Fix rediss ssl eventlet sslerror bug

This is the same as [^1].

I did a test deploy to double check that Redis on PaaS doesn't work
without this.

[^1]: a2cbe20325

gunicorn_config.py

@@ -1,6 +1,8 @@
 import os
 import sys
 import traceback
+import eventlet
+import socket
 
 import gunicorn
 from gds_metrics.gunicorn import child_exit  # noqa
@@ -17,3 +19,21 @@ def worker_abort(worker):
     worker.log.info("worker received ABORT")
     for stack in sys._current_frames().values():
         worker.log.error(''.join(traceback.format_stack(stack)))
+
+
+def fix_ssl_monkeypatching():
+    """
+    eventlet works by monkey-patching core IO libraries (such as ssl) to be non-blocking. However, there's currently
+    a bug: In the normal socket library it may throw a timeout error as a `socket.timeout` exception. However
+    eventlet.green.ssl's patch raises an ssl.SSLError('timed out',) instead. redispy handles socket.timeout but not
+    ssl.SSLError, so we solve this by monkey patching the monkey patching code to raise the correct exception type
+    :scream:
+    https://github.com/eventlet/eventlet/issues/692
+    """
+    # this has probably already been called somewhere in gunicorn internals, however, to be sure, we invoke it again.
+    # eventlet.monkey_patch can be called multiple times without issue
+    eventlet.monkey_patch()
+    eventlet.green.ssl.timeout_exc = socket.timeout
+
+
+fix_ssl_monkeypatching()