Bug 5540
Summary: | Print jobs goes to the queue and hangs. | ||
---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | jim |
Component: | lpr | Assignee: | Bernhard Rosenkraenzer <bero> |
Status: | CLOSED RAWHIDE | QA Contact: | |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 6.1 | CC: | alexp, cks-rhbugzilla, dirvine, genec, greg, henikoff, jim, lamont, meberle, michael |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | i386 | ||
OS: | Linux | ||
URL: | www.mayfieldj.com | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2000-03-22 02:48:25 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
jim
1999-10-04 23:45:46 UTC
Could you grab the latest lpr RPM from Raw Hide (ftp://ftp.redhat.com/pub/rawhide/i386/) and see if that works for you? It seems this might be more system level than just lpr. Durring a clean install, the installer does not notice printers connected to the local machine (as it did in the RH 6.0 installer), nor does printtool see any devices attached to lp0,lp1,lp2 (where in 6.0 it would for my machine:Dell Dimension XPT P 3 500 mhz ) Oh, there are multiple issues here. If local printing is broken on a new install, add alias parport_lowlevel parport_pc to /etc/conf.modules. This should fix that part of it. I've had a few reports of remote printing breaking, none of which I can duplicate; that's why I'm trying to get feedback on whether the lpr in Raw Hide fixes it for these people. Actually, if you could test the lpr RPMs at (ftp|http)://people.redhat.com/notting/lpr/ That should read 'if you could try these RPMs and let us know if they work for you'. :) should be fixed in the errata release. (0.43-2) I have this problem too and upgrading to lpr-0.44-1 has not fixed it. Another workaround is: $ rm /var/spool/lpd/lp/lock $ /etc/rc.d/init.d/lpd restart ------- Additional Comments From 10/21/99 22:03 ------- Yes -- after upgrading to lpr-0.44-1, I have a similar problem. The lpd is still running, but not responding (I am using a remote printer). lpq says "warning: no daemon running", and has a funny entry "0 bytes" at the end of the queue listing. Stopping and restarting lpd fixes the problem. This only happened once, though, and I don't know how to reproduce it. ------- Additional Comments From 10/27/99 18:20 ------- I have the same problems with RH6.1. I haven't been able to get any of the RPM packages to work, including lpr-0.46-1 and even backing off to lpr-0.35-1 from RH6.0. I have not tried installing the rhs-printfilters and print-tools packages from RH6.0 in addition to the lpr-0.35-1 RPM. What is curious is that we have RH6.0 machines that seem to function best with lpr-0.46-1, while installing lpr-0.46-1 on my RH6.1 box still leaves lpr broken. The symptom (with 0.46-1) is in trying to (remotely) print to an Intel NetPort Express 10/100 print server going to a HP LaserJet 4050, I get the familiar zero length jobs and hung jobs. What are the contents of the files in the queue? Similar problems here. lp0 worked fine in previous installs, When I try to print something, the light on the printer blinks a few times, but nuffin prints. I've tried the EOF fix in the control panel, but there was no dif. nor did 'cat file >/dev/lp0' do anything but blink the light. I've compiled lp0 into the kernel, but no joy there either. How do I make similar settings to those suggested when it's a compiled-in device (alias parport_lowlevel parport_pc) Hmm... what happens if you remove the old files from the queue and restart lpd? Hmm... is there a child lpd process running? If so, what does strace say it is doing? I hav run into the same problem. Printing was working fine until I tried to print an email from netscape 4.61. It didn't work, and the print job just sat in the queue. I restarted the printer daemon (/etc/rc.d/init.d/lpd restart). Then I was told that there was no printer daemon for lp so I started one (lpc start lp). Now the print jobs go to the queue and sit there. Removing all the entries in printcap and re-installing the printer gave no joy. There are two daemon processes running. Strace on highest PID (child lpd) one gives: rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0 open("/dev/lp0", O_WRONLY) = -1 EBUSY (Device or resource busy ) rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0 rt_sigaction(SIGCHLD, NULL, {SIG_IGN}, 8) = 0 nanosleep({32, 0}, {32, 0}) = 0 rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0 open("/dev/lp0", O_WRONLY) = -1 EBUSY (Device or resource busy etc etc strace on lower PID gives: select(20, [4 5], NULL, NULL, NULL and hangs... The paraport alias is in /etc/conf.modules Any ideas? Hmm... what does 'fuser /dev/lp0' say? (and what program is the resulting pid, if there is one?) Does lpr 0.47 (current rawhide) fix this? This APPEARS to be a problem with the permissions and ownership of the lock file in the spool directory. The lpd daemon is running as "daemon", not root. The lock file in /var/spool/lpd/lp is created -rw-r--r-- and uid/gid root:root. If you change the lock file to -rw-rw-r-- and ownership to root:daemon then it appears to be working (so far). -Scott Hi. Not sure all these lp problems are the same thing: One problem is hung jobs killing the queue. I've seen that on RH5.2 The other problem is printing not working, even though the light on the printer blinks. Help, I would like to print occasionally! What information or tests would you like me to run so we can sus this out? Michael We are experiencing the same problem here. We have ~25 Linux workstations, and have had problems with about 5 of them with recurring printing problems. They print over the network to a LaserJet 4050N with a JetDirect card. The applications the users are running include Netscape communicator, Star Office, and enscript. All systems are RH 6.1/i386 When they hang lpq shows a print job with a 0 byte size at the top of the queue. Using lprm removes the line from the lpq output but appends a line at the bottom that only says '0 Bytes' (other job info dissapears). lprm does not unhang the print-queue. Things that I've tried but do not work include: Upgrading to the latest lpr RPM in the errata (lpr-0.48-1), changing the print driver (Generic Postscript / HP Laserjet). The temporary fix is to remove all zero length files in the queue directory (find /var/spool/lpd/lp -size 0 -exec rm -f {} \;) and restart lpd (/etc/rc.d/init.d/lpd restart). This fixes the problem temporaraly. The problem appears to be intermittent, and I have had little sucess finding its cause, however it appears to occur on the same 4-5 systems without affecting the others (it could be that these are the users with the largest volume of print jobs). Hope this information helps... Only root can print on my 6.1 system, other users get the error: lpr:Cannot create /var/spool/lpd/lp/.seq I put general write on /var/spool/lpd/lp/, and then they get: lpr:connect:Connection refused couldn't start daemon I've run /etc/rc.d/init.d/lpd several times, changed the permissions on the lock file to root:daemon; still only root can print. Once root does print something, all the jobs queued by non-root users then print. Everything worked find under 6.0. I was successful in getting lpr to work for non-root users as follows: 1. installed lpr-0.50-3 2. /etc/rc.d/init.d/lpd restart 3. manually deleted everything from the print queue with lprm So far so good (I have one printer attached to /dev/lp0) Count me in on people having the same problem -- 0 byte files appearing and hanging the queue. This is with a stock RH 6.1 distribution. We have an HP 40xx series printer with a network card (that is, the printer has its own IP address). My printcap entry is: lp:\ :sd=/var/spool/lpd/lp:\ :mx#0:\ :rm=rf:\ :rp=raw:\ :lp=: This is fixed in rawhide with the move to LPRng. |