Friday, 3 December 2010

Updating Tivoli Director Integrator v7

Following my previous post, I did hit a problem whereby TDI v7 didn't appear to update correctly.

During the update process, which I performed using the command: -

cd /opt/IBM/TDI/V7.0/bin
./applyUpdates.sh -update /tmp/TDI/7.0.0-TIV-TDI-FP0005/Fixpack.zip

I did see an exception: -

CTGDKO023I Applying fix 'TDI-7.0-FP0001' using backup directory '/opt/IBM/TDI/V7.0/maintenance/BACKUP/TDI-7.0-FP0001'.
CTGDKO027I Updating SERVER.
CTGDKO009E Error while executing an operating system command. An exception occurred: /opt/IBM/TDI/V7.0/bin/tdimiggbl.sh: not found
CTGDKO027I Updating CE.

which, at the time, I ignored.

This was a bad move :-)

What I should have done is read this Technote (1391717): -

applyUpdates process fails

which directs one to replace this JAR file: -

/opt/IBM/TDI/V7.0/maintenance/UpdateInstaller.jar

and then re-run the fix pack installation: -

./applyUpdates.sh -update /tmp/TDI/7.0.0-TIV-TDI-FP0005/TDI-7.0-FP0005.zip

However, I then hit another issue;

CTGDKO032E The level of the current installation is '7.0.0.5'. The level must be between '7.0.0.0' and '7.0.0.4' to apply this fix.

Therefore, I realised that I had to first roll back from 7.0.0.5 to 7.0.0.0 before re-applying the fixpack: -

./applyUpdates.sh -rollback
./applyUpdates.sh -update /tmp/TDI/7.0.0-TIV-TDI-FP0005/TDI-7.0-FP0005.zip

Having done this, the updates installed as expected.



No comments:

Note to self - use kubectl to query images in a pod or deployment

In both cases, we use JSON ... For a deployment, we can do this: - kubectl get deployment foobar --namespace snafu --output jsonpath="{...