You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
All sample configs for testnet should be congruent across the board. It should be sufficient to copy the example config to the correct place and be able to run as is.
Even though many of the ports are completely arbitrary, we should just have a table of "standard testnet ports" with a note like this:
These ports are arbitrary, but must be consistent within the network of evo node services. Feel free to change them if they conflict with other services on your system, or for your own whimsy.
And we should mark bugs for any ports that are hard-coded to be fixed in the respective codebase or message type (such as protxreg) as soon as reasonable.
Current Behavior
It's a mix mash.
Possible Solution
Create a table and place it prominently in all repos, and update the config files as soon as reasonable.
Service
Env
Address
Default Port
Hard-Coded Port
Public
core p2p
testnet
127.0.0.1
19999
-
YES
core rpc
testnet
127.0.0.1
19998
-
-
core zmq
testnet
127.0.0.1
18009
-
-
drive api json
testnet
127.0.0.1
2501
-
-
drive api grpc
testnet
127.0.0.1
2500
-
-
drive streams
testnet
127.0.0.1
2510
-
-
tenderdash p2p
testnet
127.0.0.1
26656
-
YES
tenderdash rpc
testnet
127.0.0.1
26657
-
-
envoy-to-dapi
testnet
0.0.0.0
10000
YES
YES
Use a common config profile files across multiple services rather than having duplicate config files with the same addresses and ports for each service.
This is a general, widespread issue across this repo as well as many other pieces of documentation.
The port numbers in the various sample config files are inconsistent, incongruent, and otherwise don't actually match up.
For example:
Expected Behavior
All sample configs for testnet should be congruent across the board. It should be sufficient to copy the example config to the correct place and be able to run as is.
Even though many of the ports are completely arbitrary, we should just have a table of "standard testnet ports" with a note like this:
And we should mark bugs for any ports that are hard-coded to be fixed in the respective codebase or message type (such as protxreg) as soon as reasonable.
Current Behavior
It's a mix mash.
Possible Solution
Create a table and place it prominently in all repos, and update the config files as soon as reasonable.
Use a common config profile files across multiple services rather than having duplicate config files with the same addresses and ports for each service.
Steps to Reproduce (for bugs)
Context
It's just a lot of mental overhead to go back and forth between configs.
Your Environment
Any.
The text was updated successfully, but these errors were encountered: