deploy/ process cube issue

Hi,

I'm having difficulty to deploy and process cubes to the differenct server now.

It may be caused by our internal server problem but before checking that, please advice the following issue:

(To deploy and process cubes to local machine, and differenct server works perfectly previously. This problem just started few days ago.)

If I delopy and process the cube to the localhost then it works fine.

I also ran the sql profiler and trace the process.

But in my local machine, in BIDS if I change the target server and process cubes to the different server then

it just start deploying but keep running and never ending.

Also, I've tried to remote the target server and process cube in server machine while tracing that process.

The thing is every single process works fine but it's not ending. In trace file, I can see " finished processing cube"

But, cube process keeps running and I couldn't see " command end" eventclass in trace file.

Right now, we reload data into another fact table in test server , not production server. ( cube is in production)

but is ther any possibility to affect the cube processing?

Other then that, I have no idea why this happens.

I would appreciate if anybody can give me some advise.

Thanks.

FYI,

I've talked to another developer who are working on ETL now, and he told me that he've truncated some data in exisiting table and reloaded data. And also,he and did some changes in table structures and reload the data into another new fact table.

Does it matter to deploy and process the cube ?

(I'm guessing that table structure won't affect the processing issue since it will give an error before deploying, if it causes the table relationship or attibutes changes.)

/**

Now, I've taken small portion of the cube structure and create new project and deploy to the different server and it works perfectly. But is this the only way to solve this problem? Any suggestion?

Please let me know.

Thanks.

|||

We've occassionally had this same problem. It appears that the files that contain the cube and dimension data sometimes become corrupt. When you process the cube, all the data is written to a second set of files. There is then a drop/rename type of activitythat swaps out the new files in place of the old files. It's the swapping of files that you are waiting for.

Recommendation is to drop existing cube altogether, redeploy structure, and then reprocess. Very likely that everything will correctly at that point.

In fact, you're already seeing that your new project/cube are working correctly on the same server.

Hope this helps,

Pam W.

|||

Thanks for your valuable information.

could you tell me what exact file do I need to drop? All cube files (*.cube and *.partition) ?

What else do I need to take a look?

I've also tried to use ' deployment wizard' and overwrite xmla script and it's showing successful depolyment.

But processing still doesn't work.

I've tried to drop cube file and to rebuild the cube but it's just giving an error and can't process cubes.

I would approciate if you can give me more details in regards to fixing this problem.

Please let me know.

Thanks.

|||

I've just tried to create new database into the target server and redeploy cubes with new db then it works now.

It seems file is corrupted but still doesn't understand well. (I followed your recommendation but it doesn't work)

Anyway, thanks for your recommendation.

If anybody has a similar situation and can give some comments then I would greatly appreciate it.

Thanks.