I am using django for rest APIs.
I do have a logging code in settings.py.
LOGGING = {
'version': 1,
'disable_existing_loggers': False,
'formatters': {
'verbose': {
'format': '%(asctime)s %(levelname)s %(module)s %(process)d %(thread)d %(message)s'
},
'simple': {
'format': '%(asctime)s %(levelname)s %(message)s'
},
},
'handlers': {
'file': {
'level': 'ERROR',
'class': 'logging.FileHandler',
'filename': '/tmp/OAuthLog_Production.log',
'formatter': 'simple'
},
},
'loggers': {
'oauth': {
'handlers': ['file'],
'level': 'ERROR',
'propagate': True,
}
},
}
In the views.py it is being used like this whenever any exception occurs.
import logging
logger = logging.getLogger("oauth")
except Exception as ex:
logger.exception(ex)
response = helper.generateStandardResponse(ResponseCodes.exception_in_finding_user,
"Exception in finding user", data, False);
logger.info("get_profile_details : Exception in finding user")
return Response(response, status=status.HTTP_200_OK)
Since, I am using Elastic beanstalk to host the application I am not returning 500 status code. I am returning custom responses.
Even after this there are few 500 errors returned by application. And those are causing environment to "Degrade".
I checked my log file but it has errors like this.
2019-11-23 23:53:10,600 ERROR UserProfile matching query does not exist.
Traceback (most recent call last):
File "/opt/python/current/app/profiles/views.py", line 514, in get_profile_details
user_profile_detail = UserProfile.objects.get(mobile1=mobile, account_type=account_type)
File "/opt/python/run/venv/local/lib/python3.6/site-packages/django/db/models/manager.py", line 85, in manager_method
return getattr(self.get_queryset(), name)(*args, **kwargs)
File "/opt/python/run/venv/local/lib/python3.6/site-packages/django/db/models/query.py", line 385, in get
self.model._meta.object_name
profiles.models.DoesNotExist: UserProfile matching query does not exist.
I am not able to understand why there are 500 errors?
How can I log each and every request and response which is causing 500 errors?
So that I can debug and find out what input and client is causing the issues and resolve them.
You can write a middleware to catch all uncaught exceptions and return appropriate responses or implement custom exception handling as described here in DRF docs