logfile for bin/uv -admin -start

flashgordon · Post by **flashgordon** » Mon Oct 08, 2007 11:51 am

Hi,

We are starting our Datastage SE server and it is not coming up even though the direct message associated with the start says it's coming up. Is there a logfile for SE that shows what's it's doing and what error's it's getting when bin/uv -admin -start is issued?

... Flash

chulett · Post by **chulett** » Mon Oct 08, 2007 11:55 am

There's no log. What typically doesn't come up it the dsrpcd processes, is that your issue? If so, try searching the forums for either 'bind bombed' or '-d9' to see how to debug the issue.

Bryceson · Post by **Bryceson** » Tue Oct 09, 2007 10:02 am

I have seen this happen before . . . . I got to work by sourcing the dsenv file

1. Try to stop the Engine

2. Source the dsenv file ($DSHOME/ . ./dsenv)

3. Start the Engine again

Bryceson

flashgordon · Post by **flashgordon** » Tue Oct 09, 2007 6:57 pm

Bryseson,

Thanks. Unfortunately we are way beyond this. We have Peoplesoft support all over it. We are now starting datastage in debug mode to try to get an error message out of it.

... Flash

flashgordon · Post by **flashgordon** » Fri Oct 12, 2007 8:09 pm

Hi,

We resolved this today with the help of a very senior IBM Datastage Support person. I don't know how much will be of general interest but this is what happened. Something (probably unrelated to Datastage on the Unix side) deleted a large number of files in our $DSHOME .../DSEngine directory, all the D_* files, dsenv, uvconfig. So if your Datastage SE instance won't come up one thing to check is that all your $DSHOME files are present. A healthy $DSHOME directory should have at least 100 files in it. In our case the Datastage server would try to come, give no error messages and look a little normal. But without the UV Data Dictionary files (D_*) nothing would happen.

... Flash