DSXchange

The userid/password you enter in this location is to tell the scheduler which userid/password to run under and isn't related to the database connection at all. You should put your database User and Password information into job parameters; and either use the default values at runtime (by not explici...

A quick use of google shows that there are ODBC drivers out there for BTrieve. I didn't check whether there are freeware ones out there or not. If you have an instance of Btrieve you could always export the data into a csv or similar format from the database. You are going to need either BTrieve or ...

Manoj, you will have to do 2 passes through your source file no matter what solution path you take. If your file doesn't have millions of rows and thus lookup performance isn't of paramount importance then I would go about solving this with 3 jobs - a Sequencer and 2 server jobs. (0) write a sequenc...

Oopsie; it is "CLOSESEQ". Fortunately jobs close all open file units when they finish, but it is always good programming practice to explicitly close open files instead of letting an implicit close do it for you. Thanks.

Ray - time to reminisce about the wonders of PRIMOS. I still have a tape with the complete sources somewhere in storage; but it is 10 years old and might not be legible anymore By far my favorite OS.

if you call the job from UNIX you could use the command line escape \t when setting a parameter value. I think that if you put a tab into your paste buffer you can paste it into the parameter value field when calling it from the Director as well.

Suggestion 1: Get a new Administrator
Suggestion 2: Start the deadlock daemon
Suggestion 3: Acquaint old Administrator with blunt instruments. Repeat.

Ray,

I didn't know that the CHAP command did anything on OSs other than PRIMOS. I'll have to try that, thanks!

I think that you might best solve your issue by using a before and an after job subroutine. The before-job subroutine can use the OPENSEQ, WRITESEQ and CLOSE commands to write one or more header record(s) to the file while truncating it. The job itself writes 0 or {n} data records and then the after...

Hello Specialist,

A number of people have seen this error before and have posted it here, SEARCH is your friend.

Amongst the posts that a search would have given you is this one:

http://www.dsxchange.com/viewtopic.php?t=87878&highlight=synchronization+error

Probably this question is better posed in the Ascential Developer network; as Ascential employees post more frequently and with more authority there. I think that the balance between Server and Px is certainly going to shift towards Px, but that DataStage server is not in it's dotage and will contin...

When reading a sequential file you can specify attributes for what to do when a column is missing or null. I would suggest you define the maximum number of columns but for those that might be missing in certain rows declare them to be empty when they aren't defined in the source.

OK, I shouldn't have asked I think you have no choice but to take each of the internal selects singly until you get to the part that has a different number of resulting records. Either that or get the list of additional rows and see if you can come up with a rule.

Also think about activating the deadlock daemon so that locks like this will automatically clear themselves every n minutes (default is 15).

The reject file is always in the format of the source file (same thing happens when an output has a reject link). That is the way the system works; you could always put a transform stage on your reject output, but if your format is wrong then you wouldn't be getting useful information.