DSXchange

What happens if you ask a consultant the time?

He asks to borrow your watch, tells you and then keeps it.

I've "consulted" on this situation many times, and the down and dirty normally is a lot of data collected and stored for absoulutly no reason.

I went to the Melbourne one last week. It was fun and informative.

Learnt heaps about OSH and Info Server since IBM took it over.

Pizza and Wine wern't bad either.

It wasn't a vendor push like so many User groups turn into.

Becuase, as in all things, that's life and how it works. Deal with it.

I've seen this a bit lately. When dropping columns in a stage the incremental complier will work (force compile doesn't) and job fails. Also occurs if you have an output column in an SQL stage that does not have a matching select column from the SQL, especially user defined SQL. Happened a few times...

G'day Malini,
Congratulations. Any gotcha's future name changes might want to be aware of, of shortcuts you used to make it stress free?

Welcome Knip1 to our little community. Rules I always follow for a join stage: 1) Hash Partition on first element of join key 2) input link (or in sort stage before hand) sort on all join keys 3) Make sure youe link ordering is correct 1&2 are particularly important to make sure that the same ke...

I like the write output to a dataset in append mode if this is absolutly necessary. 1st job in sequence creates a dataset 2nd job left outer join input against the dataset to flag what has been processed, continue to append to the dataset in the job (so if it fails multiple times it will continue to...

Metric calculations and logging need to be done, but do they need to be done NOW? We write logging request data to a flat file and calculate it later. We prime the records with a server job to do the insert and update (start and end) in the sequence, but not the row calcs etc. they are done later fr...

I thibnk: SH: -c: line 0: syntax error near `(28801>' is a clue. Looks like you need a space between the variable that is reprresented by 28801 and the >. and Craig's suggestion is valid, DS executes basic sh commands and a lot of 'nix's don't . Most are either bash or ksh and you need to specify th...

I've been at two In Melbourne Australia, Current gig with 8.5 and an older one with ... .Version 4.

Can be slow, but it does work. Not as good as a locally based client, but that doesn't always work and not everyone is on big corporaste WAN's

A little off topic but I helped a collegue use a 5 bay NAS fitted with 2.5"bays to hold 5 x 256GB SSD's in a raid 5 configuration for about 1TB fast access. After the datamarts are created and cubes generated they are copied to the NAS for user access. 2 x 1GB ethernet links, paired into it to ...

You should do a search, and it may help. A couple of differences between 7.x and 8.x: 1) You need to follow the installation instructions EXACTLY. 2)This includes the post installation configuration stuff, mainly around linking infosphere users to Unix users 23 you need to make sure that the unix us...

just a quick update. worked around the problem by doing as I mentioned, an on the fly edit of the xml to change the namespace. seems to be a DS "feature".

Tell the mainframe admins you want to FTP the JCL into the JES spooler. That's the service I was thinking of :), thanks Cppwiz for remembering it, saved me unarchiving and digging in my old code. This is the easiest way to do it. You must remember to set the transfer mode to ascii (not binary) so t...

If you are connecting to the R/Ht system from a Windows box, I like to use UltraEdit by IDM, It was cheap, can access files on the Unix/Linux remote box and becomes a 1 stop editing shop for me. Combined with my favorite File tool (FileCommander) I'm in pig heavin (one without a fllu though). Just m...