django 在线程中创建的DatabaseWrapper对象只能在同一线程中使用,”

7cwmlq89  于 2023-11-20  发布在  Go
关注(0)|答案(4)|浏览(116)

我正在使用celery从CSV读取数据并将其上传到Postgres。celery任务正在工作(我认为),但django抛出了一个错误。我正在获取一个文件,将其转换为pandas,删除2列,并转换为numpy并传递给celery。

#tasks.py
@shared_task
def a(data):
    for x in data:
        date=x[0]
        date=datetime.datetime.strptime(date,"%m/%d/%Y")
        date=str(date.date())
        desc=x[1]
        amount=x[3]
        account=x[6]
        cat=x[5]
        account_type=x[4]
        #changing account type
        if account_type=="debit":
            account_type=False
        else:
            account_type=True
        #creating account if needed and assigning the the object to account
        try:
            account=t102_accounts.objects.get(c102_acc_name=account,c102_is_credit_fl=account_type)
        except:
            account=t102_accounts.objects.create(c102_acc_name=account,c102_is_credit_fl=account_type)
        #creating cat if needed and assigning the object
        try:
            cat=t100_code_lookup.objects.get(c100_code_name=cat)
        except:
            cat=t100_code_lookup.objects.create(c100_code_name=cat,c100_for_exp_fl=True,c100_for_income_fl=True)
        #creating the transaction if it is not existing already
        try:
            t106_transactions.objects.get(c106_trans_amount=amount,
            c106_trans_date=date,c102_trans_card_num=account,
            c100_trans_cat=cat,c106_trans_name=desc)
            continue
        except:
            t106_transactions.objects.create(c106_trans_amount=amount,
            c106_trans_date=date,c102_trans_card_num=account,
            c100_trans_cat=cat,c106_trans_name=desc)
    return 'done'
#calling the function in views
def upload(request):
    if request.method=="POST":
        form=UploadFile(request.POST,request.FILES)
        if form.is_valid():
            data=request.FILES["file"]
            data=pd.read_csv(data)
            del data["Notes"]
            del data["Labels"]
            data = data.dropna(axis=0)
            data=data.to_numpy()
            data=data.tolist()
            a.delay(data)
            return redirect(dashboard)

字符串
我的追溯在这里列出。追溯:

File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\celery\utils\dispatch\signal.py", line 288, in send
    response = receiver(signal=self, sender=sender, **named)
  File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\celery\fixups\django.py", line 172, in on_task_postrun
    self.close_database()
  File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\celery\fixups\django.py", line 177, in close_database
    return self._close_database()
  File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\celery\fixups\django.py", line 186, in _close_database
    conn.close()
  File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\django\utils\asyncio.py", line 24, in inner
    return func(*args, **kwargs)
  File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\django\db\backends\base\base.py", line 286, in close
    self.validate_thread_sharing()
  File "c:\users\yogab\appdata\local\programs\python\python38-32\lib\site-packages\django\db\backends\base\base.py", line 553, in validate_thread_sharing
    raise DatabaseError(
django.db.utils.DatabaseError: DatabaseWrapper objects created in a thread can only be used in that same thread. The object with alias 'default' was created in thread id 54283112 and this is thread id 8121832.```

bnlyeluc

bnlyeluc1#

我想我有一个类似的problem
你如何开始你的celery 任务呢?你可能使用--pool=eventlet
和**-游泳池=独奏**它为我工作。
范例:

celery -A yourapp.celery worker --loglevel=info --pool=solo

字符串
关于池选项的更多信息:https://www.distributedpython.com/2018/10/26/celery-execution-pool/

xbp102n0

xbp102n02#

要解决这个问题,建议关闭与数据库的连接,因为Django在Celery中执行方法时不会管理连接。

import connection

connection.close()

字符串
或者,消耗较少的资源:

connection.close_if_unusable_or_obsolete()

dw1jzc5e

dw1jzc5e3#

这不一定是操作系统相关的问题。
在我的情况下,这是由于将项目迁移到Python 3而引起的,对我有效的解决方案是:
1.更新celery和eventlet到最后版本
1.在celery配置中添加CELERY_TASK_ALWAYS_EAGER

# celery configuration

CELERY_BROKER_URL = 'redis://127.0.0.1:6379'
CELERY_RESULT_BACKEND = 'redis://127.0.0.1:6379'
CELERY_ACCEPT_CONTENT = ['application/json']
CELERY_RESULT_SERIALIZER = 'json'
CELERY_TASK_SERIALIZER = 'json'
CELERY_STORE_ERRORS_EVEN_IF_IGNORED = True
CELERY_TASK_ALWAYS_EAGER = True

字符串
阅读this article关于celery 的问题是很大的帮助

tvz2xvvm

tvz2xvvm4#

我发现了问题:我在Windows上。
一旦我部署到heroku(运行在Linux上),它就工作了。

相关问题