DSXchange

shyju,

did you split the sequential stage into 2 objects? I am 90% certain that will solve your problem.

I just saw that this is a PX forum question. In that case, I do think that a recompilation is necessary if anything in the environment has changed, particularly paths and the PX configuration file.

You don't even need to re-compile if you don't want to

Craig - I'm sure that's going to be the solution; I had the same problem a couple of weeks ago and couldn't figure out why certain things weren't working as expected.

The basic transform stage normal works without a problem. If you took a very simple job with an input sequential file with just one column (or use a row generator stage to do the same) and a BASIC transform stage that just passes the column through with no changes and write to a sequential file or d...

shyju, I think what has happened is that you have a process that is hung on the pipe - it really is necessary to make sure you have a reader AND a writer and ensure that you have timeouts set. I think that on Sun platform the fuser command might give out processes that have a fifo pipe open, it does...

did you try it in the right pane in the manager ? It's always worked for me.

hmmm... that last post needs to be corrected! The command to execute had 2 errors, it should read: Call DSExecute("UV","SELECT NAME FMT '60R' FROM DS_JOBS WHERE NAME NOT LIKE '%\%';",output,sysretcode) Your output variable "output" will contain a multivalued list of the j...

I am not sure what is causing this. Can you change your 2 sequential file definitions to another file name and use the "mkfifo {newfilename}" command and then explicitly doing a "chmod 777 {newfilename}". Perhaps it is your user's umask setting causing a problem (btw, are you running your DataStage ...

You need to delete the temporary sequential file and use the mkfifo command to create a named pipe in it's stead.

Hello reddyds, what is your question? Are you wanting to install 2 versions of DataStage on the same machine? If so, it is possible and is explained in the documentation under installation of separate instances. The two engines need to have different shared memory addresses and need to be accessed v...

Hello pavanns,

you haven't given close to enough information to be able to help. Most likely you are using a Database stage and your custom SELECT clause contains more columns than declared in the metadata.

The second sequential stage in your job that uses the named pipe should have the input filename and output filename identical and both should point to the file that you created with the mkfifo command. You should set the stage to read/write a pipe and make sure that the timeout values are set - if y...

Craig, you don't need to enable NLS for this. Just make sure your conversion in DS is none and that you have the ENVironment settings for oracle correct. The input conversion to/from Unicode should map 1:1 and not convert characters; especially since the upside-down-question-mark and other Spanish c...

I think that this functionality is limited to DataStage 7.5; you might still be on 7.1.

It make more sense to correct the configuration as Ray suggested instead of putting in an explicit handler; I would recommend you add the new pool