Currently i'm making a replay of the hive blockchain on my raspberry pi 5 and it's already running over 22hrs and only got at block height 20,090,000. I think it's makes no sense to let it run further cause i need to catch up to 101,401,366. Ideas? @forkyishere
You can use a snapshot and revive a node in a few hours...
Go into another Hived node, snapshot dump. Move the snapshot into another machine...
Restore it!
I did a full replay from an 8-core ARM processor, but I had 5x SSDs connected to it.
Thanks! Currently i dont know how to use the snapshot feature - more or less where to get it - or how to create my own and replay with it. We'll i guess give it a try once more during the day.
Replaying, at least you don't have to trust anyone else.
But either way, I was thinking, in case you had another hived node around already in sync... if so, you can just do...
hived --help is your friend... 😇🙃
It's very easy though...
Stop hived
./hived --dump-snapshot <snapshot_path>
Then copy the contents of the snapshot to another hived node.
./hived --load-snapshot <snapshot_path>
For both situations, if you add --exit-before-sync, it will exit after finishing creating/loading the snapshot...
Done
I think the recommendation is to use a snapshot and someone may make that available to download. I'll be interesting in how well a Pi can run Hive. Will it be a witness? How much storage does it have?
Yea, i think so aswell it's recommended - i was just giving a try to run it from block 1 to latest but, that just takes days. I think it's just needs very less storage if you have the option block-log-split to 0 or just 1 to just keep the latest state availabe - witness only.
We need a simple guide on how to do it all, so maybe you can make something from your experience. Someone suggested it's better to just get a mini PC that will much more powerful, but a Pi should use less power.
That's a long time, do you need to replay it all? Isn't it a lighter version for the witness or some sort?
