|
We don't like to leave something in MSGW (at least not a user-run job), because users have been known to answer the message on an update program and screw something up, leaving us to clean up the mess. I'd just as soon do my own error handling and save them (and us) the trouble. I take extra care when writing RPG code, for example, to try to cover all the bases, error-wise (for example, I never assume the record isn't there already before a WRITE - I do a SETLL first and only WRITE if NOT %EQUAL). But some of this stuff goes back to 1992 and isn't written as cleanly. I'm converting them to ILE and RPG 4 whenever I can, and adding my stuff where I can, but we have over 1,000 programs out there. This kind of error handling isn't what I do all the time - but this particular code comes from an FTP job which, if it fails, has to be restarted from scratch, since it will just overwrite the work file that it didn't get the first time. Besides, it will run under QSYSOPR - not a user who feels compelled to answer run-time error messages. In case you were curious, the MONMSG after the SNDPAGMSG is there because we only copied the command object to the development box so the programs will compile - there is no CPP for the command on development). Francis Lapeyre IS Dept. Programmer/Analyst Stewart Enterprises, Inc. E-mail: flapeyre@xxxxxxxx -----Original Message----- From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx Sent: Tuesday, January 31, 2006 4:27 PM To: Midrange Systems Technical Discussion Subject: RE: MONMSG CPF2130 was: CL program We too use LXI. I'd just as soon leave the job in a message wait and have their other monitors page that the job is stuck in a MSGW. That way manual intervention could be done. Like I said before, it's often easier to fix a problem and issue an R to retry then restart a job. Looks like we're going to have to agree to disagree. Perhaps it looks nasty in a review at your shop to have jobs stuck in MSGW. Me, I have the buggers page the VP of IS hoping he comes up with a plan to deal with these situations. Pages him every 15 minutes until the message is gone. Rob Berendt -- Group Dekko Services, LLC Dept 01.073 PO Box 2000 Dock 108 6928N 400E Kendallville, IN 46755 http://www.dekko.com "Lapeyre, Francis" <FLAPEYRE@xxxxxxxx> Sent by: midrange-l-bounces@xxxxxxxxxxxx 01/31/2006 04:53 PM Please respond to Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx> To "'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx> cc Fax to Subject RE: MONMSG CPF2130 was: CL program PGM DCL VAR(&JOB) TYPE(*CHAR) LEN(10) DCL VAR(&NBR) TYPE(*CHAR) LEN(6) DCL VAR(&USER) TYPE(*CHAR) LEN(10) /* Trap all unmonitored errors. Page the on-call programmer. */ MONMSG MSGID(CPF0000) EXEC(GOTO CMDLBL(ERROR)) /* Get job attributes to put in pager message in case of a problem. */ RTVJOBA JOB(&JOB) USER(&USER) NBR(&NBR) ... Normal processing ... RETURN /* Error processing (general). */ ERROR: LXIPAG/SNDPAGMSG MSG('Job' *BCAT &NBR *TCAT '/' *TCAT + &USER *TCAT '/' *TCAT &USER *BCAT 'had an + unexpected error. See job log and dumps + for job.') TOUSERS((PROGRAMMER AR)) MONMSG MSGID(CPF0000) DMPCLPGM ENDPGM -----Original Message----- From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx Sent: Tuesday, January 31, 2006 3:17 PM To: Midrange Systems Technical Discussion Subject: RE: MONMSG CPF2130 was: CL program Ok, so "maybe" you got 1 & 3 covered. Don't know how that translates into "None", but ok. In a case like this, I'd like to get the hard halt if the CRTDUPOBJ fails so that I can shoot the developer who bypassed Turnover (we, too, use that software. Guess which vendor discovered the V5R4 ftp anomaly the hard way? ((but quickly had a fix)) ). Or handle manually any of the myriad of other possible situations, like the RI constraints, etc. Rob Berendt -- Group Dekko Services, LLC Dept 01.073 PO Box 2000 Dock 108 6928N 400E Kendallville, IN 46755 http://www.dekko.com "Lapeyre, Francis" <FLAPEYRE@xxxxxxxx> Sent by: midrange-l-bounces@xxxxxxxxxxxx 01/31/2006 03:56 PM Please respond to Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx> To "'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx> cc Fax to Subject RE: MONMSG CPF2130 was: CL program None of those things will happen, because our production box is locked down to the point where not even programmers have command-line authority. All programming is done on a development machine and things get sent over with a change management package (Turnover). &COMMONLIB is determined at run time by a RTVOBJD on a file and returning the library it's in. -----Original Message----- From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx Sent: Tuesday, January 31, 2006 1:57 PM To: Midrange Systems Technical Discussion Subject: RE: MONMSG CPF2130 was: CL program You got the CPF3142 on the CLRPFM, not on the CRTDUPOBJ. How else could this have failed on the CRTDUPOBJ, since you use adopted authority? Well I can only think of a few dozen ways. 1 - Someone recompiled the program and took away the adopted authority. 2 - File got corrupted. Yes this even happens with DB2-UDB for the iSeries albeit damn rare. 3 - File did not exist in &COMMONLIB. 4 - &COMMONLIB got deleted. 5 - Someone added a trigger or referential constraint to the file and it doesn't like being duped into QTEMP. ... and the list goes on... Rob Berendt -- Group Dekko Services, LLC Dept 01.073 PO Box 2000 Dock 108 6928N 400E Kendallville, IN 46755 http://www.dekko.com "Lapeyre, Francis" <FLAPEYRE@xxxxxxxx> Sent by: midrange-l-bounces@xxxxxxxxxxxx 01/31/2006 02:47 PM Please respond to Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx> To "'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx> cc Fax to Subject RE: MONMSG CPF2130 was: CL program Yes there is (the global MONMSG CPF0000). Anyway, if you get a CPF3142, the file isn't in QTEMP, so how could a CRTDUPOBJ fail? (All our programs use adopted authority, so that won't be a problem). -----Original Message----- From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx Sent: Tuesday, January 31, 2006 12:20 PM To: Midrange Systems Technical Discussion Subject: RE: MONMSG CPF2130 was: CL program Like that code... But there is no checking to see what happens if the CRTDUPOBJ fails. Which seems to happen a lot here. And not because it already exists in the TOLIB. Rob Berendt -- Group Dekko Services, LLC Dept 01.073 PO Box 2000 Dock 108 6928N 400E Kendallville, IN 46755 http://www.dekko.com "Lapeyre, Francis" <FLAPEYRE@xxxxxxxx> Sent by: midrange-l-bounces@xxxxxxxxxxxx 01/31/2006 11:34 AM Please respond to Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx> To "'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx> cc Fax to Subject RE: MONMSG CPF2130 was: CL program Here is what I am doing in a program (duplicating a physical and an associated logical into QTEMP): /* Clear temporary decrypted file. Create (as a clone of AROCC) */ /* if not found. Also create AROCCL6 logical. */ CLRPFM FILE(QTEMP/AROCC) MONMSG MSGID(CPF3142) EXEC(CRTDUPOBJ OBJ(AROCC) + FROMLIB(&COMMONLIB) OBJTYPE(*FILE) + TOLIB(QTEMP)) CHKOBJ OBJ(QTEMP/AROCCL6) OBJTYPE(*FILE) MONMSG MSGID(CPF9801) EXEC(CRTDUPOBJ OBJ(AROCCL6) + FROMLIB(&COMMONLIB) OBJTYPE(*FILE) + TOLIB(QTEMP)) The program could possibly get called more than once in a job, so there is no sense in deleting and recreating it every time it's called - the file just needs to be empty. You can't do a CLRPFM on a logical, obviously, so I'm just checking for existence and duplicating it if not found. Francis Lapeyre IS Dept. Programmer/Analyst Stewart Enterprises, Inc. E-mail: flapeyre@xxxxxxxx -----Original Message----- From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx Sent: Tuesday, January 31, 2006 10:21 AM To: Midrange Systems Technical Discussion Subject: Re: MONMSG CPF2130 was: CL program Yes you can CRTDUPOBJ more than one file. You can do it a couple of ways. 1 - Use generics and one command. CRTDUPOBJ OBJ(MYFILE*) FROMLIB(MYLIB) OBJTYPE(*FILE) MONMSG ... 2 - Or single string them. CRTDUPOBJ OBJ(MYFILE1) FROMLIB(MYLIB) OBJTYPE(*FILE) MONMSG ... CRTDUPOBJ OBJ(MYFILE2) FROMLIB(MYLIB) OBJTYPE(*FILE) MONMSG... They both have their pro's and con's. You may like the latter method if you only care that a particular object did not dup - and not why it didn't dup. But I suppose that same philosophy could be applied to the former. Clean up of the objects that did dup in case of failure in the latter case may be easier to handle. Then again, if you're going to QTEMP a simple CLRLIB QTEMP is quite succinct. Rob Berendt -- Group Dekko Services, LLC Dept 01.073 PO Box 2000 Dock 108 6928N 400E Kendallville, IN 46755 http://www.dekko.com -- This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe, unsubscribe, or change list options, visit: http://lists.midrange.com/mailman/listinfo/midrange-l or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take a moment to review the archives at http://archive.midrange.com/midrange-l.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.
midrange.com
