Evidence of Resource Leak?

[Cantor, Scott]

On 2/17/15, 8:24 PM, "Marvin Addison" <marvin.addison at gmail.com> wrote:
>
>Sorry, but I thought I had responded; possibly wasn't clear. The resource 
>leak is resolved from what I can tell; our dev IdP has been running for a 
>week without issue. What's not clear is why the resolver service appeared 
>to be attempting to reload the file every 30s or so during the problem 
>period. The reload interval was set to the default PT15M. The resolver 
>config failed to load during this time, so I wonder whether there's 
>another load mechanism that takes over in a bad config situation. In any 
>case I'm fairly certain that such frequent reloads exacerbated the 
>resource leak and made it painfully apparent.

I think there probably is a bug lurking in there on failed reloads, but if 
we can't reproduce it quickly, I think I'd leave it to post 3.1 given that 
various changes have been made to the service code anyway.

-- Scott