Add retry for wait_for_ready

[odyhunter]

Add retry for wait_for_ready

[coveralls]

Coverage decreased (-0.1%) to 85.049% when pulling f5115ac on odyhunter:delay_exponential into 05b95c7 on softlayer:master.

[allmightyspiff]

Thanks, I'll take a look at this tomorrow morning. There are a few changes I'll want to make as well, but I'll take care of that, so no worries.

[gltdeveloper]

A couple of point to consider...

• AWS (I thought) uses exponential backoff (Math.pow). This does not. This may be a good first start into the test (to catch the initial set of failures/timeouts that may occur) - just something to consider. I do like the random nature given that our client will definitely have a large amount of concurrency.
• In this implementation, it considers and does "x" retries where "x" is the configured # of retry attempts. One question we should ask ourselves is - for wait for ready, we are waiting for the VSI to be ready. If there are a number of concurrent requests all doing wait-for-ready, and we are near our API request limit per second, some requests may fail. So, do want to consider that if we fail on one request per timeout (we sleep/retry), and then are successful in issuing the request (but it's still not ready), so we reset the attempt counter? i.e., add attempt=1 to AFTER get_instance passes successfully? In this way, we tolerate more timeouts as long as progress is getting made on getting results..

[allmightyspiff]

I'm closing this pull request in favor of #869

I'll make my comments about my choices in that pull request.