On Thu, Apr 18, 2013 at 10:02 PM, Niphlod <[email protected]> wrote:
> PS: on the web2py process side, it seems that at most 10 connections are
> istantiated with Redis: is there a connection pool?
I guess, because you are running the web server with 10 threads.
python redis by default use a connection pool... see below
> On the worker side, instead, I noticed that each job generates a new fresh
> connection to Redis: is this expected ?
I guess because, before starts a work, RQ forks a process.
This is how understand it, I may be wrong:
If a request for a connection is in a different process a new empty
connection pool is created.
If the request for a connection is in the same process but none connections
is available python redis starts another one.
If someone knows more details, please share.
This is the relevant bits I guess, on how python redis manage the connections:
def _checkpid(self):
if self.pid != os.getpid():
self.disconnect()
self.__init__(self.connection_class, self.max_connections,
**self.connection_kwargs)
def get_connection(self, command_name, *keys, **options):
"Get a connection from the pool"
self._checkpid()
try:
connection = self._available_connections.pop()
except IndexError:
connection = self.make_connection()
self._in_use_connections.add(connection)
return connection
def disconnect(self):
"Disconnects all connections in the pool"
all_conns = chain(self._available_connections,
self._in_use_connections)
for connection in all_conns:
connection.disconnect()
Ricardo
--
---
You received this message because you are subscribed to the Google Groups
"web2py-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
