DSXchange

I would simplify things first in order to debug it....let's take XML out of the picture. Are you putting the test messages into the queue yourself? Copy your job and remove the xml stage and put in a transformer instead. Put in one message.....how many rows in the sequential target do you get? Ernie

...look carefully thru your settings on the MQ Stage or Connector.....normally it will just read one whole message (you should just have one single large column to receive it)...but there are options to have it look for crlf's in the data, and that "might" be the issue you are seeing.....y...

Unfortunately, FastTrack only uses Connectors in its generation for sources and targets. You have to delete the Connector Stage and attach the one that you require.

Ernie

If you check the box for "Generate Operational Metadata), you will find an xml of the job run in /IBM/InformationServer/ASBNode/conf/etc/xmlFiles .... This is the formal xml file that is used to enable lineage and high level summary operational information to be used within the Metadata Workben...

It's possible that the xsd is not complete or is invalid...or it could be that it contains one of several constructs that the pack is unable to read..... As an alternative, get a complete document sample for it.....and import that...the importer can import xml documents as easily as xsd's... just be...

That's why I said you will probably need to do all the validation yourself, and then manually construct the xpath you want...not "use" the xpath.

Ernie

I think so....been awhile since I explored all the syntax options for xpath, but that's a possibility.

Exactly..this is why I think the best solution would be for you to learn more about xpath, and then build, after your invalid record appears, a custom xpath that represents the particular row and value that died. I can't remember the exact way you specify values and I don't have time to check right ...

And to Craig's point, Workbench isn't available in 7.x, but you could import 7.x Jobs into 8.x with Workbench and immediately analyze them (meaning, they don't have to be compiled and running in v8, just imported).

Ernie

My suggestion would be to look at one thing at a time, and separately validate that each "type" of lineage is working for your particular jobs and your particular configuration. ...and also, while Automated Services "will" parse user defined SQL, it's not a good way to get to kno...

Log into the Information Server "Web" Console...not the Information Server Console where you deployed your service (you can view it there too, but it's better that you learn how to do it for all your services).... Connect in your browser to: http://<your server>:9080/ That's where you will...

The generic navigation xpath is in the Stage itself...you have probably seen it, but if not, scroll to the right of your output link....this is the xpath that identifies each node and its elements/attributes that you want to retrieve. ...you'd have to build your own "value" based xpath res...

Metadata Workbench can provide you with this information also, by doing a query that looks at "Stage Columns"....which are effectively, Columns-on-a-link.....

Ernie

Thanks for sharing. Can't say I have any other suggestions though, for your Connector import. There are other ways to get tables into Info Server...I like the Connector import --- on most issues it is the most mature and robust, but obviously it has some sort of problem in your environment based on ...

log into the Info Server Web Console and look thru the Information Services Repository Tab............

Ernie