DSXchange

Only messages that are read will be removed from the queue, so in your case the 2nd call of the job will find 5,000 entries in the queue (unless some process has written more)

If you create a test job that has a DB2 stage as source, can you connect to a table from there using "view data"? If not, what is the error you get?

You can specify your year cutoff in either the date format (i.e. "%NNNNyy", page 31 of the Parallel Job book) or by specifying a new value for the $APT_DATE_CENTURY_BREAK_YEAR

There have been 8 major releases of DataStage, each major release has had between 3 and 7 minor releases and many of those have had numerous patch releases. There are hundreds, if not thousands of modifications done over time - that doesn't include the different sub-products. Nobody is going to spen...

Can you put an echo of the command itself with values in your script and then manually try to execute the same "dsjob" command?

Roopanwita wrote:...(wrote a script ,but it is not working)...

It would be helpful if you could post the part of the script that isn't working along with a description of the problem.

I think converting to the internal date first is better. To a StringToDate() call and then later a DateToString() with appropriate formatting.

MQ Queues are first-in-first-out, so the ordering will remain in place - unless you change it explicitly in your DataStage job. You can confirm this by reading and checking the internal 24 byte "MsgId" unique number - these will be incremented by 1 at the rightmost position for each record.

Ag_ram - Please don't change topics when you find that you have posted yourself into a corner. You have been asked to explain the use of uvbackup and uvrestore for this example and all you do is refer to "39 search results". That is as useful as someone asking about Oracle that "I get hundreds of se...

The "DSParams" is found in the project directory for each project.

This can be done either with string functions or by converting to date and then back to display. In your case I would do a transform derivation of "In.ShortDate[1,6]:if In.ShortDate[7,2]<'50' then '20' ELSE '19':In.ShortDate[7,2]". This assumes that all dates are from 1951 through 2050.

If you use the bulk loading method and the index rebuild fails, I recall that the DataStage job does get an error. In your case, what method are you using and what type of error do you get when rebuilding the indices?

The MFILES setting in the UVCONFIG controls how many actual Hashed files are open at any time; this setting can seriously affect performance but has not direct impact on the problems described in this thread.

BMC's Control-M is a common tool and, when implemented correctly, is a very powerful one. I haven't used HP Sitescope, but I thought it was more of a monitoring tool than a scheduling one.

Yes, it is.