No, but since it's a random number, I think it would be good to explain why it
was chosen in the source code; otherwise, it looks meaningful.
--Justin
> On Feb 24, 2015, at 10:08 AM, Andy Zhou <az...@nicira.com> wrote:
>
> It is a random number. Since the goal is to scale from tens of
> connections to hundreds of connection, 330 seems like a good first
> step. Is there another number you'd prefer?
>
> On Tue, Feb 24, 2015 at 10:01 AM, Justin Pettit <jpet...@nicira.com> wrote:
>> Can you comment on why 330 was chosen?
>>
>> --Justin
>>
>>
>>> On Feb 24, 2015, at 9:58 AM, Andy Zhou <az...@nicira.com> wrote:
>>>
>>> Raise the connection limit to allow larger number of concurrent
>>> ovsdb-server connections. Note, ovsdb-server may not perform well
>>> at the new limit. It is rather a prelude to further scaling tests and
>>> optimizations.
>>>
>>> Signed-off-by: Andy Zhou <az...@nicira.com>
>>>
>>> ---
>>> This limit is currently hard coded. Should we make it a parameter
>>> for the caller to set?
>>> ---
>>> ovsdb/jsonrpc-server.c | 2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/ovsdb/jsonrpc-server.c b/ovsdb/jsonrpc-server.c
>>> index 1092ffa..91e24ee 100644
>>> --- a/ovsdb/jsonrpc-server.c
>>> +++ b/ovsdb/jsonrpc-server.c
>>> @@ -121,7 +121,7 @@ ovsdb_jsonrpc_server_create(void)
>>> {
>>> struct ovsdb_jsonrpc_server *server = xzalloc(sizeof *server);
>>> ovsdb_server_init(&server->up);
>>> - server->max_sessions = 64;
>>> + server->max_sessions = 330;
>>> shash_init(&server->remotes);
>>> return server;
>>> }
>>> --
>>> 1.9.1
>>>
>>> _______________________________________________
>>> dev mailing list
>>> dev@openvswitch.org
>>> http://openvswitch.org/mailman/listinfo/dev
>>
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
