What if PJL is also thrown into the mix.  My perference has always been 
PJL over SNMP, but should I still take PJL over IPP?
Rodney are you listening??? <grin>
Thanks,
Charles
I would chose PJL last.  The reason being no sharing between systems 
until there are no RDY files.  A single TCP/IP port, so if the printer 
stops talking for some reason, you have no way to determine why...just 
looks like the writer is hung.
SNMP is better because it allows sharing between files (if you specify 
*IBMSHRCNN on the USRDFNOPT attribute of the printer device).  SNMP uses 
two ports.  A data port, and a status port (actual SNMP).  This allows 
you to know why the printer writer has stopped processing the spooled 
file (like out of paper condition, printer offline, etc.).
IPP has the ability to send the data encrypted, thus if security is an 
issue, and you are sending to an IBM IPP server or CUPS server, you can 
send the data without fear of losing confidential data.  IPP also 
supports finishing operations.  SNMP can support finishing if you use my 
latest version of the user data transform TSPRWPR (and modify the source 
to match the PJL commands for the printer with the finisher...unless you 
have an IBM 1145 or 1130).
IPP might be better if you are sending site to site.  IPP can be not as 
good if the IPP implementation of the printer is limited.  IPP expert 
here has run into a number of situations that can be performance 
limiting (or even won't work) due to the implementation/interpretation.
SNMP and IPP are both good choices. I use SNMP
Charles Wilt wrote:
-- 
Rodney A Johnson 
Technical Team Lead for AS/400 Spool
Dept GJC
IBM   Rochester, Minnesota
The contents of this message express only the sender's opinion.
This message does not necessarily reflect the policy or views of
my employer, IBM.  All responsibility for the statements
made in this Usenet posting resides solely and completely with the
sender.
Rodney,
Just curious, does this also mean that a spool file is deleted only when the 
print job (spool file) has completed printing. We use IPDS printers mainly 
because we want to be absolutely sure that even if a printer, for whatever 
reason, has to be turned off the print job (spool file) can still be 
resumed. In other words, (if I remember correctly) in the past when we still 
used the PCL/PJL driver, if a printer lost power, sometimes print jobs just 
disappeared (cached in printer memory). Does the SNMP driver do the job of 
protecting us from this situation as well as IPDS?
TIA,
Kaj
Thank you for the excellent information.
One thing I found with SNMP was that I couldn't use it to both printers 
on a multiple port server.  So I ended up going IPP.  Actually, in this 
case, I don't believe either printer supported SNMP anyway.
Is this the location of the latest TSPRWPR:
http://www-1.ibm.com/support/docview.wss?
uid=nas1584a2dee8678698386256624004b527e&rs=110
Also, you suggest using TSPRWPR with SNMP, the document above doesn't 
mention this possibility.  I assume all that is need is to specify this 
program on the USRDTATFM parm of the CRTDEVPRT command.
Thank again,
Charles Wilt
In article <41646...@news3.prserv.net>, rjoh...@rchland.ibm.com 
says...
On the print server (Axis 5600+) there's an option to view a log of the 
jobs processed.
It's interesting that LPD jobs show the user name of the iSeries user 
whose spool file it was.
But the IPP jobs show an anonymous.
Is there anyway to change that?
How would SNMP jobs show?
Thanks,
Charles
In article <MPG.1bcf0021a...@news.easynews.com>, 
cw...@meaa.mea.com says...
SNMP waits only for acknowledgement from the TCP/IP layer that the last 
packet of data was successfully received by the NIC.  Thus if the 
printer is shut off at the right time, you still could lose the file. 
Unfortunately IPP also has the same limitation, and in fact might be 
worse, in that once the whole file is there iSeries assumes it is 
printed, where as SNMP, it would only be the last few pages or buffers 
of data that may not get printed and are lost.
The PJL method waits for confirmation that the last page printed.  
Unfortunately, many printer implementations send back an "early" print 
complete (that is it says last page printed before the page really has 
been printed).
IPDS is probably the only method that make sure the last page is printed 
(although I have heard of problems in this area, and don't know if all 
those bugs have been fixed...some at software level others were at 
printer microcode level).
Charles,
I could try to change that.  What release of the iSeries are you
currently using?
Thanks,
Dave
I'm on v5r2 right now.
Charles
In article <41655...@news3.prserv.net>, dbmu...@hotmail.com says...
Charles Wilt wrote:
>Rodney,
>
>Thank you for the excellent information.
>
>One thing I found with SNMP was that I couldn't use it to both printers 
>on a multiple port server.  So I ended up going IPP.  Actually, in this 
>case, I don't believe either printer supported SNMP anyway.
>  
>
 >> Yes there is a limitation on multi-port servers where SNMP will only 
work to the printer connected on the first port.  For multi-port 
servers, IPP would likely be the better option.
>Is this the location of the latest TSPRWPR:
>http://www-1.ibm.com/support/docview.wss?
>uid=nas1584a2dee8678698386256624004b527e&rs=110
>
>http://tinyurl.com/46k8m
>
>  
>
 >>The version I am referring to is not available on the web.  It is a 
newer version that works on V5R2 and up that I completed a few months 
back.  I send to whoever is interested upon request.  The source will be 
in the release following V5R3 (in qusrtool).
>Also, you suggest using TSPRWPR with SNMP, the document above doesn't 
>mention this possibility.  I assume all that is need is to specify this 
>program on the USRDTATFM parm of the CRTDEVPRT command.
>
>  
>
 >>Part of the modification was to make it work with printer device 
configurations as well as with the remote writer.  When used with a 
printer writer, it still allows the printer driver to handle the page 
range parts.  In that specific case it just prepends and appends the 
appropriate PJL commands to start and stop stapling.  In theory, TSPRWPR 
would then work with the SNMP, telnet 5250 printing, and LPR (remote 
writer) to provide stapling and hole punching.  Course the version I 
send out is written to only work with IBM 1145 and 1130 printers.  It is 
also coded to supposedly work on the 4332 and 4324 printers, but was not 
able to test and confirm.  Hopefully the code is written in a manner 
that is easy to modify to match other printer finishers.
I'd be interested in getting the new version of TSPRWPR.
The email address on for this post is valid.
Thanks again for your help,
Charles
In article <41658...@news3.prserv.net>, rjoh...@rchland.ibm.com 
says...