DSXchange

Ken, I understand the relational model itself but was trying to justify the capturing of such fine details and wanted to know if I am the only one worried about them. You have answered my question. And thanks for the document. Arvind -----Original Message----- From: Kenneth Bland [mailto:kcbland_200...

Youre writing to a file on an NFS mounted filesystem. If you dont have the option of writing to a local filesystem, then you get to experience the joys of a remotely mounted file system. NFS - Not Fully Stable. Good luck! -Ken --- Rui Patricio wrote: > Hi all, > > I4m running a job under a unix shel...

Extracting job log information is tricky if youre going after the job logs directly. This is because the job log files have to be cross-referenced to another file to associate with the jobs name. In a nutshell, every job is assigned a number. Every time the job is renamed or imported, it is assigned...

If I may reply...The document in question has been freely passed around in a collaborative sense for over four years. It is (or will be again) available on www.datastagexchange.com as soon as some issues get worked out. In the meantime... In a typical warehouse environment you have periodic load pro...

Thanks everyone! Leonid Kenneth Bland cc: Subject: Re: Tab delimiter in Basic 05/07/2003 04:39 PM Please respond to datastage-users Sure, either directly use CHAR(9) or set a variable to it, like TAB=CHAR(9), then use TAB anywhere you need to. Ex: TAB=CHAR(9) blah blah blah msg="This is before the t...

That is *defininitely* not our problem. I will, however, check our MFILES setting and see about getting it bumped up from the default, which it probably is still at. The only thing that leads me to believe that it wont help is the fact that the job that failed ran in isolation. But well see... what ...

Kim, The article was on the DatastageExchange.com, but is not available anymore. I downloaded it as per the terms of the website. The process audit model in the article captures start and end times of every job in a batch; within a job it captures the number of rows passing through each link, and th...

Weve had the same type of problem. Ive called it into support, but no resolution yet. I think it has something to do with the amount of processes open at a particular time. During the time this normally runs we have over 100 jobs running that open 1 or more hash files for read/write. Some open as ma...

Right, this is another reason. The default value for the mfiles is so low (50)! You can increase it to 200. Kasia Fromudo.neumann@eds.com Todatastage-users@oliver.com Cc DateThu, 8 May 2003 11:55:22 +0200 SubjectAntwort: Jobs starting but not running or aborting > Hi Kenneth, > > yes, I see also you...

Never used DS scheduling, but times to times when launching jobs manually or through an external schedular they seem started but dont run. It was quite always because of insufficient space in the project directory. Hmm, think that everybody knows that. Kasia From"Craig Hulett" To Cc DateWed, 7 May 2...

Hi Kenneth, yes, I see also your problem in our environment. This occurs if DataStage isnt able to open more files. Each DataStage process open some files (RT_LOGxx, RT_STATUSxx, &PH&,...) . You must increase the parameter MFILES in your uvconfig file. Pleas check with support the right values for y...

What version of DataStage? If you want to code it yourself, your BASIC code can include calls to the DSSendMail function, with calls to DSGetNewestLogId and DSGetLogEvent to obtain the information. Or you can use a Notification Event in a job scheduler (DataStage 5.0 and later), checking the box to ...

Our control jobs have error detection built in, so that if required, email or SMS can be sent when a job finishes with errors or warnings. The action depends on how important the job is and what kind of error occurred. That is, it is built into the calling job, not the job which has the error. -----...

Hello everybody well I would like to know how to capture the warnings or the errors from the director logs in order to send e-mails because my sequencer run at early morning. In other words gentleman I would like to send an automatic e-mail while having errors or warnings in my sequencer. regards Ge...

Here, or more properly it *was* here: m For the moment, Kens articles seem to be Unavailable. -craig Kim Duke wrote: > Arvind Where is this article? I would like to read it. I have used > all the information that comes from monitoring a job and loaded into > Oracle and other databases. If you have j...