DSXchange

-911
THE CURRENT UNIT OF WORK HAS BEEN ROLLED BACK DUE TO DEADLOCK OR TIMEOUT. REASON reason-code, TYPE OF RESOURCE resource-type, AND RESOURCE NAME resource-name


Did you try running it on a single node ?

whats is the sql code that you are getting ?

Since it runs after restart , my guess is this is data related (you might be getting more that one records from your input stream for the same key)

Start by running you code on a single node.

and reduce the record count to the least possible.

hey All, I have a tricky requirement and I am not even certain if its possible. We have two servers (one has 8.7 version and the other has 11.3) I want to run a dsjob command on the 8.7 version server but i want to connect to the 11.3 server. dsjob syntax i am using right now on the 8.7 version serv...

Below is what I see now. If you see last few lines you ll see a lot of repeated name space declarations. Seems like a waste of space, anyway I can avoid it and have this name space towards the begining of the xml. <ACAUIBusinessHeader xmlns="urn:us:gov:treasury:irs:msg:acauibusinessheader"...

here is something you can try.
I am sure that the DB2 Database must be cataloged on both the head and the compute node. UnCatalog the database in question from all the compute nodes and run the job.
And when you run send the job to these compute nodes.

from what i remember , connections are made thru the head nodes.

yea , The XML is not using the tags the way its supposed to. If you are in UNIX or windows i would suggest just write a shell script to extract the records with entRecno. (from the look of things it looks like they are going to be on the same line).. And use the output from that as an input. Just a ...

few things i would check are . 1. Are there duplicates coming from the source links. If so remove / manage them. 2. Work with DBA to identify the active query that is in execution when the job hangs. 3. Check the partitions of the job. i.e. test by running on a single node to see if that works. 4. M...

select
table1.cols,
table2.cols
where
table1.key = table2.key
and
table2.col2 not in ('ABC')

are you certain its not running on more than one node ?
Check with the DBA to see the queries executed by the datastage user ID.

@Cranie - Can you give some more details of the DS job that you used to capture the said details.

If I understand correctly - You have a sequencer and one of the parameter variable you use is for hostname. And you want to avoid typing in Hostname again and again. If that is the case I would suggest using Parameter set Value files then. you can create 3 value files (one for each environment Dev /...

Interesting.
Best way to improve chances ?

So I talked to the DBAs.
And they said will be able to give me the table names and the bytes per record (using something called DCLGENs)
multiplying that by the count of records loaded by ETL would give a rough estimate on the data bytes moved by ETL nightly.

Whats the throughput that you see with your current design.