Fix "mysql server has gone away" in mysql cache backend

[lorien]

How to repeat:

1. Start spider, use mysql cache
2. Pause spider execution (for instance, import pdb; pdb.set_trace()) in some task handler
3. Wait for a some time, well, 5 hours, for example :)
4. Get the error: mysql has gone away

[oiwn]

I got same thing using Flask with SQLAlchemy before.

http://stackoverflow.com/questions/16341911/sqlalchemy-error-mysql-server-has-gone-away
http://stackoverflow.com/questions/6471549/avoiding-mysql-server-has-gone-away-on-infrequently-used-python-flask-server

[asakasinsky]

Hi, @lorien
There are my db.py, SQLAlchemy template for all my projects. Parsers, Flask sites etc.
«mysql has gone away» is solved.

I drop this file into some project, example:
..
./models/
./app.py
./db.py

and init db in app:

from db import db_session, init_db
SQLALCHEMY_DATABASE_URI = "mysql+mysqldb://%s:%s@%s/%s?charset=utf8" % (db_user, db_pass, db_host, db_name)
init_db(SQLALCHEMY_DATABASE_URI)
...
...
db_session.add(Model)
db_session.commit()

And Model:

from sqlalchemy import (
    Column,
    Integer,
)
from db import Base

class User(Base):
    """User Model"""
    __tablename__ = 'users'
    id = Column(Integer, primary_key=True)

[asakasinsky]

«Patch» action log:
2014-10-13 17:47:30,027 INFO sqlalchemy.pool.QueuePool Connection <_mysql.connection open to '127.0.0.1' at 7ffcb48eca20> exceeded timeout; recycling
2014-10-14 21:02:31,507 INFO sqlalchemy.pool.QueuePool Connection <_mysql.connection open to '127.0.0.1' at 7ffcb4002220> exceeded timeout; recycling

[lorien]

Пожалуй, закрою этот тикет, так как я его сам открывал, а фунциональность эта мне уже сто лет как не нужна.