Re: [Bigdata-developers] Build failed in Hudson: BigData #70

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

Ah.  Yes, I was looking at another branch.  I have not touched that code in ages so I had no expectation that there were any changes to the test suite.  I see that three of the tests you wrote fail.

The ring buffer is primarily used by the B+Tree and some other cache classes as a hard reference queue to force the retention of recently used references.  

Are any of these failures linked to higher level errors?

Thanks,
Bryan

> -----Original Message-----
> From: Bob Resendes [mailto:res...@ya...] 
> Sent: Friday, July 16, 2010 7:56 AM
> To: Bryan Thompson; big...@li...
> Subject: Re: [Bigdata-developers] Build failed in Hudson: BigData #70
> 
> Bryan,
> 
> test_remove_get_order() is a new unit (as well as about 70 
> others) that I added to TestRingBuffer.java and committed to 
> the trunk earlier this week. It was part of rev 3196 (see 
> 
> http://bigdata.svn.sourceforge.net/viewvc/bigdata?revision=319
> 6&view=revision)
> and  ticket #101 (see 
> https://sourceforge.net/apps/trac/bigdata/ticket/101) was 
> submitted to cover it. 
> 
> 
> I suspect that you haven't done an update to the latest 
> "trunk" code, yet, which is why you aren't seeing that 
> failure (or test) locally, but it is showing up in the CI builds. 
> 
> 
> Regards,
> Bob
>