This is a tool to sync data from different sources to a NetBox instance.
Available source types:
- VMware vCenter Server
- bb-ricardo/check_redfish inventory files
IMPORTANT: READ INSTRUCTIONS CAREFULLY BEFORE RUNNING THIS PROGRAM
A BIG thank-you goes out to Raymond Beaudoin for creating vcenter-netbox-sync which served as source of a lot of ideas for this project.
copied from Raymond Beaudoin
The NetBox documentation makes it clear the tool is intended to act as a "Source of Truth". The automated import of live network state is strongly discouraged. While this is sound logic we've aimed to provide a middle-ground solution for those who desire the functionality.
All objects collected from vCenter have a "lifecycle". Upon import, for supported object types,
they are tagged NetBox-synced to note their origin and distinguish them from other objects.
Using this tagging system also allows for the orphaning of objects which are no longer detected in vCenter.
This ensures stale objects are removed from NetBox keeping an accurate current state.
- python >= 3.6
- packaging
- requests==2.24.0
- pyvmomi==6.7.3
- aiodns==2.0.0
- NetBox >= 2.9
- VMWare vCenter >= 6.0
- check_redfish >= 1.2.0
- here we assume we install in
/opt
- on RedHat/CentOS 7 you need to install python3.6 and pip from EPEL first
- on RedHat/CentOS 8 systems the package name changed to
python3-pip
yum install python36-pip
apt-get update && apt-get install python3-venv
- download and setup of virtual environment
cd /opt
git clone https://github.com/bb-Ricardo/netbox-sync.git
cd netbox-sync
python3 -m venv .venv
. .venv/bin/activate
pip3 install -r requirements.txt || pip install -r requirements.txt
The vsphere-automation-sdk must be installed if tags should be synced from vCenter to NetBox
- assuming we are still in an activated virtual env
pip install --upgrade pip setuptools
pip install --upgrade git+https://github.com/vmware/vsphere-automation-sdk-python.git
Run the application in docker container
- The application working directory is
/opt - Required to mount your
settings.ini
docker build -t netbox-sync .
docker run --rm -it -v $(pwd)/settings.ini:/opt/settings.ini netbox-sync [some args...]
Run the containerized application in a kubernetes cluster
- Build the container image following the docker instructions above
- Tag and push the image to a container registry you have access to
- Create a secret from the settings.ini
- Update the image field in the manifest
- Deploy the manifest to your k8s cluster and check the job is running
docker build -t netbox-vsphere-sync .
docker image tag netbox-vsphere-sync your-registry.host/netbox-vsphere-sync:v1.2.0
docker image push your-registry.host/netbox-vsphere-sync:v1.2.0
kubectl create secret generic netbox-vsphere-sync --from-file=settings.ini
kubectl apply -f netbox-vsphere-sync-cronjob.yaml
In order to updated data in NetBox you need a NetBox API token.
- API token with all permissions (read, write) except:
- auth
- secrets
- users
A short description can be found here
usage: netbox-sync.py [-h] [-c settings.ini]
[-l {DEBUG3,DEBUG2,DEBUG,INFO,WARNING,ERROR}] [-n] [-p]
Sync objects from various sources to NetBox
Version: 1.2.1 (2021-11-03)
Project URL: https://github.com/bb-ricardo/netbox-sync
optional arguments:
-h, --help show this help message and exit
-c settings.ini, --config settings.ini
points to the config file to read config data from
which is not installed under the default path
'./settings.ini'
-l {DEBUG3,DEBUG2,DEBUG,INFO,WARNING,ERROR}, --log_level {DEBUG3,DEBUG2,DEBUG,INFO,WARNING,ERROR}
set log level (overrides config)
-n, --dry_run Operate as usual but don't change anything in NetBox.
Great if you want to test and see what would be
changed.
-p, --purge Remove (almost) all synced objects which were create
by this script. This is helpful if you want to start
fresh or stop using this script.
It is recommended to set log level to DEBUG2 this way the program should tell you what is happening and why.
Also use the dry run option -n at the beginning to avoid changes directly in NetBox.
Copy the settings-example.ini sample settings file to settings.ini.
All options are described in the example file.
In Order to sync all items regularly you can add a cron job like this one
# NetBox Sync
23 */2 * * * /opt/netbox-sync/.env/bin/python3 /opt/netbox-sync/netbox-sync.py >/dev/null 2>&1
READ CAREFULLY
The program operates mainly like this
- parsing and validating config
- instantiating all sources and setting up connection to NetBox
- read current data from NetBox
- read data from all sources and add/update objects in memory
- Update data in NetBox based on data from sources
- Prune old objects
Request all current NetBox objects. Use caching whenever possible. Objects must provide "last_updated" attribute to support caching for this object type.
Actually perform the request and retry x times if request times out. Program will exit if all retries failed!
Check out the documentations for the different sources
If you have multiple vCenter instances or check_redfish folders just add another source with the same type in the same file.
Example:
[source/vcenter-BLN]
enabled = True
host_fqdn = vcenter1.berlin.example.com
[source/vcenter-NYC]
enabled = True
host_fqdn = vcenter2.new-york.example.com
[source/redfish-hardware]
type = check_redfish
inventory_file_path = /opt/redfish_inventoryDevelopers: If you are interested in adding more source types please open an issue/discussion because the documentation of implementing a new source hasn't been finished yet. 🤷
Prune objects in NetBox if they are no longer present in any source. First they will be marked as Orphaned and after X (config option) days they will be deleted from NetBox.
Objects subjected to pruning:
- devices
- VMs
- device interfaces
- VM interfaces
- IP addresses
All other objects created (i.e.: VLANs, cluster, manufacturers) will keep the source tag but will not be deleted. Theses are "shared" objects might be used by different NetBox objects
You can check out the full license here
This project is licensed under the terms of the MIT license.