I am using the n3h server on heroku as a test of holorelay type connections
When using remote wss connection, IPC connects before P2P and starts messaging, but that causes an unhandled error to be thrown and the n3h process to die.
(@hackmode@) [t] Received IPC: {"method":"requestState"}
Strict Fail on Unhandled Promise Rejection
TypeError: Cannot read property 'getId' of undefined
at N3hHackMode._handleIpcJson (/Users/connor/code/holochain/n3h/lib/hackmode/index.js:153:25)
I was able to confirm that this same error does NOT occur if you just launch n3h from outside the conductor, just manually, because no IPC communications start to occur.
In my initial testing, I've found that adding the following to the top of _handleIpcJson prevents the process from dying, but networking still doesn't work afterwards. Seeing no gossip between nodes
if (!this._p2p) {
return
}
I have this config in my conductor-config.toml file
I would have to go back and check, based on all the work that's happened since. I haven't seen mention of anything that I specifically thought addressed this though...
I am using the n3h server on heroku as a test of holorelay type connections
When using remote wss connection, IPC connects before P2P and starts messaging, but that causes an unhandled error to be thrown and the n3h process to die.
I was able to confirm that this same error does NOT occur if you just launch n3h from outside the conductor, just manually, because no IPC communications start to occur.
In my initial testing, I've found that adding the following to the top of _handleIpcJson prevents the process from dying, but networking still doesn't work afterwards. Seeing no gossip between nodes
I have this config in my conductor-config.toml file
and my
network_config.json
file looks like this: