DSXchange

there should be "max buffer size" row in teradata connector properties under limits section, which is by default set to 0 to use default buffer size of 64KB. You will have to change that to push the record larger than 64kb.

Bulk method only supports up to 64kb in an array, if the single record is exceeding that, you may not be able to use bulk at all.


This link should help you.

Since you mentioned you have a query, a user defined query will execute in sequential mode only. You have to select entire table to make it run in parallel mode. When you say you get the result in 4 mins from oracle client, did you fetch entire 45 million records there, if not, that 4 mins are irrel...

Great work, Craig.

There are few which may be relevent to you. But from the time they started selling everything bundled, even the fixes are shown and bundled for all products in Information Server. You will have to find the ones you need by looking at their descriptions.

When you are getting null values are you getting any warning or info message from transformer which ends with something like "Use Default value" ? I think you have spaces etc in data which is causing it to default which converting from varchar to decimal. Try using IsValid() to find if the...

Did you verify that the unique constraint violation is for the columns you are mentioning as key in change capture, are there any other unique key such as the surrogate key which is generating old values?

Does it run without problem on one node?

Another excerpt from same communication with IBM support... At some point in 7.5.x, there was a bove to make environment variables case insensitive. Case sensitive environment variables was considered a bug under the case referenced with the following title: "Make environment variable not case ...

can you check xmeta if stage information is not present in the DSSTAGESXXXX or something like that but it exist in DSJOBSXXXXX. You should be able to join these tables on RIDs.

and they have mention that, such behavior can not be changed to match older version. They might be converting everything to uppercase under the covers but not much idea on why or if they are even doing. I am not on datastage box at the moment to check if I can define lower case unix env variables in...

Yes, Unix environment variables are case sensitive. You mentioned a generalized comment, hence clarifying that the datastage environment variables are not case sensitive anymore. Here is an excerpt from the mail sent by support. To followup on our phone call, the migration had an issue with the dupl...

Just to note, From a communication from IBM, environment variables are not case sensitive anymore. 8.5 and prior versions had case sensitive environment variables but that behaviour has been changed in 8.7. We figured that out when migrating a project from 7.5.3 to 8.7. But that doesn't change the a...

Did you change lookup reference partitioning from default to any other type of partioning? Is it same as partitioning in the stream link and is it a key based partitioning that you have selected? easiest way to check if the problem is related to partitioning, change partitioning in reference link to...

You can try doing a double convert. Doing a copy paste may not work. In those cases, do have a look at that character in hex editor to identify it and then use Char() with decimal value of that character. The character I can see here a unicode character used for replacement character with decimal co...

Here you go.

New and changed features of datastage 11.3.1 can be accessed here.

Just a note from the same page that File connector is not supported on Windows (Although I noticed that you are using unix).