Socket Timeouts to Combat Zombie Agents

We're getting close to something usable!

The largest 'customer' using this was running into issues where the daemon simply stopped reporting. It was still running, just waiting on a response from something.

Some debugging indicated that it was waiting for a response from a tcp connection to new replic's API that had timed out.

SO we added a socket timeout value (currently hard-coded to 5 seconds) to combat this. Test VM's have been up and running for ~10 days with no lock-ups.

Better Logging and Exception Handling

adding real logging facility and exception passthrough for the data collection functions

Wide Distribution Beta

Ready for more widespread deployment / testing

http proxy functionality

confirmed that we have a functional http proxy configuration.

ready for testing through https proxy

ready to test through an https proxy, or directly to new relic

Initial Release

The first release.

No TRUE QA, but I squinted my eye really hard and stuck my thumb out while looking at the code.