smartfile / django-mysqlpool Goto Github PK

I guess pip package isn't the last version, because I had to install the package as a git repo.

items = [(n, tuple(v)) for n, v in self.items if isiterable(v)]
TypeError: 'builtin_function_or_method' object is not iterable 

About my platform:

platform.release       3.14.19-100.fc19.x86_64 (Fedora Linux 19);
python.version          2.7.5;
django.version          1.6b2;

May be it'll better

items = [(n, tuple(v)) for n, v in self.items() if isiterable(v)]

The Readme says:

Once we decorated all functions that utilized a connection, this service used less connections than it's total thread count.

In my case, I have a Django command where I want to use the decorator. This command imports some functions and classes from the project. So, for example, in a 5-level execution flow would be something like (tree levels are ok, markdown messed a little):

• Django command
  • view1/function1
    • view1/function2
      • view2/function1
        • view3/function1, which has SomeModel.objects.___ call

Do I have to decorate the command all the execution tree down to the bottom function where Django uses a connection, or decorating just the main function (in my example, the Django command one) is enough?

Thanks!

why don't you use a middleware instead of a decorator for closing the connection?

When I am start using your lib, we are facing high load avarage and CPU utilisation in my production servers. we have to handle high traffic in our website. can you give me any solution for this?

Can you suggest best configuration for High Traffic websites using django and alchemy connection pool?

When we use multiple database on same MySQL host, e.g.:

• test_db_202010
• test_db_202011
• test_db_202012
• test_db_202101
• ...

Can those database connect with same connection pool ? We don't want to create a pool for each of them, since it will break the connection limit when we have a lot of django process. e.g: 100 machine * 8 process * 6 (month) database conn pool * 10 connection in each pool .

But actually they could use the same pool base on host