I'm using Python and Flask, served by Waitress, to host a POST API. I'm calling the API from a C# program that posts data and gets a string response. At least 95% of the time, it works fine, but sometimes the C# program reports an error:
(500) Internal Server Error.
There is no further description of the error or why it occurs. The only clue is that it usually happens in clusters -- when the error occurs once, it likely occurs several times in a row. Without any intervention, it then goes back to running normally.
Since the error is so rare, it is hard to troubleshoot. Any ideas as to how to debug or get more information? Is there error handling I can do from either the C# side or the Flask/Waitress side?
Your flask application should be logging the exception when it occurs. Aside from combing through your logs (which should be stored somewhere centrally) you could consider something like Sentry.io, which is pretty easy to setup with Flask apps.
Flask supports registering error handlers. Defining error handlers enables customizing the server error logs as well as returning more meaningful information to the client.
import logging
import traceback
from flask import Flask, jsonify
from werkzeug.exceptions import HTTPException
app = Flask(__name__)
logger = logging.getLogger()
@app.errorhandler(HTTPException)
def handle_http_exception(error):
error_dict = {
'code': error.code,
'description': error.description,
'stack_trace': traceback.format_exc()
}
log_msg = f"HTTPException {error_dict.code}, Description: {error_dict.description}, Stack trace: {error_dict.stack_trace}"
logger.log(msg=log_msg)
response = jsonify(error_dict)
response.status_code = error.code
return response