printing robustness issue

[Dave Jones]

I'm using VMS samba (2.2.8) to service about 10 print queues to about 100 PCs.
Every now and then, something goes wrong with a print operation and the PC and
the SMBD process end up in a loop chatting about their predicament.  The SMBD
process will show 2000-3000 I/O's per second.  I haven't been able to track
down the exact set of circumstances that causes this loop, has anyone else seen
this in a way they could reproduce reliably?

If you kill the smbd process, the spooler process on the PC (under XP pro),
immediately reconnects to the server and appears to want to retry sending
the print job (I make the new smbd start with a higher debug level to see what
is going on).  The new smbd process doesn't recognize the printer handle and
apparently sends a WERR_BADFID error response, to which the PC reacts by
repeating the attempt a short time later.  The net ammount of I/O from this
loop is relatively low, but it never stops until you reboot the machine.

-----------------------------------------------------------------------------
David L. Jones               |      Phone:    (614) 292-6929
Ohio State University        |      Internet:
140 W. 19th St. Rm. 231a     |               jonesd at er6s1.eng.ohio-state.edu
Columbus, OH 43210           |               vman+ at osu.edu

Disclaimer: I'm looking for marbles all day long.