> From [EMAIL PROTECTED] Thu Aug 10 11:13:53 2000 > From: [EMAIL PROTECTED] > To: [EMAIL PROTECTED] > Date: Thu, 10 Aug 2000 12:58:27 -0400 > Subject: LPRng: Possible issue eith LPRng-3.6.22 and ifhp-3.3.19 > > > > I have put LPRng-3.6.22 and ifhp-3.3.19 into production and they are working > well. I have two strange problema that I'd like Patrick to take a look at. > > 1) I had a print job that was a runaway. Some user turned off the printer to > stop it. When the user did an LPQ they got the following status back: > > > Printer: laser7@mercury 'Next to FAX' > Queue: 11 printable jobs > Server: pid 8507 active > Unspooler: pid 8518 active > Status: cannot open 'laser7%2501' - 'Connection timed out', attempt 1, sleeping > 10 at 16:32:13.129 > Filter_status: (of) FD 1 IS CLOSED!!! LOGIC ERROR IN CODE!!! at 16:23:02.770 This is a problem that was reported, and I am trying to find out if it is OS dependent, OS version dependent, of ifhp (version) dependent. For some reason fstat fails on a socket that has never been closed or had a 'shutdown' done it. This is really strange. Now, unfortunately in such a circumstance you cannot WRITE to it, so you cannot get status, so you have to abort any accounting that you try. I am really really really REALLY interested in seeing if you can get this problem to occur, and if so, occur repeatably. Or if you can determine the circumstances under which it occurs. > > Unfortunately I did not get to the status log before it have rotated. When I > turned the printer back on and 'lprm'ed the bad job the rest of the jobs in the > queue printed just fine. You could have done a 'lpc redo' and rerun the job. I would have been interested in seeing the status. > > 2) On my HPIVsi I often have a problem where a job prints but ifhp hangs out > afterward waiting for the page count status to come back. It misses (or it is > never sent) the feedback from the printer and has to time out. The lpq command > often returns "code = 30010, 'status buffer overflow'" when this condition > happens. On the next attempt to get status it's successful and printing > continues. The printercap is: We are seeing a horrible problem between the TCP/IP support, the PostScript Engine, and status reporting. After the job is sent to the printer, and the status reporting is enabled in PostScript mode, the ifhp filter periodically sends out a ^T (Control T) to prompt the printer to give it status. It also sends a job that would cause it to send back job status. Unfortunately, these do not get processed until all of the current job is done. Then, in a fit of passion, the PostScript interpreter tries to generate all of the output and send it. Thus, we have 'status buffer overflow'. Now, I am puzzled why it is getting buffer overflow. Usually this happens only when LOTS and LOTS of time is taken by the job. > > It's annoying to have to wait for the timeouts between jobs. It doesn't happen > all the time but it's frequent enough, especially on small jobs, to be a pain. > Any ideas? > > - Justus ----------------------------------------------------------------------------- If you need help, send email to [EMAIL PROTECTED] (or lprng-requests or lprng-digest-requests) with the word 'help' in the body. For the impatient, to subscribe to a list with name LIST, send mail to [EMAIL PROTECTED] with: | example: subscribe LIST <mailaddr> | subscribe lprng-digest [EMAIL PROTECTED] unsubscribe LIST <mailaddr> | unsubscribe lprng [EMAIL PROTECTED] If you have major problems, send email to [EMAIL PROTECTED] with the word LPRNGLIST in the SUBJECT line. -----------------------------------------------------------------------------
