you'll notice, the 'b' in the 
first column above indicates that these files are now block special files 
(i.e. block device interfaces), not very useful as executables ;-)  
You'll need to reinstall the base distribution to fix this.

> Also, as per the stylewriter set up instructions, I have a symbolic link 
> from a file /dev/stylewriter to /dev/tty01, which according to the 
> instructions is the print port for NetBSD.  To which user should tty01 
> belong?  Mine shows an ownership of uucp wheel.  Should it be root wheel, 
> or is there a reason that tty01 (and I notice now, tty00) should belong 
> to uucp?

I think that the reasons for this are leftover from when uucp was a 
fairly common way for unix machines to communicate.  I think that 
/etc/mtree shows to whom all the files should belong.

> What can you all recommend?  Do I need to re-install etc12 and/or base12? 
>  Is there something in my rc that could be screwed up?

Although something in /etc/rc or whatever could be screwed up, 
reinstalling base12 should probably fix the difficulties.

I hope this helps.

Later.

-- 
Colin Wood                                      ender@is.rice.edu
Consultant                                        Rice University
Information Technology Services                       Houston, TX