DSXchange

You did "ps -ef |grep dsrpc" and you don't see the process out there?

You need to decide your threshold for pain. Fewer jobs in a project allow for a faster graphical experience using Director. It makes no difference what project in which a job lives, as long as internal hashed files are not shared across your 3 separate job classes. Since you're not developing these ...

Because it's just an interface to At on Windoze and cron on Unix. That is not the same as an enterprise scheduler, which has interdependent process scheduling, escalation activities, monitoring, graphical interface, etc. Cron/At run a task at given time with an optional repeating interval. If you do...

In the world of computers files have no datatypes. Datatypes are a prejudice used by programs when reading data from files. DS Sequential stages write ASCII text files. There is no concern on your part about datatypes, as files do NOT have datatypes. You don't need to do anything.

Primary input stream has --> @INROWNUM
Each output stream has --> @OUTROWNUM

Day 0 in the DS BASIC internal date world is 1967-12-31. Somehow you're getting to there in your manipulations.

If you're trapping to a sequential file, the datatypes are really irrelevant except for a few quirks. Choose a delimited file and just set the metadata to VARCHAR. Set the length and display to something you like because it affects the VIEW DATA display. If you choose fixed-width, you get into issue...

I think Ray neglected to read the entire discussion, otherwise he hopefully wouldn't have given a bad answer.

Table Definitions stored in Manager are not necessary for executing or importing a job design. Proper metadata management is a different topic. You should consider keeping all design information together no matter which environment. A table definition is copied into a job design with a silent link b...

Your project is a repository for jobs designs, runtime logs, etc. Do you think that's a good location for work files and such? Would you write sequential files there as well? Explicit external paths allow you to more easily manage all work files. Consider removing hashed files from an external path ...

Sound idea. Seems to me a better import/export Wizard is in order. Something with "Tagged by developer" or "Searching for changed since" capabilities, etc.

Keep your job logs purged, don't use your project to house hashed or sequential files, and make sure you periodically sweep for accidental sequential files in the project.

Use Hashed file 2 as your primary input stream, and reference lookup Hashed file 1.

There's no way, security is at a project level. You can mark jobs as read-only using the version control tool or manually updating the repository.

All hashed, ODBC, and OCI lookups to the primary input stream are equi-joins, think of it like a cursor lookup. A found row is not required.