unexpected response from server [500]: Too many open files

[pdreiter]

On a single program that I'm running algorithm type genetic, I'm seeing the following exception occur a few times:

<title>requests.exceptions.ConnectionError: ('Connection aborted.', OSError(24, 'Too many open files')) // Werkzeug Debugger</title>

I captured the STDOUT to a debug.log, so if this is something that is of interest to debug, I can attach that log.

errata: I have a large population size and number of generations that probably contributes to the expression of this error:

algorithm:
  type: genetic
  population: 200
  generations: 200
  tournament-size: 20
  mutation-rate: 0.8
  crossover-rate: 0.4
  # look at entire test suite for test sampling [subset of testsuite is 100%]
  test-sample-size: null

Looks like this ConnectionError ultimately results in the darjeeling process to hang. I'm guessing that it's waiting for evaluation data from the lost Candidates.

---

submitted from GitQ

[pdreiter]

Attached is the STDOUT from the darjeeling run [large file, gzip'd] anonymized.debug.log.gz

[pdreiter]

Follow-up : reducing the population size to 50 from 200 did not prevent the ConnectionError from occurring, but did prolong the time before the Error occurred.

[ChrisTimperley]

This sounds like a resource leak related to the requests API calls that are used to talk to the BugZoo server. From reading a few similar issue reports for requests, it sounds like it may be necessary to explicitly close requests.

I'll put together a potential fix now.