Okay, I have written an AWS Lambda function which pulls data from an API and inserts the data into a DocumentDB database. When I connect to my cluster from the shell and run my python script it works just fine and inserts the data no problem.
But, when I implement the same logic into a lambda function is does not work. Below is an example of what would work in the shell but not through a Lambda function.
import urllib3
import json
import certifi
import pymongo
from pymongo import MongoClient
# Make our connection to the DocumentDB cluster
# (Here I use the DocumentDB URI)
client = MongoClient('mongodb://admin_name_here:<insertYourPassword>my_docdb_cluster/?ssl=true&ssl_ca_certs=rds-combined-ca-bundle.pem&retryWrites=false')
# Specify the database to use
db = client.my_db
# Specify the collection to use
col = db.my_col
col.insert_one({"name": "abcdefg"})
The above works just fine in the shell but when run in Lambda I get the following error:
[ERROR] ServerSelectionTimeoutError: my_docdb_cluster timed out, Timeout: 30s, Topology Description: <TopologyDescription id: ***********, topology_type: ReplicaSetNoPrimary, servers: [<ServerDescription (my_docdb_cluster) server_type: Unknown, rtt: None, error=NetworkTimeout(my_docdb_cluster timed out')>]>
From my understanding, this error is telling me that the replica set has no primary. But, that is not true there definitely is a primary in my replica set. Does anyone know what could be the problem here?
