DSXchange

I know that as this mapping between TYPE and SEVERITY is basically in the table definition posted in my second post. This "event type" (perhaps a bad name? "message code" or so is perhaps better?) would be a lot more detailed than TYPE and allows searching for specific messages. ...

The TYPE column is basically redundant to SEVERITY as SEVERITY is created from it. This is not what I mean. The kind of "event type" I mean has entries like "TFCN000001" which for example stands for the message that datastage has been started. So it's more or less redundant to th...

The output of the real hashed file is actually the same, I tried that before.

Unfortunately it's not listed: DICT RT_LOG 18:00:50 01-11-10 Page 1 Type & Field......... Field. Field........ Conversion.. Column......... Output Depth & Name.......... Number Definition... Code........ Heading........ Format Assoc.. @ID D 0 RT_LOG 10'0'R S TYPE D 1 Type 2R S WAVE.NO D 2 Wa...

Using a Hashed File Stage it is possible to read alle fields without knowing how they are called. I found a field which I would call something like "eventtype". There are entries like TFCN000001 or TODC000085 in it which classify the messages (seem to be those codes for which message handl...

Hello, I've got a job which sorts by a Timestamp-field (read from DB2 API with microseconds flags, non-nullable in DB). On execution, I get the following error after about 3 Mio. rows: Sort_123,2: Trouble encoding sort keys: Field "xxx" has keyprep export format error; value: *************...

uvsh "RESIZE RT_LOGxxxx *" worked so far and recovered about 200MB. Analyze.File before resize: File name .................. RT_LOGxxxx Pathname ................... RT_LOGxxxx File type .................. DYNAMIC NLS Character Set Mapping .. NONE Hashing Algorithm .......... GENERAL No. of...

I've done the purge from the base job. In the instance logs, the option to clear the log is greyed out...
The log exists since mid october (job was created then), so it's been 2 months now. Thus, I though that deleting 30 days would reduce the size by about the half...

We've built a job which lists logfiles which get too huge. The job is a MultipleInstance job which got > 1GB. After deleting entries older than 30 days via the director, the size stayed almost exactly the same (the older log entries are really deleted): 1207MB before, 1206MB now. The deletions made ...

Is it possible to reduce the size of the joblog after purging log entries (I've purged all entries older than 30 days)? Perhaps by some kind of Universe Database command like "Optimize Database" or so (I already know the "RT_LOGXXX" table which I need via uvsh)? I know that it's ...

The partitioning and sorting of the data in a dataset is the same like it was when writing into the dataset. However, Datastage doesn't know of this sorting and so you have to explicitly declare this in the job reading from the dataset. You can do this by using a Sort stage with "Don't sort/Pre...

You may also use the "External File" stage. Under "Source Program" you can enter a shell script. Each line this shell script outputs becomes a datastage row. You will probably want to use the "Column Import" stage after that to get the fields out of each row.

Thanks a lot! Works perfectly :D

Is it possible to read (and possibly write) special keys like "//JOB.STARTED.NO" from a shell script (unix ksh) and is there a list of these keys? It is possible to do issue commands like Select (with some rather straightforward SQL syntax) using the "uvsh"/"dssh" comma...