skip to navigation
skip to content

celery-cloudwatch 1.1.1

A monitor for celery queues that reports to AWS CloudWatch

Celery CloudWatch

Monitor your celery application from within AWS CloudWatch!


The following events are tallied per task:

  • CeleryEventSent
  • CeleryEventStarted
  • CeleryEventSucceeded
  • CeleryEventFailed

You can then see how many tasks/day, tasks/week etc are being completed.

Also, we record the total waiting/running per task:

  • CeleryNumWaiting
  • CeleryNumRunning

Also, statistics on task duration are sent in the metrics:

  • CeleryQueuedTime
  • CeleryProcessingTime

These metrics are sent with all supported stats (No. Events, Sum, Max, Min), allowing you to gain insight into your task processing and match requests and capacity.

Getting Started

  1. Set up an IAM Role for your instance.

    It must include a policy to perform ‘PutMetricData’, eg:

      "Version": "2000-01-01",
      "Statement": [
          "Effect": "Allow",
          "Action": [
          "Resource": [

    (Note: Alternatively, you can set up a User with the same policy and provide access details that way)

  2. Install via python-pip (and upgrade pip & boto)

    sudo apt-get install -y python-pip
    sudo pip install --upgrade pip boto
    # Install directly
    sudo pip install celery-cloudwatch
    # OR, install in a virtualenv
    sudo apt-get install -y python-virtualenv
    mkdir /var/python-envs
    virtualenv /var/python-envs/ccwatch
    source /var/python-envs/ccwatch/bin/activate
    pip install celery-cloudwatch
  3. Create your own boto.cfg at /etc/boto.cfg-

    # if not using an IAM Role - provide aws key/secret
    aws_access_key_id = xxx
    aws_secret_access_key = yyy
    cloudwatch_region_name = my-region
    cloudwatch_region_endpoint =
  4. Create your own config file in /etc/ccwatch.yaml

      broker: null
      camera: celery_cloudwatch.CloudWatchCamera
      verbose: no
      frequency: 60.0
      verbose: no
      dryrun: no
      namespace: celery
        - myapp.mytasks.taskname
        - myapp.mytasks.anothertask
        - myapp.mytasks.thirdtask
        - name: myapp.secondarytasks
            task: myapp.secondarytasks
            customDim: value
        - name: myapp.tertiarytasks
            task: myapp.tertiarytasks
            customDim: value
  5. Install upstart

    Create a file /etc/init/celery-cloudwatch.conf-

    description "Celery CloudWatch"
    author "nathan muir <>"
    setuid nobody
    setgid nogroup
    start on runlevel [234]
    stop on runlevel [0156]
    exec /var/python-envs/ccwatch/bin/ccwatch


    sudo initctl reload-configuration
    sudo service celery-cloudwatch start
  6. Start Celery your celery workers with the -E (or CELERY_SEND_EVENTS=1 and CELERY_TRACK_STARTED=1) options, and, start celery clients with CELERY_SEND_TASK_SENT_EVENT=1

  7. All done! head over to your CloudWatch monitoring page to see the results!

File Type Py Version Uploaded on Size
celery-cloudwatch-1.1.1.tar.gz (md5) Source 2017-04-04 10KB