DSXchange

Please advise how it was resolved, to assist future searchers.

Initial thought was that the parameter DSCAPIOP__TSS_DB_NAME_STAGING was not being resolved correctly.

But second thought is that you're trying to update something in the repository.

How about a few more details about what the database is and what you're trying to accomplish here?

Sender address?

Please don't use "u" when you mean "you". I find it particularly annoying. And it gives those of our members whose first language is not English a wrong impression. We strive here for a professional standard of written English.

Are the indexes in the project in need of rebuilding?

It can be done from UNIX server, it can be done from Administrator client, it might even have been done from a job - either accidentally or maliciously. I can't tell if it's just the catalog entry or the code as well that's missing. The workaround I gave you will be faster than trying to find out. F...

Browsing through the Pattern Action Language manual (as one does) for version 8.5, I notice that both CONVERT_S and CONVERT_P are supported (as well as CONVERT and CONVERT_R). CONVERT changes tokens. CONVERT_P concatenates a prefix onto specified tokens. CONVERT_S concatenates a suffix onto specifie...

Each of these processes is executing something in the DataStage engine. It needs to communicate via shared memory, so has to have a shared memory key, which is what the ipcs command is showing you.

The ps command for these processes may show what each is doing.

No, Arnd can't do that, U can't do that, and I can't do that. Because each server is different.

Are you giving the correct SMTP server name in your new environment?

Some tweaks may be needed for version 8, such as -file and/or -domain options for dsjob.

What I mean by "at global level" is "for all fields in the record structure". This is set on the Format tab.

Give it the correct pathname for the range map file or make sure that that file is readable by the process that is performing the range partitioning.

Neither of these affect MQ.

Transport blocks are used when data are re-partitioned. Physical dataset block size refers to physical storage of Data Sets.

You're wrong.

There's always a condition.

The default is Fail.

ketanshah123 wrote:What will be the reason for this error ?

How would we know? It was unexpected even to DataStage! Maybe there was a blip in your network - we have no way of knowing.

No idea. I'd prefer to prevent them being sent to Oracle in the first place. Hence my suggested approach.