home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.aix
- Path: sparky!uunet!cs.utexas.edu!geraldo.cc.utexas.edu!portal.austin.ibm.com!awdprime.austin.ibm.com!jaws1.Berkeley.EDU!jaws
- From: jaws@jaws1.austin.ibm.com (Jason A. Woods)
- Subject: Re: enq fills up /var and leaves it
- Sender: news@austin.ibm.com (News id)
- Message-ID: <C17y8J.2290@austin.ibm.com>
- Date: Thu, 21 Jan 1993 19:31:30 GMT
- Reply-To: YOURLOGIN@YOURCOMPUTER.austin.ibm.com
- References: <1jmfraINN9bs@rave.larc.nasa.gov>
- Organization: IBM Austin
- Lines: 25
-
- >I found an interesting behavior problem on my RS6K today. A user tried to
- >print a 10MB file using lpr. When the spool file filled up /var, enq reported
- >that the spool device was full (this message is directed to the user who
- >issued the lpr command) and aborted, leaving the spool file in
- >/var/spool/qdaemon. Of course, having a full /var causes all kinds of
- >problems, since nobody can save edited files or spool other printouts then.
- >When I called IBM, they said that this behavior was "expected", since enq
- >didn't "know what to do when the spool device filled up". Does this make sense
- >to anyone out there? Of course I have instituted the obvious workaround,
- >which is making people use lp instead of lpr, but I feel that it shouldn't be
- >that hard to have enq delete the temporary spool file if the spool device
- >fills up.
-
- There is an apar out there that addresses almost the same thing,
- ix31756; but it was a change to qdaemon so that it would remove
- the files in /var/spool/qdaemon when a job was stopped/canceled
- abnormally. Without looking closer, I couldn't say that this would
- fix this or not; however, the fact remains that we should be cleaning
- up after ourselves, so I would think that if ix31756 doesn't work
- that another apar is in order.
-
- Jason Woods - Go Ahead, Take A RISC . . .
- AIX Software Defect Support
-
- DISCLAIMER - "I did it during lunch . . . really!"
-