Author |
Message
|
rajaramanl |
Posted: Thu Apr 11, 2002 3:44 am Post subject: |
|
|
Novice
Joined: 07 Apr 2002 Posts: 13 Location: India
|
Hi
i put messages in a queue with some Message Id and Correlation Id. but in queue i find these id in different format(a string of special characters). This are generated by CICS program while Message PUT. What may be the reason for different format. Can any one breif abt Message Id and Correlation Id how it is been handled by MQ Series. |
|
Back to top |
|
 |
StefanSievert |
Posted: Thu Apr 11, 2002 7:10 am Post subject: |
|
|
 Partisan
Joined: 28 Oct 2001 Posts: 333 Location: San Francisco
|
Hi,
messageId and correlId are binary fields. So, if you put in a string or something like that and look at the message descriptor, you will most likely see the hex representation of that string. Be aware that, if you use strings for these fields and your messages leave the mainframe, these values are not subject to character set conversion but are treated as pure binary fields.
Is that anywhere near to what you are trying to do?
Stefan
_________________ Stefan Sievert
IBM Certified * WebSphere MQ |
|
Back to top |
|
 |
rajaramanl |
Posted: Thu Apr 11, 2002 8:50 pm Post subject: |
|
|
Novice
Joined: 07 Apr 2002 Posts: 13 Location: India
|
i generate these id's as string, and find special characters in the queue. i need the same id which i generate, for that which format to be set while message put. do i need to convert the format of the string. |
|
Back to top |
|
 |
StefanSievert |
Posted: Thu Apr 11, 2002 11:40 pm Post subject: |
|
|
 Partisan
Joined: 28 Oct 2001 Posts: 333 Location: San Francisco
|
What do the id's look like when you generate it and what do they look like when you browse the queue? Can you provide examples?
How (with what tool/program) do you look at the queue? |
|
Back to top |
|
 |
rajaramanl |
Posted: Fri Apr 12, 2002 1:25 am Post subject: |
|
|
Novice
Joined: 07 Apr 2002 Posts: 13 Location: India
|
When i generate i give it as ACCTLKUP for both Message Id and Correlation Id. In the Queue i find it as.
Message Id :ÃâØ@ÔØâã@@@@@@@@·nYÊn
Correlation Id :ÁÃÃãÓÒä×@@@@@@@@@@@@@@@@
I user MQ Series Explorer and look into the Queue. When i retrieve the Messages using VB program also it looks the same.
|
|
Back to top |
|
 |
muralidhar |
Posted: Fri Apr 12, 2002 9:37 am Post subject: |
|
|
 Acolyte
Joined: 28 Feb 2002 Posts: 50
|
If you want to messageID in readable format, use MessageIDHex instead of MessageID property of the MQMessage. This should solve your problem. |
|
Back to top |
|
 |
StefanSievert |
Posted: Fri Apr 12, 2002 9:57 am Post subject: |
|
|
 Partisan
Joined: 28 Oct 2001 Posts: 333 Location: San Francisco
|
You say you generate the Id's in CICS, so I am assuming on the mainframe. Is that correct?
Now, you say you are looking at the queue using MQExplorer and/or a VB application, so I am assuming non-mainframe.
I would really suspect that the whole problem is caused by the different character sets; EBCDIC on the mainframe, ASCII on Windows. As I said, the ID fields are treated as binary values and are not subject to EBCDIC/ASCII character set conversion. It is very likely that you see the EBCDIC values. The '@' looks very much like an EBCDIC 0x40 (blank).
You will run into problems if you use string values in the ID fields if your messages go across platforms with different character sets.
Cheers,
Stefan
_________________ Stefan Sievert
IBM Certified * WebSphere MQ |
|
Back to top |
|
 |
rajaramanl |
Posted: Sat Apr 13, 2002 12:03 am Post subject: |
|
|
Novice
Joined: 07 Apr 2002 Posts: 13 Location: India
|
Thanks Stefan and your assumption is correct. Instead of using String values for ID's what other values can be used. can you tell some examples. will conversion help.
i understand ID fields will create problems in cross platforms. is it applicable for messages also. in my situation iam getting my messages correct. |
|
Back to top |
|
 |
|