Search found 53125 matches

by ray.wurlod
Fri Jan 13, 2006 11:56 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: SQLSTATE=HY000, DBMS.CODE=-302
Replies: 4
Views: 2497

Start by getting your DBA to tell you what an error code -302 is in DB2.
by ray.wurlod
Fri Jan 13, 2006 11:55 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Maximum Hash file size
Replies: 2
Views: 1254

It's hashed file, not "hash" file. A hash file is a tool for reducing the size of a block of hash. 1. What is the maximum size of a hash file that can still give a good performance? Lets assume we have 2 to 3 fields, each of length < 10 and integer/varchar type. It depends on the operating system. U...
by ray.wurlod
Fri Jan 13, 2006 11:44 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: SHARED CONTAINER
Replies: 1
Views: 817

The answer to this question is completely documented at the end of Chapter 2 in the Parallel Job Developer's Guide (qv)
by ray.wurlod
Fri Jan 13, 2006 3:29 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Copy one row to multiple rows
Replies: 16
Views: 8137

Or use a job parameter?
by ray.wurlod
Fri Jan 13, 2006 3:28 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Problem in modifying sql statement in db2/udb stage
Replies: 6
Views: 1676

Custom SQL means that you write the SQL. It must have exactly the columns in its SELECT clause as you have columns in your Columns grid in the stage. Apart from that restriction it can be any SQL statement that is legal for the database and which returns rows containing those columns.
by ray.wurlod
Fri Jan 13, 2006 3:25 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Using flat-file to stage data
Replies: 4
Views: 1199

Flat files are very fast. Extract to staging area can be run in a convenient time window for extraction. Decoupling means that you don't have to do the load phase at the same time; it can be run at a convenient time for the target database. Some folks, with complex transformation/cleansing requireme...
by ray.wurlod
Fri Jan 13, 2006 3:18 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Terrible Performance due to log file
Replies: 25
Views: 8432

Did you even look at the DataStage log (via Director)? My first question would be why so much stuff is being logged; in an ideal world there should be only a few tens of entries per run, and no warnings at all. Is auto-purge disabled, so that the log grows indefinitely? If so, manually purge the log...
by ray.wurlod
Fri Jan 13, 2006 3:14 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Copy one row to multiple rows
Replies: 16
Views: 8137

You can't perform any kind of lookup without a reference key.
by ray.wurlod
Fri Jan 13, 2006 3:09 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Failed to open RT_STATUS538 file
Replies: 4
Views: 1467

Another possibility is that the job was created and compiled by someone else whose umask setting and group membership was such that you don't have permission to open the RT_STATUS538 file.

This is less likely, but may need to be considered as a possibility.
by ray.wurlod
Fri Jan 13, 2006 3:08 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Failed to open RT_STATUS538 file
Replies: 4
Views: 1467

My guess would be a corrupted RT_STATUS538 file.
You could try clearing the status file then recompiling the job.
by ray.wurlod
Fri Jan 13, 2006 3:06 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Job status for running
Replies: 15
Views: 7669

Find a list of the active stage names in a job with DSGetJobInfo() and find the stage name you require in that list. Find a list of the links connected to that stage using DSGetStageInfo() and find the link name that you require in that list. Use DSGetLinkInfo() to determine the link row count. Alte...
by ray.wurlod
Fri Jan 13, 2006 2:59 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Rollback the tables
Replies: 6
Views: 1412

If you've committed you can't rollback. Databases 101.

You have to "unwind", by deleting. Based on job start timestamp is OK for rows newly inserted but not for undoing change. That's why type 2 SCD is a good approach; you never update rows (except maybe with expiry information).
by ray.wurlod
Fri Jan 13, 2006 2:56 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: jobnotify: System call error number 904.
Replies: 4
Views: 1274

If it's a memory leak I'd expect to see a segmentation violation (or SIGBUS) error somewhere. Please post their response. I'm curious.
by ray.wurlod
Fri Jan 13, 2006 2:51 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Replacing a substr with a single value
Replies: 3
Views: 1561

Is this a literal string "%3D" or are you referring to a single character 0x3d ? Convert() is a character-by-character conversion; Convert('%3D',':',Arg) is the same as Convert('%',':',Arg) then Convert('3','',Arg) then Convert('D','',Arg) - which is not the effect you intended. In server jobs and B...
by ray.wurlod
Fri Jan 13, 2006 2:44 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: passing parameter as group by key to agg.
Replies: 1
Views: 657

No. You've found the appropriate technique. The Aggregator derivations do not support the use of job parameter references.