decprint: error getting a port

[suttorpa@hotmail.com]

Hello again, another print issue, new server.

So am configuring printing on a new server to Xerox C500 (same as the old server) using dec print, libraries, and setup files. But for some odd reason when I send a job to the queue it processes for a moment and then disappears, the printer does not receive anything.

Some details:

write sys$output f$getsyi( "version")
V8.4-2

product show product dcps
------------------------------------ ----------- ---------
PRODUCT KIT TYPE STATE
------------------------------------ ----------- ---------
VSI I64VMS DCPS V2.8 Full LP Installed
------------------------------------ ----------- ---------

I have enabled and started LPR/LPD service using @sys$startup:tcpip$config
And have configured the printer using lprsetup.exe:

$run sys$system:tcpip$lprsetup.exe
TCPIP Printer Setup Program
Command < add delete view help exit >: add
Adding printer entry, type '?' for help.
Enter printer name to add : EBBC500Spare
Enter the FULL name of one of the following printer types:
remote local : remote
Enter printer synonym:
Enter full file specification for spool directory
SPOOLER DIRECTORY 'sd' : [TCPIP$LPD_ROOT:[EBBC500SPARE]] ?
Set LPD PrintServer extensions flag 'ps' [] ?
Set remote system name 'rm' [] ? EBBC500SPARE
Set remote system portnumber 'rt' [515] ?
Set remote system printer name 'rp' [] ? lp
Set printer error log file 'lf' [/TCPIP$LPD_ROOT/000000/EBBC500SPARE.LOG] ?
Enter the name of the printcap symbol you wish to modify. Other
valid entry is : 'q' to quit (no more changes)
The names of the printcap symbols are:
sd for the printer spool directory
lf for the printer error log file
lp for the name of the local printer
ps for the LPD PrintServer extensions flag
rm for the name of the remote host
rp for the name of the remote printer
rt for the port number of the remote host
fm for the printer form field
pa for the /PASSALL flag
nd for the /NODELETE flag
cr for the cr flag
sn for the setup NoLF flag
p1-p8 for the /PARAMETER=(p1,...,p8) field
Enter symbol name: q
Symbol type value
------ ---- -----
Error log file : lf STR /TCPIP$LPD_ROOT/000000/EBBC500SPARE.LOG
Printer Queue : lp STR EBBC500SPARE
PS extensions flag: ps STR
Remote Host : rm STR ebbc500SPARE
Remote Printer : rp STR lp
Spool Directory : sd STR /TCPIP$LPD_ROOT/EBBC500SPARE
Remote Port : rt INT 515

Are these the final values for printer EBBC500 ? [y] y
Adding comments to printcap file for new printer, type '?' for help.
Do you want to add comments to the printcap file [n] ? :
Do you want the queue to default to print flag pages [y] : N
Do you want this procedure to start the queue [y] :
Creating execution queue: EBBC500SPARE
Updating TCPIP$LPD_SYSTARTUP.COM
Updating TCPIP$LPD_SYSHUTDOWN.COM

*************************************************
* TCPIP$LPD_SYSTARTUP.COM, the printcap file *
* and TCPIP$LPD_SYSHUTDOWN.COM *
* have been updated for this printer *
* *
* Set up activity is complete for this printer *
*************************************************
Command < add delete view help exit >: exit

I generally test prints with .lis files, because that is normally all we print from OpenVMS; but whatever I send, the job goes to the queue for a moment (if I stop the queue, I will see the job pending), but then the job disappears completely.

an example of our print command:
print rp1ebb625.lis /que=EBBC500SPARE

When I look at the log file for this printer I see an error:

%%%%%%%%%%%% 13-JUL-2022 14:42:31.45 %%%%%%%%%%%%

%TCPIP-I-LPD_SYMBRUN, symbiont is running the queue EBBC500SPARE


TCPIP LPD configuration data:
LPD Spooler Directory : TCPIP$LPD_ROOT:[000000]
Retry-Interval : 0 00:05:00.00
Retry-Maximum : 0 01:00:00.00
Idle-Timeout : 0 00:05:00.00
Inbound-Queues-Per-Node : 1
Utility-Queues-Per-Node : 0
Receiver-Debug : 0
Symbiont-Debug : 0
Utilities-Debug : 0
Loop-Max : 0
Droptime : 0
Probetime : 0
PS-Extensions : LPS
Stream-Passall : FALSE
VMS-Flagpages : FALSE
Persistent-Server : FALSE
Keepalive : FALSE
Synchronize-All-Jobs : TRUE
Setup-NoLF : FALSE
1st-VFC-Prefix-Special : FALSE
Setup-Buffer-Size : 1024
Broadcast-Msg : TRUE

PCB for printer EBBC500SPARE
RT: 515
LP: EBBC500SPARE
RM: EBBC500SPARE
RP: lp
SD: /TCPIP$LPD_ROOT/EBBC500SPARE
LF: /TCPIP$LPD_ROOT/000000/EBBC500SPARE.LOG
BP: EBBC500SPARE
PA: FALSE
ND: FALSE
CR: FALSE
SN: FALSE


%%%%%%%%%%%% 13-JUL-2022 14:42:43.25 %%%%%%%%%%%%

%TCPIP-F-LPD_BADPORT, error getting a port
-TCPIP-W-EAI_AGAIN, temporary failure


%%%%%%%%%%%% 13-JUL-2022 14:43:32.98 %%%%%%%%%%%%

%TCPIP-F-LPD_BADPORT, error getting a port
-TCPIP-W-EAI_AGAIN, temporary failure

in show dev I can see lpd configured, it looks correct:

Device_socket Type Local Remote Service Host
bg25598 STREAM 515 0 LPD *

LPD is enabled in show services..

TCPIP hosts file is configured, and I am able to ping the printer either by hostname or ip

I tried printing with my own profile, and as System.. I am kind of run out of ideas..

I was hoping someone here could help?

Thank you.

[sms]

Code: Select all

   I still do almost no printing from VMS, and I haven't touched LPR/LPD
in decades, so I know nothing, but...

> So am configuring printing on a new server to Xerox C500 (same as the
> old server) using dec print, libraries, and setup files.

   You've compared all this stuff between the two VMS systems?

> write sys$output f$getsyi( "version")
> V8.4-2

   "tcpip show version" might reveal more.  (Amazing that there's no
simple command (like, say, "show version") for the OS version, isn't it?)

ITS $ tcpip show version

  HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.7 - ECO 5
  on an HP rx2600  (1.50GHz/6.0MB) running OpenVMS V8.4-2L3

> VSI I64VMS DCPS V2.8 Full LP Installed

   Mine's V2.8-1, but I wouldn't bet that that matters.

> I have enabled and started LPR/LPD service [...]

   Why?  Why use LPR/LPD?  Acording to:
      https://www.office.xerox.com/latest/VC5SS-01U.PDF
your printer can use the following "Network Protocols":

      TCP/IP: HTTP/HTTPS, Internet Printing Protocol, LPR/LPD,
      Raw Socket Printing/Port 9100, IPv4/IPv6, WSD LDAP
      Bonjour/AirPrint

   It also says, "Most protocols not in use can be disabled".  Can you
Telnet to your printer (remote host) at port 515?

      telnet EBBC500SPARE 515

(Is "EBBC500SPARE" resolved to the printer's actual IP address?)

>  TCPIP hosts file is configured, and I am able to ping the printer
> either by hostname or ip [address]

   _Which_ "hostname"?  "ping" is one thing; listening at port 515
(LPD/LPR) is another.

   DCPS should be able to use the raw-9100 scheme, which would eliminate
all the LPR/LPD software (at both ends).  Around here, with my cheap,
old, needs-a-cartridge Dell 1720dn, for example:

ITS $ show queue sys$print
Printer queue DELL_DCPS, idle, on ITS::"IP_RawTCP/dell:9100", 
mounted form DCPS$DEFAULT (stock=DEFAULT)

      show queue <wherever_you're_sending_things>

> in show dev I can see lpd configured, it looks correct:

   So, someone could send a print job _to_ your VMS system?  Do you
care?  What _I_ would want is DCPS on the VMS system to be able use the
LPR/LPD protocol to talk to the LPD (server) on the printer.  I don't
see what an LPD (server) on the VMS system buys you.

> %TCPIP-F-LPD_BADPORT, error getting a port

      pipe tcpip netstat -an | search sys$pipe 515

   My Web search for that message didn't find much, but (knowing what
little I do) a search for "lpr lpd port" did turn up a few places which
claim that RFC 1179 says, "The port on which a line printer daemon
listens is 515. The source port must be in the range 721 to 731,
inclusive."  That seems pretty restrictive to me.  Is it possible that
all those "must be" source ports are busy on your system?

   Does the DCPS documentation say anything about it?

[suttorpa@hotmail.com]

Thank you for the response!

let me go over your questions... I'll try to clear out text to clean up this post:

> So am configuring printing on a new server to Xerox C500 (same as the
> old server) using dec print, libraries, and setup files.

You've compared all this stuff between the two VMS systems?

I recently started with this company, and I am no print expert myself. They have been struggling with updating their printer fleet for a while (years - some of the printers here are 20 years old); they can print to the new devices but all parameters are removed. I took over the project and was working on a dev/backup server. In their notes they were using LPD; so I continued from there. I managed to figure out how to send parameters to the new printers which was to use libraries/setup files and PCL code. Now I am trying to replicate it on the live server. But, well.. it is not even sending the job to the printer, as mentioned before the jobs just disappear. Every setting that I changed personally I have gone over and they match server to server. The only difference is maybe the socket number that I do not believe would matter... I must be missing something else

ITS $ tcpip show version

HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.7 - ECO 5
on an HP rx2600 (1.50GHz/6.0MB) running OpenVMS V8.4-2L3

> VSI I64VMS DCPS V2.8 Full LP Installed

show prod hist gives me this:
VSI I64VMS DCPS V2.8 Full LP Install Val 21-DEC-2017

> I have enabled and started LPR/LPD service [...]

Why? Why use LPR/LPD? According to:
https://www.office.xerox.com/latest/VC5SS-01U.PDF
your printer can use the following "Network Protocols":

TCP/IP: HTTP/HTTPS, Internet Printing Protocol, LPR/LPD,
Raw Socket Printing/Port 9100, IPv4/IPv6, WSD LDAP
Bonjour/AirPrint

I was following the setup of previous printers on the system, they all appear to be LPD, and also to match the printer on the dev server. I need LPR/LPD running to use the lprsetup.exe to configure the printer

It also says, "Most protocols not in use can be disabled". Can you
Telnet to your printer (remote host) at port 515?

telnet EBBC500SPARE 515

(Is "EBBC500SPARE" resolved to the printer's actual IP address?)

It does, at least it appears to work:
telnet EBBC500SPARE 515
%TELNET-I-TRYING, Trying ... 172.23.13.195
%TELNET-I-SESSION, Session 01, host EBBC500SPARE, port 515
-TELNET-I-ESCAPE, Escape character is ^]
%TELNET-S-REMCLOSED, Remote connection closed
-TELNET-I-SESSION, Session 01, host EBBC500SPARE, port 515

I am not exactly sure how to make use of telnet printing... I found an article here:
https://support.hpe.com/hpesc/public/do ... cale=en_US

I will work through that to test it.

> TCPIP hosts file is configured, and I am able to ping the printer
> either by hostname or ip [address]

_Which_ "hostname"? "ping" is one thing; listening at port 515
(LPD/LPR) is another.

Just to be sure I changed the hostname/device name on the printer to match EBBC500SPARE.
PING EBBC500SPARE (172.23.13.195): 56 data bytes
64 bytes from 172.23.13.195: icmp_seq=0 ttl=64 time=1 ms
64 bytes from 172.23.13.195: icmp_seq=1 ttl=64 time=0 ms
64 bytes from 172.23.13.195: icmp_seq=2 ttl=64 time=365 ms
64 bytes from 172.23.13.195: icmp_seq=3 ttl=64 time=0 ms

DCPS should be able to use the raw-9100 scheme, which would eliminate
all the LPR/LPD software (at both ends). Around here, with my cheap,
old, needs-a-cartridge Dell 1720dn, for example:

ITS $ show queue sys$print
Printer queue DELL_DCPS, idle, on ITS::"IP_RawTCP/dell:9100",
mounted form DCPS$DEFAULT (stock=DEFAULT)

show queue <wherever_you're_sending_things>

SHOW QUEUE /FULL EBBC500SPARE
Server queue EBBC500SPARE, idle, on EBB017::, mounted form DEFAULT
/BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT) /LIBRARY=DCPS$DEVCTL Lowercase /OWNER=[SYSTEM] /PROCESSOR=TCPIP$LPD_SMB
/PROTECTION=(S:M,O:D,G:R,W:S)

> in show dev I can see lpd configured, it looks correct:

So, someone could send a print job _to_ your VMS system? Do you
care? What _I_ would want is DCPS on the VMS system to be able use the
LPR/LPD protocol to talk to the LPD (server) on the printer. I don't
see what an LPD (server) on the VMS system buys you.

It just lets me configure the printer, its the only instructions I have to work with. If you have a different method i could try it

> %TCPIP-F-LPD_BADPORT, error getting a port

pipe tcpip netstat -an | search sys$pipe 515

pipe tcpip netstat -an | search sys$pipe 515
tcp 0 0 *.515 *.* LISTEN

My Web search for that message didn't find much, but (knowing what
little I do) a search for "lpr lpd port" did turn up a few places which
claim that RFC 1179 says, "The port on which a line printer daemon
listens is 515. The source port must be in the range 721 to 731,
inclusive." That seems pretty restrictive to me. Is it possible that
all those "must be" source ports are busy on your system?

Does the DCPS documentation say anything about it?

It is possible I guess, there are about 14 print queues on this server and 6 generic queues. Printing is not all that busy though, you would be lucky to see 2 print jobs in the queue at any given time. Any large print jobs are done after hours. Data our users print have been formatted into reports that they can view through HTML and print via windows, so most printing is done there. we are just stuck with a few older reports that still print through OpenVMS

[sms]

Code: Select all

> I was following the setup of previous printers [...]

   I assumed that.  But if I were having LPR/LPD problems, then I might
investigate a different/simpler scheme.

> I need LPR/LPD running to use the lprsetup.exe to configure the
> printer

   Ok.  But not using LPR/LPD should avoid all that.

> %TELNET-I-TRYING, Trying ... 172.23.13.195
> %TELNET-I-SESSION, Session 01, host EBBC500SPARE, port 515
> -TELNET-I-ESCAPE, Escape character is ^]

   Ok.  The printer seems receptive, so the problem seems to be on the
VMS side.

>  I am not exactly sure how to make use of telnet printing... [...]

   I wouldn't worry about it.  That use of Telnet was simply a test that
the printer was listing on the LPR/LPD port.  (Other/simpler programs
can do the same kind of test, but Telnet is more widely available.)

> pipe tcpip netstat -an | search sys$pipe 515
> tcp 0 0 *.515 *.* LISTEN

   That says that the LPD server on the VMS system is working, but I
claim that you don't care much about that.  You want to send jobs _from_
the VMS system, not accept them from elsewhere, which is what that
diagnoses.

   The fact that you're running current VSI software suggests that you
have a support agreement with VSI.  In such a case, I'd exercise it,
starting with asking what the %TCPIP-F-LPD_BADPORT message actually
means.

> SHOW QUEUE /FULL EBBC500SPARE
> Server queue EBBC500SPARE, idle, on EBB017::, mounted form DEFAULT
> [...]

   Your arrangement is more complicated than mine.  I have a "Printer
queue" with a complex "device" spec ("IP_RawTCP/dell:9100").

Printer queue DELL_DCPS, idle, on ITS::"IP_RawTCP/dell:9100",

   You have a "Server queue" with /PROCESSOR=TCPIP$LPD_SMB.  Not
something I've done.

   The TCPIP docs include "Chapter 24. Setting Up and Managing the
LPR/LPD Print Service", which explains some "Debug" capabilities
("24.12. Solving LPD Problems").  Might be informative.

https://docs.vmssoftware.com/docs/vsi-tcpip-services-for-openvms-management.pdf

   Safe to assume that you've been exercising the TCPIP$LPD_STARTUP.COM
and TCPIP$LPD_SHUTDOWN.COM scripts (to no avail)?

[suttorpa@hotmail.com]

simpler scheme? TELNETSYM? I guess I could give it a shot, it does not "look" simpler , I just felt I would be starting from square one. I'll consider it.

Our support agreement is mixed... VSI/HP, we have certain functionality that is unavailable with full VSI licensing, or at least is not expected to be available until 2023... so we are in a bit of a pickle in terms of a "support contact."

I am not sure how you created "IP_RawTCP/dell:9100".. I was just following the LPD printer setup procedure as per the tcpip manual you sent, or the above notes.

in that TCP document you sent in the previous post, at the end of Chpt.24 it had configuration changes for symbiont-debug: 8. basically, to increase the log file details.. would you happen to know how to configure that?

I tried changing it in TCPIP$LPD_IN_Servername_1.LOG

Code: Select all

%%%%%%%%%%%%                   13-JUL-2022 09:34:06.29  %%%%%%%%%%%%

%TCPIP-I-LPD_LOGSUC, using log file TCPIP$LPD_ROOT:[000000]TCPIP$LPD_IN_Servername_1.LOG


%%%%%%%%%%%%                   13-JUL-2022 09:34:06.29  %%%%%%%%%%%%

%TCPIP-I-LPD_SYMBRUN, symbiont is running the queue TCPIP$LPD_IN_Servername_1


TCPIP LPD configuration data:
LPD Spooler Directory         : TCPIP$LPD_ROOT:[000000]
Retry-Interval                : 0 00:05:00.00
Retry-Maximum                 : 0 01:00:00.00
Idle-Timeout                  : 0 00:05:00.00
Inbound-Queues-Per-Node       : 1
Utility-Queues-Per-Node       : 0
Receiver-Debug                : 8
Symbiont-Debug                : 8
Utilities-Debug               : 0
Loop-Max                      : 0
Droptime                      : 0
Probetime                     : 0
PS-Extensions                 : LPS
Stream-Passall                : FALSE
VMS-Flagpages                 : FALSE
Persistent-Server             : FALSE
Keepalive                     : FALSE
Synchronize-All-Jobs          : TRUE
Setup-NoLF                    : FALSE
1st-VFC-Prefix-Special        : FALSE
Setup-Buffer-Size             : 1024
Broadcast-Msg                 : TRUE

PCB for printer
PA: FALSE
ND: FALSE
CR: FALSE
SN: FALSE

I actually have not tried the TCPIP$LPD_STARTUP.COM and TCPIP$LPD_SHUTDOWN.COM scripts (it might be as simple as requiring the service to be restarted), just testing it out... and both scripts run without issue.

when I try printing again the job is getting stuck in the queue with an error message:

Server queue EBBC500SPARE, idle, on Server::, mounted form DEFAULT

Entry Jobname Username Size Status
----- ------- -------- --------- ------
999 rp1ebb625 SUTTORP 2.00KB Retained on error
%NONAME-F-NOMSG, Message number 07649F5C

which is better than before, but still I cannot find any information on that error, in the logs, the error is the same.

Added in 3 hours 5 minutes 36 seconds:
You know what...

I decided to just delete the queue entirely and recreated it.. changing absolutely nothing, in my setup process.. ran the same print statement of the same file..

and guess what?

it worked...

Thank you for bearing with me sms, I truly appreciate it...

[sms]

Code: Select all

> simpler scheme? TELNETSYM? [...]

   No, what HP called "Raw Socket Printing", and DCPS calls "IP_RawTCP".

> Our support agreement is mixed... [...]

  I'd ask one or both of those organizations.  You might get an answer
from someone, whether you deserve one or not.

> I am not sure how you created "IP_RawTCP/dell:9100".. [...]

   I can't find any record of what I actually did, so I'd guess that I
used INITIALIZE /QUEUE with all (most?) of the stuff shown by SHOW QUEUE
/FULL.

   However, my quick search of SYS$MANAGER"*.COM for "9100" found some
interesting stuff in DCPS$STARTUP.COM, which shows examples of set-ups
using various comm schemes, including "IP_RawTCP" (port 9100).  (I
wouldn't bet on AppleTalk still working, but it probably did long ago.)

> [...] it had configuration changes for symbiont-debug: 8. basically,
> to increase the log file details.. would you happen to know how to
> configure that?

   I'd try adding something like:
      Symbiont-Debug: 8
to my TCPIP$LPD_ROOT:[000000]TCPIP$LPD.CONF file (if I had one).  The
nearby template file, TCPIP$LPD_CONF.TEMPLATE, says:

!    2. The field names for each supported field are listed below. [...]

which sounds like a lie, given that the TCPIP docs mention ones like
"Symbiont-Debug" which I don't see in my template file, but other info
there might be helpful.

> I tried changing it in TCPIP$LPD_IN_Servername_1.LOG

   The ".LOG" files are for _you_ to read, not for the programs.  The
programs read the ".CONF" files.

> %NONAME-F-NOMSG, Message number 07649F5C

   You'll like this:

ITS $ @ utility:sysmsg %x07649F5C
%TCPIP-F-LPD_BADPORT, error getting a port

Sure would be nice to know what that means.

      http://antinode.info/ftp/misc/sysmsg.com
      
         You might also try asking in the comp.os.vms news group.  The
audience there is bigger.  (The penalty being that the discussion might
soon degenerate into noise about the future of COBOL, or the merits of
different C language standards, or whatever else is disturbing someone
this week.)

[volkerhalle]

Cross-posted in HPE OpenVMS forum:

https://community.hpe.com/t5/Operating- ... 51#M105660

Solution:

I decided to delete the queue and recreated it. the only thing i changed was for rp=lp, i changed to rp=lp:\

Volker.

[suttorpa@hotmail.com]

yeah.. I thought later that might of fixed it. It might have been that I set up the hosts file after I ran lprsetup.exe, and that after redoing it with the host file created maybe then it was able to properly set up the printer? I don't know. I find it odd because, on my dev server, I just entered lp, the :\ was added automagically. Only after comparing inside "lprsetup.exe > view" did I notice the discrepancy. this whole thing just gave me a big headache. lesson learned.. when you feel you did everything right. delete it all and try again before harassing people in forums

[sms]

Code: Select all

> [...] lesson learned.. when you feel you did everything right. delete
> it all and try again before harassing people in forums ;)

   That's one lesson.  Or, when someone asks:

>    You've compared all this stuff between the two VMS systems?

You might actually compare the stuff carefully.

   But it's good to hear that you got it working.  I'd still suggest
investigating the IP_RawTCP scheme, when you get bored.