atlas

MongoDB Atlas Integration for Honeycomb

STATUS: this project is archived.

We recommend using honeytail to move Atlas data into Honeycomb.

Questions? You can chat with us in the Honeycomb Pollinators Slack. You can find a direct link to request an invite in Spread the Love: Appreciating Our Pollinators Community.

Installation

go get github.com/honeycombio/atlas

Usage

Preparing MongoDB logging

Atlas does not currently allow you to set the log verbosity level. To enable logging, you will need to configure profiling on each DB that you are interested in collecting from. To do this, run the following in your Mongo shell:

dbnames = db.getMongo().getDBNames()
for (var k in dbnames) { adb = db.getMongo().getDB(dbnames[k]); adb.setProfilingLevel(2, -1); }

Atlas Permissions

Your user will need ownership permissions on the cluster to pull the logs. See Atlas Logs API for more information.

Running the Atlas integration

# Your Atlas API Key, configured at:
# https://cloud.mongodb.com/v2#/account/publicApi
API_KEY=abcd1234-ab12-ab12-1234567890ab
# Your Atlas Username
USERNAME=myuser@example.com
# This is the Atlas Group ID/Project ID
# You can find this in your Atlass "Project Settings" page
GROUP_ID=abc1def2abc1def2ab9999
# Your Honeycomb Write Key
WRITE_KEY=abcdef1234567890999999999
# Target Honeycomb Dataset
DATASET=atlas-logs
atlas \
  --api_key=${API_KEY} \
  --username=${USERNAME} \
  --group_id=${GROUP_ID} \
  --cluster=Cluster0 \
  --writekey=${WRITE_KEY} \
  --dataset=${DATASET}

You can specify more clusters by providing the --cluster parameter multiple times.

Kubernetes

Atlas is a good candidate for running inside k8s. You can modify the included spec file (kubernetes/atlas.example.yml) with the required arguments (USERNAME, API_KEY, WRITE_KEY, etc) and apply with kubectl apply:

kubectl apply -f kubernetes/atlas.example.yml