sqlite3.OperationalError: database is locked - non-threaded application

I have a Python application which throws the 开发者_JS百科standard sqlite3.OperationalError: database is locked error. I have looked around the internet and could not find any solution which worked (please note that there is no multiprocesses/threading going on, and as you can see I have tried raising the timeout parameter). The sqlite file is stored on the local hard drive.

The following function is one of many which accesses the sqlite database, and runs fine the first time it is called, but throws the above error the second time it is called (it is called as part of a for loop in another function):

def update_index(filepath):
    path = get_setting('Local', 'web')
    stat = os.stat(filepath)
    modified = stat.st_mtime
    index_file = get_setting('Local', 'index')

    connection = sqlite3.connect(index_file, 30)
    cursor = connection.cursor()
    head, tail = os.path.split(filepath)
    cursor.execute('UPDATE hwlive SET date=? WHERE path=? AND name=?;', (modified, head, tail))
    connection.commit()
    connection.close()

Many thanks.

You might particularly check for functions that keep a read-lock (unfinished cursor). That would block the commit from the update function. Note that there is a dedicated mailing list for Python-sqlite problems: http://groups.google.com/group/python-sqlite

Do you really need to continuously open and close the database file for every single UPDATE? If you do the same thing in every function that accesses the database, is it possible that you have called the update_index function from another function that already opened the database using a different connection and is in the process of modifying the database?