DSXchange

If you were to write the same job into a table with upsert rather than a hashed file the statistics there would be the same. Hashed files can be checked from the admin tool or the command shell with "COUNT {HashedFileName}" assuming you created a normal hashed file instead of one with a pa...

You can copy a dataset from one server to another, but it is not very simple. Do an "orchadmin ll {dataset}" to your descriptor file. That will give you the names and paths to the actual data files. Copy the descriptor and data files to the new machine, the descriptor file can go anywhere,...

The two "fi" in the odbc.sh are obviously wrong, the second one should be removed. This error remains at my copy of Version 8.
The question I have is whether or not this script needs to be executed, I've never had to.

Have you addressed the first severe error, the lack of a domain name for the AIX server? Although I didn't know that it was necessary, I have to admit that I've never been on a machine where the domain wasn't set.

Those commands are used in the command tool or from a dssh shell; but they won't necessarily solve your problem. Deleting the UV.ACCOUNT entry does not remove the schema and does not address any entries in xmeta.

I should have read the whole thread, not just the last post.

Kittu - which UNIX are you running? Each flavour has a different way of specifiying the hard and soft limits.

HTML files are sequential files. If you don't need the contents parsed but just loaded into strings in the database, it is easy. HTML is a structured format which has no builtin parsers as one has for XML, though. So a lot depends upon what your HTML contains and how it needs to go into your database.

in the before-job routine do an execute shell of "ulimit -a" and you will see the run-time values displayed in your log file.

I assume you mean you want to do this outside of a job, right?

From UNIX you could do something like:

You can get the exact statement that is causing the error by using "VLIST"; if you search the forum the syntax and use is described a couple of times.

Reset the job in the director, do you have an entry "from previous run..." and, if so, please post that.

This can be done by using the complex flat file stage, which is used for COBOL type record formats such as yours. You can declare multiple record types for header, data, and footer and filter out what you don't require.

The line numbers can be off by a couple of lines, this is due to compiler line length issues and has been around in UniVerse for many years. Just look for a line around 347 which does a mathematic computation on a variable and you have found your culprit.

Those are a lot of questions for one post. Since a server job link collector cannot be set to take input from one link, then the other you cannot optimize that way. Do you have PX available?

Earlier you indicated that the data was "IBM-1386" which would be simplified Chinese. Why not try using "ibm-1386_P100-2002" in your sequential file Stage -> NLS Map settings?