So the first bisect seemed to go off just fine, just didn't reproduce much of a regression. The 2nd hit timed out fetching DEPS during startup:

Traceback (most recent call last):
  File "/b/rr/tmpdyjYuu/rw/checkout/scripts/slave/recipe_modules/auto_bisect/resources/fetch_file.py", line 51, in <module>
    main()
  File "/b/rr/tmpdyjYuu/rw/checkout/scripts/slave/recipe_modules/auto_bisect/resources/fetch_file.py", line 47, in main
    print fetch_file(args.repo, args.commit, args.path)
  File "/b/rr/tmpdyjYuu/rw/checkout/scripts/slave/recipe_modules/auto_bisect/resources/fetch_file.py", line 37, in fetch_file
    encoded_content = urllib2.urlopen(url).read()
  File "/usr/lib/python2.7/urllib2.py", line 127, in urlopen
    return _opener.open(url, data, timeout)
  File "/usr/lib/python2.7/urllib2.py", line 410, in open
    response = meth(req, response)
  File "/usr/lib/python2.7/urllib2.py", line 523, in http_response
    'http', request, response, code, msg, hdrs)
  File "/usr/lib/python2.7/urllib2.py", line 448, in error
    return self._call_chain(*args)
  File "/usr/lib/python2.7/urllib2.py", line 382, in _call_chain
    result = func(*args)
  File "/usr/lib/python2.7/urllib2.py", line 531, in http_error_default
    raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)
urllib2.HTTPError: HTTP Error 503: Service Unavailable

I'll try restarting that one. We're switching the android bots to pinpoint soon so most likely won't try to improve the error message from the old recipe version.