This Python application is used in conjunction with the ACI Monitoring Stack to convert the ACI Object Model Database in a Graph Database.
This is done by following these steps:
- Load the ACI access details from the aci-exporter config map
- Generate an ACI Backup
- Export it locally
- Parse the backup in a set of CSV Files
- Load The backup inside Memgraph
Backup2Graph
and Memgraph
are sharing a common PVC so that the data processed by Backup2Graph
is directly accessible by Memgraph
This is taken care by the Helm chart automatically.
Currently Multiple Fabric are supported but only 1 snapshot is preserved.
You can use the docker file to build your own images however in my lab I have to deal with a proxy if you are as unfortunate as me follow these steps:
- If you are using in insecure registry (i.e. self signed certificate) set you key/cert/ca in the buildkitd config file.
- If you have a proxy set up your
http_proxy/https_proxy/no_proxy
environment variables in/etc/environment
docker buildx create --use --driver-opt '"env.http_proxy='$http_proxy'"' --driver-opt '"env.https_proxy='$https_proxy'"' --driver-opt '"env.no_proxy='$no_proxy'"' --config ./buildkitd.toml
For my build I just need https_proxy
but you can add the other options if needed.
docker login harbor.cam.ciscolabs.com -u <USER> -p <PASS>
docker build . --build-arg HTTPS_PROXY=$https_proxy --platform linux/amd64,linux/arm64 --push --tag harbor.cam.ciscolabs.com/library/backup2graph:test
If you push to a registry that needs authentication do
docker login <CREDENTIALS>