astronowbot

Jump to:. Sponsored Links Next. 1.

• Dspmsg Qsysopr

RE: MessageQ Break-Handling Programs -. If the break handling program isn't handling the message properly. Pro*C/C++ Programmer's Guide - Contents. The system has different types of message queues: workstation message queue, user profile message queue. The CTRL+C and CTRL+BREAK key combinations receive special handling by console processes. CTRL+BREAK is always treated as a signal. A better way of sending break messages. This program is simple as I am only accepting one user profile to send the message to and only allowing the message to be.

Clemente wrote: I have as400 system with os400 v4r5 and pc client access express v4r5 wirh so w2k sp2. The as400 works as file server through NetServer. Under this conditions everything works fine. We have just updated the pc clients to w2k sp3.

After that the clients has problems to create new folders and files of any type in the server. The control access parameters looks like no tot work properly. Are you sure it was updated to SP3? I do believe that SP4 was just released and there were numerous problems with many installs after updating. Can I change an override in a cl and have it affect the calling cl? Simplied sample pgma: ovrdbf input tofile(a) mbr(one) call pgmb.process file input. Dltovr input end pgmb: ovrdbf input tofile(b) mbr(two) end I thought I needed to use ovrscope(.job) and/or secure(.yes), but it doesn't seem to work (based on putting a dspovr in pgma).

I have an apache server which I will protect by a validation list. The setup is simpel, but after I have restarted the server and want to logon, it will not accept my password. After 3 attempts I get an error in the browser saying the password has expired. If I look in the log on the Iseries I get the following error: ZSRVMSG0302: User XXX: authentication failure for '/': 1. Any suggestion???

Does an AS/400 program written using RPG have better performance than using C? Could RPG provide low-level system control such as File access in IFS, job control and hardware control?

Is it a rule that a message queue break handling program cannot be called recursively? Chgmsgq msgq(qsysopr) dlvry(.break) pgm(mypgm) When a message arrives in the message queue my break handling program is called. The issue I am having is when a 2nd message arrives in the message queue before the break handling program has returned from handling the first message, the 2nd message does not break into the break handling program. Is this clear? Am I correct that the message queue break handling program has to return before the system calls the break handler program with the next message? Thanks, -Steve.

I also found that behaviour. I think that the msgq knows the break handling program is already active so even if the program is recursive/re-entrant it will not be called.

It dosnt always seem to be called when the user is displaying messages including program messages like the 'enter CDIR' type. I use a remote screen view program which uses this method to interrupt the users job & grab their screen. The other alternative I have considered is strsrvjob then trcjob with an exit program specified.

It works OK on a 1 off basis (eg a 1 line cl pgm with rollback as the command) but the exit program is called multiple times & you have to take that into account. The advantage is that a batch job with no workstation message queue can also be targetted.

That is the behavior I see also. The system appears to call the break handler for every message in the queue, regardless of when the message arrives in the queue. What I am trying to do is mimic the behavior of the QSYSOPR message queue when it is in break mode, with the addition of automated handling of some of the messages.

The problem is that when the break handler calls my DSPMSG equivalent program to display the contents of the message queue, the break handler does not get called again until the message queue display program exits. To get around this I am spawning a 2nd job which will run the break handler code. Then I have a 2nd message queue that is used to break into the interactive job and display the just arrived QSYSOPR messages. A bit more code than I had planned for. I dont think I studied it so well as you - I imagined some messages did not lead to the break program running.

I think its behaviour makes som sense though. If you had so many messages the break handler might never complete before the next message arrives and so at th end of the day just as everyone leaves the final message will be processed & appear on the 2nd users screen, followed by the penultimate message & so on leaving the 1st message to come out last. Or am I just being a bit pessimistic? I think you would get some swapping of messages if it was allowed to happen. Maybe you could try making the 1st instruction of the handler could reset the message Q to be in break mode & point to the correct program & see if this overrides its waiting. That seems to be overly restrictive wording on the part of IBM.

Regardless of what is done by the msgq break handler, you can still dspmsg qsysopr to display and respond to messages in the qsysopr msgq. What you lose is ability to break the qsysopr messages on the display. There are many shops that dont have someone monitoring the message queue, so I dont see the reason for high concern. Anyway, I have written my own version of DSPMSG QSYSOPR to break and display the messages to the operator.

Because I only want the messages that are not automatically responded to to break on the display in the computer room for the operator to respond to. If an.inq message breaks on the qsysopr message queue the operator will assume a response is needed. But if the background job is responding every interval to some messages things could ge confusing.

Or perhaps IBM knew that it wasn't going to be that easy to do it - as you are now finding out.;-) So, why not have a second message queue (QSYSOPR2) also in break mode in the QSYSOPR job. In your break handing program, pass any messages that need attention to the QSYSOPR2 queue. In your break handling program, loop around before returning and read any messages that have arrived in the meantime. This (should) hopefully get around the not-being-called-if-already-active problem and the operator will still respond to break messages in his/her usual way. Similar Threads: 1. I'm trying to do what I thought would be simple in SQL.

Dspmsg Qsysopr

That is making an old fashion control break, where one field only prints when it changes and the other fields print for each record. I've hunted and hunted for an examples of this with out success, leaving me to believe that no one does this any more or it's so simple no one needs to discuss it.

Thanks in advance for any help. Sample SQL statement is. I want 'Board Name' to only print when it changes. SELECT ADMLIB.ADP0001.CMLNAM AS 'Board Name', Trim(BDMFST) ' ' BDMLST AS 'Member's Name', substr(BDMFDT,1,2) '/' substr(BDMFDT,3,2) '/' substr(BDMFDT,5,4) AS 'First Term', substr(CDMCDT,1,2) '/' substr(CDMCDT,3,2) '/' substr(CDMCDT,5,4) AS 'Current Term Started', substr(BDMEDT,1,2) '/' substr(BDMEDT,3,2) '/' substr(BDMEDT,5,4) AS 'Current Term Ends' FROM ADMLIB.MEMBERS INNER JOIN ADMLIB.ADP0001 ON ADMLIB.MEMBERS.BDCODE = ADMLIB.ADP0001.CMCODE ORDER BY 'Board Name' ASC 2.

Hello All, While developing a service program to perform I/O for files I have noticed that exceptions are being handled differently to how they work in regular ILE programs. All error handling is performed as follows: - At the beginning and end of every sub procedure a condition handler is registered and unregistered. The condition handler procedure, when called, determines the program, module, procedure, line number etc that the exception takes place and emails this info to the program administrator. It then percolates the exception. Every procedure also includes a PSSR subroutine which gracefully shutdowns the program.

This PSSR routine takes control of the exception after the condition handler percolates it. This results in every exception being caught and dealt with in the procedure which caused the error. I have written the service program procedures in the same manner but have noticed that when an exception occurs. The service programs procedure runs the condition handler registered with it, percolates the exception as usual.

However, after this the service program procedure's PSSR routine is not called. Instead the excepetion percolates to the calling program, who's condition handler once again informs me of the error, then the calling programs PSSR routine processes the exception. Because the condition handler is being called twice from two different procedures i am receiving an email for the same error twice. The only difference is that the system reports different libraries and jobs for where the exception occured Email Output 1 - From Service program: Program: PRODUCT Module: PRODUCT Procedure: OPENTESTMST Library: GPL Job: IT02A User: HARRISON Job Number: Error: Error on an explicit Open/Close operation. Statement Number: 12100 Error message CPF4326 appeared during OPEN.

Cause.: RPG procedure OPENTESTMS in program GPL/PRODUCT received the message CPF4326 while performing an explicit OPEN operation on file TESTMST. The actual file is TESTMST. Recovery.: Check the job log for a complete description of message CPF4326, and contact the person responsible for program maintenance. Email Output 2 - From caller program: Program: PRODUCT Module: PRODUCT Procedure: OPENTESTMST Library: SHC Job: IT02A User: HARRISON Job Number: Error: Error on an explicit Open/Close operation.

Statement Number: 12100 Error message CPF4326 appeared during OPEN. Cause.: RPG procedure OPENTESTMS in program GPL/PRODUCT received the message CPF4326 while performing an explicit OPEN operation on file TESTMST.

The actual file is TESTMST. Recovery.: Check the job log for a complete description of message CPF4326, and contact the person responsible for program maintenance.

The service program is run from the.CALLER activation group. I would like the service program to be able to gracefully shutdown itself and the program that called it, but due to the nature of service programs how can I tell which program called it, or what job it was running in? Forgive the length of the post. Hi Steve, You can use the 'List ILE Program Information (QBNLPGMI)' API, format PGML0200 to retrieve a list of all service programs bound to a program.

Furthermore the 'List Service Program Information (QBNLSPGM)' API, format SPGL0800 can be used to get a list of all signatures of a given service programs. So basically combine these two APIs to check for signature violations.

Thomas Raddatz. PS: I do not have any idea how to automatically handle a MCH4431 message. Steve Richter wrote: any ideas how to automatically handle escape message mch4431 - program signature violation? it is easy enough to write a cl program that would read the message data and rebind the program in error. But how to call that program automatically, without using monmsg that uses custom code to redo the call that failed. baring that, can I check every program and service program in a library for existing program signature violations?

Any ideas how to automatically handle escape message mch4431 - program signature violation? It is easy enough to write a cl program that would read the message data and rebind the program in error. But how to call that program automatically, without using monmsg that uses custom code to redo the call that failed.

Baring that, can I check every program and service program in a library for existing program signature violations? I have this external USB HD that I only turn it on when needed. One time, when trying to mount, it asked if I would like to check the HD. I clicked yes, and it started checking for a while but stopped at 99%. Maybe I was too impatient, I turn it off and tried to mount it again. After attempting with disk reading noise, it couldn't.

It gave a message that the HD needed to be formatted to use. The drive changed its name to 'Local Drive.' I tried a few programs and it seems that I can recover files with programs such as R-Studio but I can't mount the drive. It's rather inconvenient to retrieve files this way.

I'd like to mount the drive and copy some files and then reformat the drive. Is there a program to mount a HD like the condition my HD is in? Thanks, liu 6.

I have never liked the Send Break Message command, SNDBRKMSG, as it is not possible to send the message to a user without knowing which work station they are. There is an alternative: the Send Message API, QEZSNDMG. This API allows me to send a message to directly to the user, without caring where they are. It is the only API I can think of that has an user interface that can be used interactively, and the same API can be used in a program to run without the user interface. Below I am going to describe how this can be used with the user interface, and then how to call the API in a program. User interface All I have to do to use the user interface is just a program call: CALL QEZSNDMG Below is the screen that is displayed: Send a Message Type information below, then press F10 to send.

Message needs reply. N Y=Yes, N=No Interrupt user. N Y=Yes, N=No Message text.

Send to. Name, F4 for list More. F1=Help F3=Exit F10=Send F12=Cancel The parameters displayed are pretty much self evident. Message needs reply If this is 'Y' when I send the message the receiving user(s) can enter a reply to the message is returned to me. Interrupt user If 'Y' when the message is sent is will 'break'/interrupt the user. Message text This is the message you want to send.

Send to This is my favorite part of the API. I can list one of more user profiles who I wish to send the message to. F10=Send You need to press F10 to send the message. If you press Enter a message is displayed informing you: 'Press F10 to send message'. When I get one of those dreaded call at night from the operator I use this to communicate with him. For example: Send a Message Type information below, then press F10 to send. Message needs reply.

Y Y=Yes, N=No Interrupt user. Y Y=Yes, N=No Message text. The message has been handled. You can now continue Send to.

OPERATOR1 Name, F4 for list If this message is sent to someone who is signed on I get the following message: Message sent to OPERATOR1. If OPERATOR1 is not signed on then I get this instead: Message sent.

User OPERATOR1 not signed on. OPERATOR1 receives the following message that interrupts them on all their active sessions: From.: SIMON DD/DD/DD TT:TT:TT OPERATOR1: The message has been handled. You can now continue Reply. They enter their reply press Enter.

I receive their reply as a message. From.: SIMON DD/DD/DD TT:TT:TT OPERATOR1: The message has been handled. You can now continue Reply. Ok Including in a program I wish that QEZSNDMG was as simple to include in a program as it is to use interactively.

I cannot remember when I first wrote a program to use this API, it was many years and releases ago. I chose to write a 'black box' type program, one were the user does not have to deal with all of the parameters, formatting, etc. They just pass a couple of parameters and the code in the program does the rest.

I chose to write it as a CL program as I found it easier than writing the equivalent in RPGIII. Karl Hanson It would be fairly easy to revise your CL (or RPG) program to be a Command Processing Program (CPP), and create a CL Command interface for this function. The current program input parameters (&USER and &MSG) could be required parms for the CL command, and if you wish, other optional CL command parms could be defined for other API parameters, such as MSGTYPE. The required user/user profile input could be defined as a list CL parm, with a minimum of 1 value required, but a maximum of a larger number if desired. The command analyzer then automatically passes the number of values specified by the user to the CPP - that is, the user need not specify a separate parameter value for the number of input USERs.