DSXchange

As far as I know it needs a remote shell (rsh) capability between both the servers and not ssh. If you can go through the Install and Upgrade guide then it is clearly mentioned over there.

For Case 1:- ---- If you are sure that all your dates are of this century then your first transform will work correctly but I think there are spaces coming in the input column due to which its giving the wrong answer. So just use the following transform StringToTimestamp('20':Trim(DSLink40.TODAYS_BU...

See this post http://dsxchange.com/viewtopic.php?t=97064&highlight=&sid=08f59dd73a28246b707ac31e99f3db07 Steps:- 1) check if you are able to login on both the servers using rsh ( without prompt ) for password. 2) Please put an entry for both the hosts in /etc/hosts file of both the servers. ...

I am not able to recall this but I have done this in past nearly 2 years ago where I have tried to use the resources from the other server.I have even posted the complete solution for that, it seems you may need to enable some of the Environment variables if your rsh is working properly. Just check ...

[quote="uppalapati2003"]Hi dsusr, i understand what u r saying but my doubht is how i pass the previous effectve date to the particular satge variable.if u dont mine can u explain brefly..that was very help ful to me[/quote] Use the Following stage variables in the transformer:- StageVaria...

Bit of crude logic on top of my mind..... 1) Just create your reference file as each word in a separate row with same key value asume it to be 1 so that your file wud look like:- 1, ACCT 1, ACCOUNTANT 1, TRUST Even if you have a file with single row you can do such kind of transformation in the job ...

Just write the logic in stage variables i.e. to capture the date of the previous record as end_date if the surrogate key column is same.

But please make sure the data need to be sorted based on key values ( surr key, date)


----
dsusr

[quote="ray.wurlod"]Please note that [i]dsrecords [/i]is an independent command, not an [i]orchadmin [/i]sub-command. ...[/quote]

Sorry my mistake, i wrongly put orchadmin in front of it.

Following are thevarious options to count the nmber of records:- 1) As Ray said use the Dataset management tool 2) Use the Orchadmin commands i.e. just go into your PXEngine/bin directory type the following commands:- orchadmin dump < dataset name with path > | wc -l or use orchadmin dsrecords < dat...

Hi all, In past I have used SQL queries on the Hashed file using Command Line ( either in Administrator or on Unix ) but at that time the Hashed files were Located in the Default project directory. Now when the Hash files are getting created in a separate directory other than the Default project dir...

Input ---> Transformer ---> Two links from Transformer i.e. Link1 and Link2 Link1 ( only the year column ) -----> Remdup (on the years ) ----- Transformer Transformer writes a single record which contains all the distinct years concatenated with any delimter ( assume it to be pipe) and Text 'EmpId' ...

If you have dsadm access then either try releasing the job using dssh or thru DS Administrator. Steps using dssh:- 1) Go to the DSHOME/bin/ directory and get the Userno from the 'list_readu' command. 2) type dssh and use the command LOGTO <ProjectName> 3) Type UNLOCK <Userno> Steps using DS Administ...

Prabu - I hope the column is nullable - yes. Since the the error is at export, for test, try with a sequential file to a dataset. Hi Kumar, Yes, the column is nullable. I miss the point here, why a dataset as target :roll: . my target is a sequentia file and it expects the null to be handled as a p...

[quote="sankarsadasivan"]well, it was a typo, i meant 7.5. We have 7.1 as of now, and we are getting our new box and s/w installed only after 2 weeks. Till then we have to wait. Ray- It was not an interview question. If any of you have any information, let me know. Thanks[/quote] there are...

Not in all cases. Untill the lookup data is small enough to handled in lookup memory, lookup approach is better. Once it is beyond certain limit, it is advisable to switch join, which doesnt pre load the data into lookup memeory. Yes Kumar you are right that it depends on the data but performance o...