tobiasbp/float_exporter repository overview

⁠float_exporter

An exporter for Prometheus exposing data from project management service Float at float.com

When running on localhost, float_exporter expose metrics at http://localhost:9709/metrics

⁠Metrics

The metrics exposed by float_exporter. The label days, is the number of days beeing looked in to the future when generating a report. A metric with days="7" would cover the period from the time the data was scraped, to days days in the future.

• float_accounts{account_type="['account_owner'|'admin'|'billing'|'member'|'project_manager']"}
• float_clients{}: Total number of clients.
• float_department_id{name="string"}: Mapping of department name to departmen_id.
• float_department_members{department_id="int"}: Number of people in department FIXME: Add department_id to people
• float_people{people_type=[1|2|3]}: Total number of people
• float_people_report_billable_hours{days="int",department_id="int"} Number of billable hours.
• float_people_report_capacity_hours{days="int",department_id="int"} Number of work hours available.
• float_people_report_nonbillable_hours{days="int",department_id="int"} Number of non billable hours.
• float_people_report_overtime_hours{days="int",department_id="int"} Number of overtime hours.
• float_people_report_scheduled_hours{days="int",department_id="int"} Number of scheduled hours.
• float_people_report_timeoff_hours{days="int",department_id="int"} Number of timeoff hours.
• float_people_report_unscheduled_hours{days="int",department_id="int"} Number of unscheduled hours.
• float_project_report_clients{days="int"} Number of clients worked for
• float_project_report_projects{days="int"} Number of projects worked on
• float_projects{active="[0|1]"} Total number of projects
• float_projects_billable{active="[0|1]"} Total number of billable projects
• float_projects_budget_sum{type"=[1|2|3]"} Sum of project budgets
• float_projects_with_budget{type"=[1|2|3]"} Number of projects with a budget
• float_tasks{days="int",priority="[0|1]",status="['complete'|'confirmed'|'tentative']} Number of tasks.
• float_tasks_hours{days="int"} Number of task hours.
• float_tasks_people{days="int"} Number of people with assigned tasks.
• float_up{} Is data beeing pulled from Float? 0 = no, 1 = yes.
• float_accounts{priority="[0|1]",status=['complete'|'confirmed'|'tentative']}

⁠Configuration

How to configure float_exporter.

⁠Parameters

The following parameters make up the configuration of float_exporter.

• email: An email where you can be contacted. Required by Float.
• user_agent: A string to use as User-Agent along with your contact email. Required by Float.
• log_file: The path to the configuration file you want to use.
• log_level: The level of logging
• access_token: The Float access token to use (Get in settings/integration in GUI)
• port: The port to accept traffic on. Default is 9709.
• report_days: A list of days in the future to create reports for.

⁠Priority

float_exporter has three sources of configuration. They are the following, in order of priority.

1. Arguments supplied via flags when running float_exporter
2. Environment variables
3. Configuration file

When setting a parameter using an environment variable, you should prefix the parameter name in uppercase with the string FLOAT_. To set the access token to use you would need to specify the environment variable FLOAT_ACCESS_TOKEN.

⁠Run in Kubernetes with helm

Install float_exporter in a Kubernetes cluster from a helm chart. You are assumed to have helm running on your machine. The following will install float_exporter in the namespace default.

1. Clone repository
2. Go to dir: cd helm
3. Copy config.yaml to config.local.yaml: cp config.yaml config.local.yaml
4. Update config.local.yaml with your configuration
5. Install float_exporter in your cluster: helm install -f config.local.yml my-float-exporter ./float_exporter
6. You should see my-float-exporter running: helm list

To uninstall: helm uninstall my-float-exporter

To see the Kubernetes files generated by helm without installing float_exporter in your cluster, run helm with flag --dry-run like this: helm install -f config.local.yml my-float-exporter ./float_exporter

⁠Run in Kubernetes

Run float_exporter as a service in a Kubernetes cluster managed by kubectl. This section refers to files in directory k8s. All files uses namespace monitoring. If you want to use a different namespace, you can change the configuration files accordingly.

1. Clone repository
2. Add namespace monitoring: kubectl create -f fe-namespace.yml
3. Copy fe-secret.yml to fe-secret.local.yml and add your Float access token
4. Create secret in Kubernetes: kubectl create -f fe-secret.local.yml
5. Copy fe-configmap.yml to fe-configmap.local.yml and change to you liking. You must add a valid email.
6. Create configmap in Kubernetes: kubectl create -f fe-configmap.local.yml
7. Create the fe-svc service and fe-dep deployment in Kubernetes: kubectl create -f fe-service.yml
8. Forward you local port 9709 to the float_exporter service: kubectl port-forward -n monitoring service/fe-svc 9709:9709
9. You should be able to see your metrics at http://localhost:9709/metrics

When running, you should see a pod, a deployment (with replicaset) and a service in the monitoring namespace:

kubectl get all -n monitoring

NAME                                  READY   STATUS      RESTARTS   AGE...
pod/fe-dep-xxx-xxx                    1/1     Running     0          18h

NAME                           TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)     AGE
service/fe-svc                 ClusterIP   xx.xx.xx.xx      <none>        9709/TCP    18h

NAME                             READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/fe-dep           1/1     1            1           18h

NAME                                        DESIRED   CURRENT   READY   AGE
replicaset.apps/fe-dep-xxxx                 1         1         1       18h

⁠Run locally

1. Clone repository
2. Install Python packages: pip3 install -r requirements.txt
3. Copy config file float_exporter.yml to float_exporter.local.yml
4. Update float_exporter.local.yml with your configuration.
5. Run float_exporter.py: ./float_exporter.py ./float_exporter.local.yml
6. You should see the log from float_exporter confirming a successful run.
7. You should be able to see your metrics at http://localhost:9709/metrics

⁠Run in Docker

⁠Build image

Build an image with tag float_exporter.

1. Clone repository
2. Build the image: docker build --tag float_exporter .

⁠Configuration in file

How to run with configuration in a file.

1. Copy float_exporter.yml to /example/path/float_exporter.yml (An example path).
2. Update /example/path/float_exporter.yml with your configuration.
3. Run the container: docker run --rm -p 9709:9709 -v /example/path/float_exporter.yml:/etc/float_exporter.yml float_exporter
4. You should see the log from float_exporter confirming a successful run.
5. You should be able to see your metrics at http://localhost:9709/metrics

⁠Configuration with environment variables

If you don't want to map an external config file to the container, you can pass environment variables for the container to use in stead of the values in the configuration file /etc/float_config.yml. Here is an example of setting the access token to be used by float_exporter.

• docker run --rm --env FLOAT_ACCESS_TOKEN=****** -p 9709:9709 float_exporter

You should probably always use the flag --rm to remove the container when stopping float_exporter. If you don't do that, you will end up with a lot of unused containers. If you want to keep the log file, you should map an external directory to /var/log in the container as a volume.