Firestore 集合快照退出且永远不会再次启动 - `Thread-ConsumeBidirectionalStream exiting`

如何解决Firestore 集合快照退出且永远不会再次启动 - `Thread-ConsumeBidirectionalStream exiting`

出于某些工作目的，我需要在两个不同 Firestore 实例的两个集合之间重定向数据。为此，我创建了一个回调，将其作为参数传递给 collection.on_snapshot 方法。我无法显示整个函数，但它与下面的代码非常接近，至于我如何运行它，我遵循了 google exemple。

# Create a callback on_snapshot function to capture changes
def redirect_data(col_snapshot,changes,read_time):
    for change in changes:
        # Document update
        if change.type.name == 'MODIFIED':
            ts = datetime.datetime.now().astimezone()
            document_data = change.document.to_dict()
            document_id = change.document.id
            document_data['updated_at'] = ts
            for k,v in document_data.items():
                data_id = document_id + k
                data_to_add = {'updated_at': ts,'is_active': v['is_active']}
                logging.info('Updating slot : {0}'.format(data_id))
                try : 
                    # Updating data to another project
                    FIRESTORE_DATABASE.collection(COLLECTION) \
                        .document(data_id) \
                        .update(data_to_add)
                except:
                    # If the document does not exist we create it without
                    # forgetting to add the slot type
                    DB.collection(COLLECTION) \
                        .document(data_id) \
                        .set(data_to_add)
    callback_done.set()

当我运行我的代码时，一切正常，至少一开始是这样。然而，一段时间后（可能是几个小时或几天），处理此集合快照的线程退出并且不再启动。

这里有一些日志：

2021-06-11 15:28:49,102| INFO  | root| redirect_data()| L89  | Updating slot : slot_00002| call_trace=path_to_python_file/script.py| L89  
2021-06-11 15:28:49,161| INFO  | root| redirect_data()| L89  | Updating slot : slot_00004| call_trace=path_to_python_file/script.py| L89  
2021-06-11 15:28:50,307| INFO  | google.api_core.bidi| _thread_main()| L676 | Thread-ConsumeBidirectionalStream exiting| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L676 
2021-06-11 16:03:35,309| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 16:09:45,092| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 16:09:45,190| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 

在此事件之后，我所看到的只是重新打开线程的一些不成功尝试。

2021-06-11 16:27:25,161| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 16:27:25,235| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 17:03:35,433| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 17:03:35,507| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 17:09:45,731| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487 
2021-06-11 17:09:45,810| INFO  | google.api_core.bidi| _reopen()| L487 | Re-established stream| call_trace=path_to_venv/venv/lib/python3.7/site-packages/google/api_core/bidi.py| L487

我很难理解导致这种行为的原因，因为它没有固定的模式，而且我无法获得更详细的日志。

此脚本通过 sudo nohup ... & 命令在 1CPU/2GB 虚拟机上运行，​​并使用 Debian GNU/Linux 10。 我使用 python 3.7.3，这是我的 venv 规范：

Package                  Version
------------------------ ---------
CacheControl             0.12.6
cachetools               4.2.0
certifi                  2020.12.5
cffi                     1.14.4
chardet                  4.0.0
firebase-admin           4.5.0
google-api-core          1.24.1
google-api-python-client 1.12.8
google-auth              1.24.0
google-auth-httplib2     0.0.4
google-cloud-core        1.5.0
google-cloud-firestore   2.0.2
google-cloud-storage     1.35.0
google-crc32c            1.1.0
google-resumable-media   1.2.0
googleapis-common-protos 1.52.0
grpcio                   1.34.0
httplib2                 0.18.1
idna                     2.10
msgpack                  1.0.2
pip                      21.1.2
pkg-resources            0.0.0
proto-plus               1.13.0
protobuf                 3.14.0
pyasn1                   0.4.8
pyasn1-modules           0.2.8
pycparser                2.20
python-dotenv            0.17.1
pytz                     2020.5
requests                 2.25.1
rsa                      4.7
setuptools               40.8.0
six                      1.15.0
uritemplate              3.0.1
urllib3                  1.26.2

更新

我有一个开发环境和一个生产环境，所以脚本运行了两次，每个实例都在一个特定的虚拟机上。 当我在生产环境中运行脚本时，问题开始出现。 谢谢