JCW in Batch - more
Deskin, Bob
Bob.Deskin@Cognos.COM
Wed, 21 Oct 1998 09:20:39 -0400
I didn't respond because it didn't seem needed. The only way you can get by
the recompile message about subfile format is to recompile without the
permanent subfile being there. Recompiles and testing should be done in the
same environment as production, including having permanent subfiles. There's
lots of warning about the subfile format change in the Upgrade Guide. QTP
has always been like this. Granted one could argue that it's a loophole, but
...
Bob Deskin
Senior Product Advisor bob.deskin@cognos.com
Cognos Inc. (613) 738-1338 ext 4205 FAX: (613) 228-3149
3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA
> ----------
> From: conleys2@nationwide.com[SMTP:conleys2@nationwide.com]
> Sent: Wednesday, October 21, 1998 8:56 AM
> To: Powerh-l@lists.swau.edu
> Subject: Re: JCW in Batch - more
>
>
> Thanks to all who responded. The correct answer is that at compile time
> it
> is considered to be a warning and can be trapped for. At run time however
> the message is merely informational and no warning or error is issued. I
> consider this to be a HOLE but fortunately it should only occur once when
> you convert to 8.19. Hopefully, you do not have job(s) running after the
> normal SUCCESSFUL? completion! This is one of the few times that I
> never
> received a single reply from COGNOS on this listserv :-(.
>
> For all of you going to 8.19 BEWARE of this relatively minor issue and I
> hope you don't mess up what ever it is that is running!
>
> Regards Steve
>
>
>
>
>
> Stephen A Conley
> 10/20/1998 02:15:54 PM
>
> Sent by: Stephen A Conley
>
> To: Powerh-l@lists.swau.edu
> cc:
> Subject: JCW in Batch - more
>
> Update,
>
> Not entirely true. I only got the correct JCW if I compiled
> interactively. When I ran the same test running the exe using auto= then
> I
> got the same problem as they did in batch.
>
> Any ideas????
>
> [969-100]WCDEV<CONLEYS>#qtp auto=testsfd
>
>
> Executing request 1 ...
> --------------------------------------------------------------------------
> -
> ----
> The requested subfile exists but is not format 8 - won't
> overwrite/append.
> File: NEWSFD
> --------------------------------------------------------------------------
> -
> ----
>
> Records read:
> CHIST-DET 0
>
> Transactions processed: 0
>
> Records processed: Added Updated Unchanged
> Deleted
> NEWSFD 0 0 0
> 0
>
> Finished.
>
> END OF PROGRAM
> [969-100]WCDEV<CONLEYS>#showjcw
> CIERROR = 0
> JCW = 0
> SECURITYANSWER = 0
> NULL = 1
> QTP = 0
> CJCW = 0
> [969-100]WCDEV<CONLEYS>#showvar @jcw
> JCW = 0
> CJCW = 0
> [969-100]WCDEV<CONLEYS>#
> ---------------------- Forwarded by Stephen A Conley/User/NWExternal on
> 10/20/1998 02:22:03 PM ---------------------------
>
>
> Stephen A Conley
> 10/20/1998 02:05:43 PM
>
> Sent by: Stephen A Conley
>
> To: powerh-l@lists.swau.edu
> cc:
> Subject: JCW in Batch
>
>
> Below I have include a sample of a problem they Ran into at my current
> client and I cannot help them because
> I have very little time in the HP3000 world. Apparently if they have a
> problem (any problem) in a job stream
> and they want to check the status once the QTP has completed the JCW does
> not get set (see below). I
> ran a similar test interactively and the JCW gets set just fine. This
> became an issue when they converted
> from 7.29.c6 to 8.19.c1 and had missed some old subfiles. They were
> surprised that the JCW check did
> not pick up on this!
>
> They contacted COGNOS and found little help here. Is this true? If
> someone sees anything inherently wrong with the code below then
> I would really appreciate any advice that you could offer up plus I could
> stand a lesson or two in HP3000 world.
>
> Thanks In Advance for taking the time to read.
>
> Regards Steve
>
> ALSO note that when I ran interactively that I got JCW=16385 and SHOWJCW
> produced JCW=WARN1 which is
> what I would expect for the same format 8 won't overwrite/append warning!
>
>
> JOB JTEMP,MGR.WCDEV,PUB.
> Priority = DS; Inpri = 8; Time = UNLIMITED seconds.
> Job number = #j1164.
> TUE, OCT 20, 1998, 11:14 AM.
> HP3000 Release: C.55.00 User Version: C.55.00
> MPE/iX HP31900 C.05.08 Copyright Hewlett-Packard 1987.
> All rights reserved.
> STREAMED BY HECKLERK,MGR.WCDEV (#S736) ON LDEV# 137
> STREAM DATE: TUE, OCT 20, 1998, 11:14 AM
>
> :FILE PHD=WCPDLC.PH
> :FILE CLMSDB=CLMSDB.AIGDEV
> :FILE TT=SISFBAKE.LUIKARTR
> :
> :QTP INFO="AUTO=QTTT.LUIKARTR"
>
>
>
>
> 10/20/1998 11:14 GATESMCDONALD
> PAGE
> 1
>
> Run:
> Request: 1
>
>
> Executing request 1 ...
>
> --------------------------------------------------------------------------
> -----
>
> The requested subfile exists but is not format 8 - won't
> overwrite/append.
> File: TT
>
> --------------------------------------------------------------------------
> -----
>
>
> Records read:
> SI-SF-CLM-DET 0
>
> Transactions processed: 0
>
> Records processed: Added Updated Unchanged
> Deleted
> TT 0 0 0
> 0
>
> Finished.
>
>
>
>
> END OF PROGRAM
> :
> :SHOWJCW
> CIERROR = 0
> JCW = 0
> SECURITYANSWER = 0
> QTP = 0
> CJCW = 0
> :
> :IF JCW>WARN
> *** EXPRESSION FALSE: COMMANDS IGNORED UNTIL MATCHING ELSEIF/ELSE
> /ENDIF
> : SETJCW JCW FATAL
> :ENDIF
> *** RESUME EXECUTION OF COMMANDS
> :
> :TELL HECKLERK,MGR.WCDEV TEST IS FINISHED.
> :
> :EOJ
> CPU sec. = 1. elapsed min. = 1. TUE, OCT 20, 1998, 11:14 AM.
>
>
>
>
>
>
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
> =
> Subscribe: "subscribe powerh-l" in message body to
> majordomo@lists.swau.edu
> Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
> powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
> This list is closed, thus to post to the list, you must be a subscriber.
>
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
This list is closed, thus to post to the list, you must be a subscriber.