I’m looking to setup a chirpstack suite for my network. I’m primarily a network administrator rather than systems/db admin. It looks like there’s only documentation for a single server doing all items though. What differences would I be looking at to have the DB on it’s own server? We’re looking to use Amazon Linux for the network/application server and Amazon RDS for the postgres DB. What items need to be installed on which servers? Does REDIS and MQTT broker have to be installed on the server hosting the DB? Or do they go onto the NS/AS server?
The architecture doc clearly explains the functions and respective config files of each needs to be changed and updated to include the distributed systems ips and authentications to ensure they talk to each other.
Not really, the architecture doc doesn’t mention REDIS at all so my question there still stands. Also it just shows the relationship between broker and gateway, not which server (DB or WEB) each should be installed on when doing a web server / DB server separately.
Couple of examples to nudge you in right directions. All of these parts can be separated which is why there is a config file to define where they are located.
# Redis configuration.
[redis]
# Server address or addresses.
#
# Use rediss:// in case of a TLS secured connection.
#
# Example formats:
# redis://127.0.0.1:6379
# rediss://127.0.0.1:6379
# redis://:password@127.0.0.1:6379
# redis://username:password@127.0.0.1:6379
#
# Set multiple addresses when connecting to a cluster.
servers=[
"redis://127.0.0.1/",
]
replace 127.0.0.1 with redis servers ip and so on.
Same with other applications defined in this config and other configs.