| Author |
Message
|
| blorro |
 Posted: Thu Jul 13, 2017 7:31 am Post subject: CCSID conversion problem 1130 to UTF-8 |
 |
|
Acolyte
Joined: 09 Jan 2014
Posts: 57
Location: Sweden
|
Problem happens when consuming messages with MQ 7.5.0.5 Client from Windows based Queuemanager running 8.0.0.5.
(Receiver of Messages is SAP with PI, JMS based client)
Messages are sent with CCSID 1130 and this has been confirmed as they arrive on Queuemanagers LQ.
Consuming application tries to convert messages to CCSID 1208 (UTF8) and JMS Exception is thrown:
"Error receiving JMS message : JMSCMQ1000: Failed to create JMS message. JMSCMQ1046: The character set 'ccsid:1130' is not supported"
Converting the message with other tools, from CCSID1130 to UTF-8 works fine
So my question is :
What CCSID:s are supported with MQ Client 7.5.0.5 ? Or even the JMS 1.1 Java files for MQ Version 7.5.0.5 ?
I have looked extensively with the help of Google, and here on this wonderful forum.
Service request has been submitted regarding this , and also this failure is happening in a Test enviroment which is good, so were dealing with it in time.
Any help or direction is appreciated.
_________________
"Anything is possible, all the time."
Last edited by blorro on Wed Sep 06, 2017 11:52 pm; edited 2 times in total |
|
| Back to top |
|
 |
| JosephGramig |
| Posted: Thu Jul 13, 2017 10:08 am Post subject: |
|
|
Grand Master
Joined: 09 Feb 2006
Posts: 1244
Location: Gold Coast of Florida, USA
|
The PMR will give you the definitive answer. The conversion is being requested on the MQGET and the Qmgr uses OS routines to do the conversion if there is a routine for the CCSID installed for the OS. You get the error because there is no routine installed for that CCSID.
I see it is supported for MQ V9.0 and V8.0
BTW, I stopped looking when I found it...
You need to install it from the Windows media.
That was my first Google hit with "Windows conversion for CCSID 1130".
Searching is an art. |
|
| Back to top |
|
|
| blorro |
| Posted: Thu Jul 13, 2017 12:24 pm Post subject: |
|
|
Acolyte
Joined: 09 Jan 2014
Posts: 57
Location: Sweden
|
Really?
I thought the conversion is done by the client side when doing the mqget ?
At least that's the impression I got from all the reading I did on KC.
We don't do queue manager conversion for the other Asian languages landing on this queue manager so I was convinced it had to do with what ccsid:s were complied into the MQ Java binaries for the client v 7.5.0.5?
_________________
"Anything is possible, all the time." |
|
| Back to top |
|
|
| JosephGramig |
| Posted: Fri Jul 14, 2017 9:22 am Post subject: |
|
|
Grand Master
Joined: 09 Feb 2006
Posts: 1244
Location: Gold Coast of Florida, USA
|
When I pull up MQGET and look at the GMO at this page.
I see
| Quote: |
| The message data is converted by the queue manager for built-in formats, and by a user-written exit for other formats. |
So that would indicate it is converted at the server where the Qmgr lives. The client will indicate the code page and encoding it would like it converted to. Reading the second bullet at MQGMO_CONVERT, it says if the convert fails, the code page and encoding will be changed to how the msg was actually MQPUT.
Then you can do your own convert via Java libs.
I know the page was for MQ 7.1 but I'm sure this has not changed in many years. Are you sure your msg format is MQSTR?
I know I have to tell my AIX admins all the time to add converters for IBM PC and z/OS formats. |
|
| Back to top |
|
|
| tczielke |
| Posted: Fri Jul 14, 2017 11:11 am Post subject: |
|
|
Guardian
Joined: 08 Jul 2010
Posts: 941
Location: Illinois, USA
|
You may want to review this link, if you have not already:
https://www.ibm.com/support/knowledgecenter/SSFKSJ_7.5.0/com.ibm.mq.dev.doc/q032130_.htm
Possibly what is happening here is that you are relying on the data conversion that is happening only within JMS/Java, and the JRE does not support the 1130 to 1208 data conversion. If you use the queue manager to convert the data for the JMS client (as mentioned in that link), then the queue manager should be able to convert the data from 1130 to 1208 and then hand that 1208 message to your Java application for the GET.
_________________
Working with MQ since 2010. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Sat Jul 15, 2017 1:31 am Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
|
| Back to top |
|
|
| gbaddeley |
| Posted: Sun Jul 16, 2017 3:46 pm Post subject: |
|
|
Jedi Knight
Joined: 25 Mar 2003
Posts: 2538
Location: Melbourne, Australia
|
Agree, 1130 is EBCDIC Vietnamese
_________________
Glenn |
|
| Back to top |
|
|
| blorro |
| Posted: Sun Jul 16, 2017 11:17 pm Post subject: |
|
|
Acolyte
Joined: 09 Jan 2014
Posts: 57
Location: Sweden
|
Hello.
Lively thread now, which is good.
Vietnamese is correct. Data is being sent from our APAC region (Vietnam) into SAP PI (JMS based).
SAP techs explained they consume everything with UTF-8 why i also suspect the MQ 7.5.0.5 Client side Java files might be missing conversion from CCSID 1130 to UTF-8 but not sure about it why PMR is registered and yet unanswered by the way.
_________________
"Anything is possible, all the time." |
|
| Back to top |
|
|
|
|
