DSXchange

Information Analyzer analysis tasks are run as DataStage jobs in the ANALYZERPROJECT DataStage project.

The Information Analyzer database (IADB) is used to store the results of analysis - for example the results of column analysis are used in performing table analysis and cross-table analysis.

Welcome aboard. The phrase "not working" isn't very helpful. Can you be more specific about the symptoms?

Can you also please verify the delimiter character between the seconds and the fractional seconds part of your source data?

Read the on-line help to learn about this control. Essentially, though, it allows you to override the default behaviour. Be very careful in your reading to identify WHERE this buffering happens (it's probably not what you're thinking). Please post the results of your research here.

Yes, don't keep asking the same question over and over.


We're trying to get to the right answer in your other thread.

Please advise what you mean by "IPC_out stage". I am not aware of any such stage type, particularly in version 5.

Please advise all the stage types used in your job design.

The bin folder should never contain VOC. You need to attach to the Engine directory and, while there, execute the command bin\dssh which will invoke "TCL" with its ">" prompt. It is there that you need to execute the given COPY command. The QUIT command will exit from TCL and dss...

As you will have observed, the LIST.READU command is available in every project but, by default, the UNLOCK command is not. You can install the UNLOCK command in a couple of ways, as noted in my earlier post. If you have used any of the unlocking options from the DS.TOOLS menu, the UNLOCK command wi...

How are you actually deriving the value that is supplied as the first column in the metadata? That is, please show us the expression that generates the "comma delimited list of values".

Read again that the job was first run on a single node. I'm guessing, therefore, that something "sequential" has happened, maybe in the mechanism that initializes the state file or something within the function itself.

You get zero issues because/if no other processes touch the same hashed file. You could do the same in a routine. But it's best practice to own the lock before inserting/updating.

Does it have to search a DSX file? Would it be OK to search the Repository database instead? If so there are examples here that query the DS_JOBOBJECTS table in various ways, and from which you could probably derive an appropriate query.

The Surrogate Key Generator stage will, like the Transformer stage, do exactly what you tell it to do, though its defaults are more likely to be well behaved. In a Transformer stage construct your expression using @PARTITIONNUM (plus any initial constant) as the initial value, and increment by @NUMP...

In particular, setting of row buffering is entirely irrelevant in parallel jobs. Buffering is automatic, and tuned by a half-dozen or so environment variables with BUFFER in their names. Intra- or inter-process row buffering is only relevant to server jobs.

If their backups are good (taken when nothing was happening) and if the restore is done properly (in the correct sequence for the incrementals from a known good point) then all should be well.

Identify (from the score) exactly what partitioning is being used at each stage in your job. Post that information here. Without it it is not possible to provide cogent advice.