Re: [Bigdata-developers] Errors in bulk load tests (regression)

[Bryan T. <br...@sy...>]

Fred,

Please retry.  Martyn has committed a change against the trunk which should resolve this issue.  Please let me know if the problem is resolved, in which case we can also close out https://sourceforge.net/apps/trac/bigdata/ticket/118.

Thanks,
Bryan

________________________________
From: Fred Oliver [mailto:fko...@gm...]
Sent: Wednesday, July 28, 2010 5:55 PM
To: Bryan Thompson; and...@no...
Subject: Re: [Bigdata-developers] Errors in bulk load tests (regression)

Bryan,

I agree that performance comparison for the new version is a good thing, though I'm a little surprised that you waited until after the announcement to test. I'll refer you to Andrew about the machines.

The problems I'm seeing are on a single workstation, if you could take a look at the exceptions....

Fred

On Wed, Jul 28, 2010 at 5:38 PM, Bryan Thompson <br...@sy...<mailto:br...@sy...>> wrote:

Fred, I was trying to setup the ci performance machine for this purpose but ran into OS configuration issues which I passed to Brad to look at.

We really need to test against a cluster to have confidence here.  Can I get access to the 16 node cluster for a few days?

Bryan

Fred Oliver <fko...@gm...<mailto:fko...@gm...>> wrote:



I'm running a test which generates (essentially lubm) data files and invokes MappedRDFDataLoadMaster.main(). The configuration is multiple services running on a single machine.

After svn update to the current branch, the test appears to run forever with many exceptions seen in the log file, apparently relating to IO. I've attached two versions of the error.log file.

One shows errors from FileChannelUtility, BufferOverflowException, etc.
The other shows AssertionErrors being throw between processes.

Would one of you take a look, please?

Fred