Running samba-3.0.3pre2... I'm having some problems with controlling individual print jobs on a cups configuration ( printing = cups ; printcap name = cups ) both as an administrator and as the job creator... When, from the Windows printer window (2000/XP), a job is pointed out (eg. on a paused queue), right-clicked and "Pause Job" or "Cancel Job" is clicked, an error occurs in the bottom right corner of the printer window (status area) saying "Error processing command" Queue controls, such as pause/resume an entire printer works without any problem. ldd against smbd shows that it has been linked against libcups. cups job controls from terminal or via the cups web-interface works fine (logged in as root). An attempt to recreate the problem on an intel machine failed (meaning I could NOT recreate the problem on intel - it all seems to work fine there even with same version of samba)
Created attachment 471 [details] smb.conf from my print server
Created attachment 472 [details] log (level 10) of the problem happening what sticks out to even the untrained eye are the WERR_INVALID_PRINTER_NAME entries in there... I hope they will give you guys a clue as to what's going on!
By the way this is on 31 bit SuSE SLES 8 (service pack 3) for s390
Adding Julius who will be looking at this to the cc: list
ordering problem somehow.... $ grep jobid failed 0014 jobid: 00000060 $ grep s.m.b.p.r.n failed 0168 string: s.m.b.p.r.n...0.0.0.0.0.0.6.9. .T.e.s.t. .P.a.g.e... (gdb) print 0x60 $1 = 96 The INVALID_PRINTER_NAME error comes from _spoolss_setjob() when the print_job_exists() fails. I'll come back to this (but not tonight).
Does this mean that you can actually see print_job_exists failing or COULD this be a cups problem???
Looks like this shares some similarities with bug 1007?
grepping the 1007 log doesn't show this same hex vs reversed decimal phenomenon you noted (though that "0014 jobid:" is certainly wild): $ grep s.m.b.p.r.n 3.0.3pre1.log 0258 string: s.m.b.p.r.n...0.0.0.0.0.0.0.1. .U.n.t.i.t.l.e.d. .- . .N.o.t.e.p.a.d... 02fc string: s.m.b.p.r.n...0.0.0.0.0.0.0.2. .U.n.t.i.t.l.e.d. .- . .N.o.t.e.p.a.d... $ grep jobid: 3.0.3pre1.log 0000 jobid: 00000001 0000 jobid: 00000002 0014 jobid: 00000eae
We were storing the print job using the platform-architecture-endian key, but in certain places comparing it against a byte-swapped version. It now just uses the little-endian version all the time. Checked in at revision 560
originally reported against 3.0.3pre1. Moving back to version to 3.0.2a to remove preX and rcX versions.
originally reported against 3.0.3pre2. Moving back to version to 3.0.2a to remove preX and rcX versions.
sorry for the same, cleaning up the database to prevent unecessary reopens of bugs.
database cleanup