With this PR, nanorc serves the configuration over the network rather than relying on a shared file system.
This should work with and without k8s and no matter if the configuration was uploaded to the MongoDB configuration service.
There is also quite a bit of refactoring in cfgmgr.py, that code still is... not great.
I suggest testing this the following way:
running nanorc with ssh
same as above with a partition offset
uploading the configuration to MongoDB and running ssh nanorc with db://config-name as the name of the configuration
same as above with a partition offset
running nanorc with k8s
same as above with a partition offset
uploading the k8s configuration to MongoDB and running k8s nanorc with db://config-name as the name of the configuration
same as above with a partition offset
running 2 DAQ subsystems (with a topjson) using ssh
running 2 DAQ subsystems (with a topjson) using k8s
With this PR, nanorc serves the configuration over the network rather than relying on a shared file system. This should work with and without k8s and no matter if the configuration was uploaded to the MongoDB configuration service.
There is also quite a bit of refactoring in
cfgmgr.py
, that code still is... not great.I suggest testing this the following way:
db://config-name
as the name of the configurationdb://config-name
as the name of the configuration