Rasa: Issue with database and RASA connection
Rasa Core version: 0.13.0
Python version: 0.14.1
Operating system (windows, osx, ...): Linux
Issue:
I am facing some issues while trying to fetch data from MySQL database.
I am getting an error as follows:
2019-03-11 17:37:50 DEBUG rasa_core.processor - Predicted next action 'action_db' with prob 1.00.
2019-03-11 17:37:50 DEBUG rasa_core.actions.action - Calling action endpoint to run action 'action_db'.
2019-03-11 17:37:50 ERROR rasa_core.actions.action - Failed to run custom action 'action_db'. Action server responded with a non 200 status code of 404. Make sure your action server properly runs actions and returns a 200 once the action is executed. Error: 404 Client Error: NOT FOUND for url: http://localhost:5055/webhook/
2019-03-11 17:37:50 ERROR rasa_core.processor - Encountered an exception while running action 'action_db'. Bot will continue, but the actions events are lost. Make sure to fix the exception in your custom code.
2019-03-11 17:37:50 DEBUG rasa_core.processor - Failed to execute custom action.
Content of domain file (if used & relevant):
slots:
personid:
type: text
stockid:
type: text
matches:
type: unfeaturized
intents:
- greet
- bye
- affirmative
- negative
- personid
- stockid
- inform
entities:
- personid
- stockid
templates:
utter_greet:
- 'Hello! How can I help?'
utter_goodbye:
- 'Talk to you later.'
- 'Bye bye :('
utter_ask_personid:
- 'Tell me your id please'
- 'May I know your id please?'
utter_ask_stock:
- 'Tell me the stock name please'
- 'Could you please tell me the stock name?'
actions:
- utter_greet
- utter_goodbye
- utter_ask_stock
- utter_ask_personid
- action_db
Content of action file :
from __future__ import absolute_import
from __future__ import division
from __future__ import unicode_literals
from rasa_core_sdk import Action
from rasa_core_sdk.events import SlotSet
import MySQLdb
class ActionDb(Action):
def name(self):
return "action_db"
def run(self, dispatcher, tracker, domain):
db = MySQLdb.connect("localhost","root","root","DBforChatbot")
cursor = db.cursor()
PersonID = tracker.get_slot('personid')
StockID = tracker.get_slot('stockid')
q = "select StockName, StockBalance from Persons WHERE (PersonID = '%d' and StockName = '%s');" % (PersonID, StockID)
try:
cursor.execute(q)
if cursor.execute(q)==0:
print("Sorry, could not find any relevant data. Please contact 1234 for further assistance.")
results = cursor.fetchall()
not all(results)
for row in results:
stockname = row[0]
stockbal = row[1]
# Now print fetched result
details = ("stockname=%s,stockbal=%d" % (stockname,stockbal))
print(details)
except:
print ("Error: unable to fetch data")
db.close()
response = """The detail is as follows {}.""".format(results)
dispatcher.utter_message(response)
return [SlotSet('stockid',StockID)]
Content of endpoints file (if used & relevant):
action_endpoint:
url: "http://localhost:5055/webhook/"
Content of dialogue_management_model file:
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals
import logging
import rasa_core
from rasa_core.agent import Agent
from rasa_core.policies.keras_policy import KerasPolicy
from rasa_core.policies.memoization import MemoizationPolicy
from rasa_core.interpreter import RasaNLUInterpreter
from rasa_core.utils import EndpointConfig
from rasa_core.run import serve_application
from rasa_core import config
logger = logging.getLogger(__name__)
def train_dialogue(domain_file = 'testbot_domain.yml',
model_path = './models/dialogue',
training_data_file = './data/stories.md'):
agent = Agent(domain_file, policies = [MemoizationPolicy(), KerasPolicy(max_history=3, epochs=200, batch_size=50)])
data = agent.load_data(training_data_file)
agent.train(data)
agent.persist(model_path)
return agent
def run_test_bot(serve_forever=True):
interpreter = RasaNLUInterpreter('./models/nlu/default/testbotnlu')
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook/")
agent = Agent.load('./models/dialogue', interpreter=interpreter, action_endpoint=action_endpoint)
rasa_core.run.serve_application(agent ,channel='cmdline')
return agent
if __name__ == '__main__':
train_dialogue()
run_test_bot()
Hi @tmbo @akelad @EPedrotti , could you please help here. Thanks.
Ujjwalsas9
All 11 comments
Did you run the actions server before starting interactive mode?
python -m rasa_core_sdk.endpoint --actions actions
https://rasa.com/docs/core/customactions/#custom-actions-written-in-python
ortsaCniveK
on 11 Mar 2019
@ortsaCniveK Yes, I ran it. But, I did not run it in interactive mode. I just trained it using rasa nlu and core commands as follows:
python3 nlu_model.py
python3 -m rasa_core.train -d testbot_domain.yml -s data/stories.md -o models/dialogue/
python3 -m rasa_core.run -d models/dialogue -u models/nlu/default/testbotnlu --endpoints endpoints.yml --debug --enable_api
Ujjwalsas9
on 11 Mar 2019
You'll want to have the actions endpoint running in order for interactive mode to establish connection to it, which will allow for action_db to work. That's why it's 404'ing.
so running inteactive mode would be likely similar to below from here:
python -m rasa_core_sdk.endpoint --actions actions& # runs action server in the background
python -m rasa_core.train \
interactive -o models/dialogue \
-d domain.yml -c policy_config.yml \
-s data/stories.md \
--nlu models/default/testbotnlu \
--endpoints endpoints.yml # starts interactive mode, pointing to the running action server
I have executed all the commands which you have mentioned (though in my case, it was:
python3 -m rasa_core_sdk.endpoint --actions action, as my action file's name is action.py).
I am not using interactive training to train the dialogue model. I am running the below commands to train and run code:
python3 -m rasa_core.train -d testbot_domain.yml -s data/stories.md -o models/dialogue/
python3 -m rasa_core.run -d models/dialogue -u models/nlu/default/testbotnlu --endpoints endpoints.yml --debug --enable_api
Ujjwalsas9
on 12 Mar 2019
I do not know, if I am right or not, but my rasa server is running on port 5055 and my sql server is running on port 3306. I have specied rasa's port in endpoints.yml file but do I have to specify sql's port as well?
Thanks
Ujjwalsas9
on 12 Mar 2019
Below are my story and training data (for reference):
Stories:
## Generated Story 1
* greet
- utter_greet
* inform
- utter_ask_personid
* personid{"personid": "1"}
- utter_ask_stock
* stockid{"stockid": "SB123"}
- action_db
* bye
- utter_goodbye
data.json (training data):
{
"rasa_nlu_data": {
"regex_features": [],
"entity_synonyms": [],
"common_examples": [
{
"text": "hello",
"intent": "greet",
"entities": []
},
{
"text": "howdy",
"intent": "greet",
"entities": []
},
{
"text": "hola",
"intent": "greet",
"entities": []
},
{
"text": "adieu",
"intent": "bye",
"entities": []
},
{
"text": "goodbye",
"intent": "bye",
"entities": []
},
{
"text": "gtg",
"intent": "bye",
"entities": []
},
{
"text": "nope",
"intent": "negative",
"entities": []
},
{
"text": "no",
"intent": "negative",
"entities": []
},
{
"text": "no, thanks",
"intent": "negative",
"entities": []
},
{
"text": "si",
"intent": "affirm",
"entities": []
},
{
"text": "ok",
"intent": "affirm",
"entities": []
},
{
"text": "right",
"intent": "affirm",
"entities": []
},
{
"text": "hola ]I would like to know the total stk bal for the stocks ST4053.",
"intent": "inform",
"entities": [
{
"end": 66,
"entity": "stockid",
"start": 60,
"value": "ST4053"
}
]
},
{
"text": "Tell me the stock balance for stocks SB123",
"intent": "inform",
"entities": [
{
"end": 42,
"entity": "stockid",
"start": 37,
"value": "SB123"
}
]
},
{
"text": "I would like to know the total stk blnc for the stock AB1233.",
"intent": "inform",
"entities": [
{
"end": 60,
"entity": "stockid",
"start": 54,
"value": "AB1233"
}
]
},
{
"text": "thank you",
"intent": "thanks",
"entities": []
},
{
"text": "Thank you",
"intent": "thanks",
"entities": []
},
{
"text": "Thanks",
"intent": "thanks",
"entities": []
},
{
"text": "1",
"intent": "personid",
"entities": [
{
"end": 1,
"entity": "personid",
"start": 0,
"value": "1"
}
]
},
{
"text": "its 2",
"intent": "personid",
"entities": [
{
"end": 5,
"entity": "personid",
"start": 4,
"value": "2"
}
]
},
{
"text": "it is 3",
"intent": "personid",
"entities": [
{
"end": 7,
"entity": "personid",
"start": 6,
"value": "3"
}
]
},
{
"text": "it is ST4053",
"intent": "stockid",
"entities": [
{
"end": 12,
"entity": "stockid",
"start": 6,
"value": "ST4053"
}
]
},
{
"text": "its AB1233",
"intent": "stockid",
"entities": [
{
"end": 10,
"entity": "stockid",
"start": 4,
"value": "AB1233"
}
]
},
{
"text": "ST4053",
"intent": "stockid",
"entities": [
{
"end": 6,
"entity": "stockid",
"start": 0,
"value": "ST4053"
}
]
},
{
"text": "hi",
"intent": "greet",
"entities": []
},
{
"text": "bye",
"intent": "bye",
"entities": []
},
{
"text": "no thanks",
"intent": "negative",
"entities": []
},
{
"text": "yes to that",
"intent": "affirm",
"entities": []
},
{
"text": "I wanna know about the total number of stock AB1233 in my account.",
"intent": "inform",
"entities": [
{
"end": 51,
"entity": "stockid",
"start": 45,
"value": "AB1233"
}
]
},
{
"text": "tnxs",
"intent": "thanks",
"entities": []
},
{
"text": "it is 2",
"intent": "personid",
"entities": [
{
"end": 7,
"entity": "personid",
"start": 6,
"value": "2"
}
]
},
{
"text": "its SB-123",
"intent": "stockid",
"entities": [
{
"end": 10,
"entity": "stockid",
"start": 4,
"value": "SB-123"
}
]
}
]
}
}
MySQL database and table (for reference):
mysql> show databases;
+--------------------+
| Database |
+--------------------+
| information_schema |
| DBforChatbot |
| mysql |
| performance_schema |
| sys |
+--------------------+
5 rows in set (0.00 sec)
mysql> use DBforChatbot;
Reading table information for completion of table and column names
You can turn off this feature to get a quicker startup with -A
Database changed
mysql> select * from Persons;
+----------+----------+-----------+-----------+--------------+
| PersonID | LastName | FirstName | StockName | StockBalance |
+----------+----------+-----------+-----------+--------------+
| 2 | Anand | Utsav | SB-123 | 231045 |
| 1 | Anand | Ujjwal | SB123 | 231095 |
| 1 | Anand | Ujjwal | sb123 | 231095 |
+----------+----------+-----------+-----------+--------------+
3 rows in set (0.00 sec)
Can you post the log of the action server?
akelad
on 12 Mar 2019
Hi @akelad , thanks for quick response.
Below are the logs of action server (now I am getting error 500):
python3 -m rasa_core_sdk.endpoint --actions action
2019-03-12 14:33:54 INFO __main__ - Starting action endpoint server...
2019-03-12 14:33:54 INFO rasa_core_sdk.executor - Registered function for 'action_db'.
2019-03-12 14:33:54 INFO __main__ - Action endpoint is up and running. on ('0.0.0.0', 5055)
[2019-03-12 14:35:52,091] ERROR in app: Exception on /webhook [POST]
Traceback (most recent call last):
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask/app.py", line 1982, in wsgi_app
response = self.full_dispatch_request()
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask/app.py", line 1614, in full_dispatch_request
rv = self.handle_user_exception(e)
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask_cors/extension.py", line 161, in wrapped_function
return cors_after_request(app.make_response(f(*args, **kwargs)))
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask/app.py", line 1517, in handle_user_exception
reraise(exc_type, exc_value, tb)
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask/_compat.py", line 33, in reraise
raise value
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask/app.py", line 1612, in full_dispatch_request
rv = self.dispatch_request()
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask/app.py", line 1598, in dispatch_request
return self.view_functions[rule.endpoint](**req.view_args)
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/flask_cors/decorator.py", line 128, in wrapped_function
resp = make_response(f(*args, **kwargs))
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/rasa_core_sdk/endpoint.py", line 86, in webhook
response = executor.run(action_call)
File "/home/xmplar/anaconda3/lib/python3.6/site-packages/rasa_core_sdk/executor.py", line 177, in run
events = action(dispatcher, tracker, domain)
File "/home/xmplar/xfact/ChatBot/RasaBot/examples/testbot/action.py", line 73, in run
q = "select StockName, StockBalance from Persons WHERE (PersonID = '%d' and StockName = '%s');" % (PersonID, StockID)
TypeError: %d format: a number is required, not str
127.0.0.1 - - [2019-03-12 14:35:52] "POST /webhook HTTP/1.1" 500 412 0.012302
Thanks!
Ujjwalsas9
on 12 Mar 2019
Thanks a lot @akelad ! After I posted the logs, I noticed that, it was because of type conversion, my bad!
Ujjwalsas9
on 12 Mar 2019
Hi Ujjwalsas9 can you elaborate what you fixed to get this resolved.
dezykum
on 14 May 2019
Can @Ujjwalsas9 please tell me what you fixed in this? I am doing similar kind of thing.
siddharthchauhan
on 17 Feb 2020
Related issues
johnson7788
路
3Comments
igormiranda001
路
3Comments
edouardlp
路
3Comments
rayush7
路
3Comments
alonsopg
路
3Comments