DSXchange

Tell me which parts of it aren't clear.

It's got nothing at all to do with the Aggregator. You have misdiagnosed this. The problem is that the lock table in DB2 is becoming full due to a combination or your activity and all other activities against the database. You can modify your job's locking strategy, or you can have the DBA create la...

You need something against which to lookup to determine whether to insert or update. Whether it's a hashed file or the target table directly is up to you, but the hashed file will be faster. TargetTable1 ---> HashedFile | | V insert SourceTable -----> Transformer --------> TargetTable2 --------> upd...

It means that the generic "open" function, which connects to the database and gets the SQL statement prepared, has not succeeded for some reason. Are there any other warnings, such as DSN could not be found, near this one in the job log? Do you get more diagnostic information when you reset the abor...

Job numbers (from DS_JOBS) are also unique across a project, but I believe what the original question was really asking for was a Run ID (rather than a Job ID).

In parallel jobs it is explicitly forbidden to "block" operations. In server jobs you can write data into a sequential file then read from that file, but the read operation does not begin until the write operation completes. (The assertion made by OttMAdpttch is not true in this regard.) This is an ...

It seems likely that you have been insufficiently diligent in accurately setting null field values. You have an empty field at column position 90 (the figure 89 is zero-based numbering) which is the field LINE_UTL_IN_PER; has this field's Null Field Value property been set to ""? Can you see the NUL...

uv -admin -stop stops all services (not just dsrpcd); uv -admin -start starts all services provided that there are no blocked connections on port number 31538. There is also a script called something like S99ds.rc in your UNIX auto-start directory (named something like /etc/rc2.d) that takes either ...

(1) Before/after subroutine in job
(2) Call DSExecute() subroutine with "SH" as Arg1
(3) Execute Command activity in job sequence

(There are other ways, but these are the most commonly used.)

Which database and which database stage are you using?

You could certainly use an SQL file from a shell script executed from an Execute Command activity in a job sequence.

Not sure, but they have been around for a long time - certainly 4.2, probably earlier.

You can learn about Oracle errors using the oerr command on the server. For example

You could use either of the SDK Key Management routines invoked via a Routine activity. These emulate a sequence in a database; the next value self-initializes to 1 on the first call, and otherwise is preserved in a record (whose key is the sequence name) in a Repository table called SDKSequences.

Oconv() is not available in parallel Transformer stages. You can learn about the "G" (group) conversion in online help for Oconv(), but it won't help you.

Oracle also doesn't have a specific TimeStamp data type. All Date data types are stored with time component.