I edited my config.json, changing “EndpointAddress”, then goal node restart.
Now the node is re-syncing… which is going to take a long time.
Is this expected behavior?
When you restarted the node, you lost synchronization with the blockchain. If the restart took only few seconds, then it means that you might have lost few seconds worth of blocks.
It should sync up quite quickly, as there shouldn’t be too many blocks required in order to catch up.
Yes, you describe the behavior I expected. Normally a restart takes a short time to re-synchronize.
I’m trying to figure out why, this once, the re-sync started from genesis.
I recall that I first edited “NetAddress” and restarted, then changed NetAddress back to “”, edited “EndpointAddress” and restarted again. (Edited NetAddress by mistake, EndPoint address is what I wanted.) Anyway, somehow during the restarts, algod decided to resync from genesis, and I was surprised (and disappointed) to see that.
Just a random guess (which might be wrong): Setting NetAddress makes the node a relay, which automatically makes it Archival. If your node was not archival before, algod had to start retrieving all the blocks from the beginning.
The default value for Archival was always false. However, it seems that we’re automatically turning on the Archival when NetAddress is not empty. I think this is a bug, but it wouldn’t be easy to fix it.
( i.e. since we would need to ensure all the relays are being deployed with a modified configuration file )