tobiasbp/ldap2float repository overview

⁠ldap2float

A tool for syncing data available via LDAP with the project management service Float at float.com. No data is written to the LDAP server.

⁠Run locally

• Clone repository
• Install Pythob packages: pip3 install -r requirements.txt
• Copy ldap2float.conf to ldap2float.local.conf
• Update ldap2float.local.conf with your configuration.
• Run ldap2float: ./ldap2float ./ldap2float.local.conf

⁠Run in Docker

• Clone repository
• Copy ldap2float.conf to /example/path/ldap2float.conf (An example path).
• Update /example/path/ldap2float.conf with your configuration.
• Build the image: docker build --tag ldap2float .
• Run a container to test: docker run --rm -v /example/path/ldap2float.conf:/etc/ldap2float.conf ldap2float
• You should see the log from ldap2float confirming a successful run.
• Set up a cron job to run the container as in the example above.

Using the flag --rm removes the container after each run of ldap2float. If you don't do that, you will end up with a lot of unused containers.

⁠Run with Kubernetes

This section assumes you have a Kubernetes cluster up and running. Files mentioned in this section are in the directory called k8s.

• Clone repository
• Copy ldap2float-config.yml to ldap2float-config.local.yml and edit it to match your configuration.
• Create new namespace automation in Kubernetes: kubectl create -f ldap2float-namespace.yml
• Add your ldap2float configuration to Kubernetes: kubectl create -f ldap2float-config.yml
• Add cronjob to Kubernetes: kubectl create -f ldap2float-cronjob.yml

Let's see what's in the namespace automation: kubectl get all -n automation

You should see something like this:

NAME                       SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
cronjob.batch/ldap2float   */5 * * * *   False     0        <none>          4m29s

After the cron job has run successfully, you should see something like this:

NAME                              READY   STATUS      RESTARTS   AGE
pod/ldap2float-1591876800-6jwkj   0/1     Completed   0          110s

NAME                              COMPLETIONS   DURATION   AGE
job.batch/ldap2float-1591876800   1/1           24s        110s

NAME                       SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
cronjob.batch/ldap2float   */5 * * * *   False     0        111s            6m20s

Let's list all the pods in the namespace: kubect get pods -n automation

NAME                          READY   STATUS      RESTARTS   AGE
ldap2float-1591877400-f4wf8   0/1     Completed   0          11m
ldap2float-1591877700-p9h4f   0/1     Completed   0          6m44s
ldap2float-1591878000-p5wsk   0/1     Completed   0          102s

See the log for a pod: kubectl logs -n automation ldap2float-1591878000-p5wsk

2020-06-11 12:00:22,440:INFO:Running ldap2float.py
...
...
2020-06-11 12:00:23,596:INFO:Done running ldap2float.py

Kubernes will keep the pods for the last 3 runs of ldap2float. Older pods will be deleted automatically. Since the logs are a part of the completed jobs, you will only be able to see the logs for the last 3 jobs (In the default configuration)

⁠To do

• Format of expected date string must be configurable.
• Mapping of LDAP fields to Float fields must be configurable.
• Raise an exception if date from ldap could not be converted to valid Float date YYYY-MM-DD.
• Add support for sending events via WebHook.
• Add dry-run option from CLI. Should be default.
• Option for creating departmens in Float based on LDAP structure.
• Add configuration files for running with Docker.
• Add configuration files for running with Kubernetes as a cron job.
• If no LDAP group is specified, all users found beneath the LDAP base should be added to Float
• Run flake (or similar) to check code
• Accept configuration parameters as environment variables
• Allow for logging to stdout and file at the same time