S3 folder structure possibly wrong

[Eric Ferreira]

Hey, @allada thanks for your excellent work on this repo.
I've been using the GETH client for quite a long time, and it becomes harder to maintain due to its size.
I've been running some tests with Erigon for the past week and I need a full archival node with the whole chain data since the genesis block, and I thought about using your snapshot to speed thing up. However, I noticed something "wrong" with the s3 folder structure or maybe I'm misunderstanding something.

Within the s3://public-blockchain-snapshots/bsc/erigon/archive/latest/v1/nodes/ there are like 19 levels of /nodes under it. Is this really expected? I mean, there's like s3://public-blockchain-snapshots/bsc/erigon/archive/latest/v1/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/.

[Nathan (Blaise) Bruer]

Yeah that's an error. But the nodes database is not required, it only holds info about previously connected clients and ip info.

…

-------- Original Message --------

On Feb 7, 2023, 12:17 PM, Eric Ferreira wrote: Hey, ***@***.***(https://github.com/allada) thanks for your excellent work on this repo. I've been using the GETH client for quite a long time, and it becomes harder to maintain due to its size. I've been running some tests with Erigon for the past week and I need a full archival node with the whole chain data since the genesis block, and I thought about using your snapshot to speed thing up. However, I noticed something "wrong" with the s3 folder structure or maybe I'm misunderstanding something. Within the s3://public-blockchain-snapshots/bsc/erigon/archive/latest/v1/nodes/ there are like 19 levels of /nodes under it. Is this really expected? I mean, there's like s3://public-blockchain-snapshots/bsc/erigon/archive/latest/v1/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/nodes/. — Reply to this email directly, [view it on GitHub](#50), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/AAN7CIX5BDISYIM7TMKN54TWWKGVFANCNFSM6AAAAAAUUJQRFM). You are receiving this because you were mentioned.Message ID: ***@***.***>

[Eric Ferreira]

Hmmm. So in this case, if I want to build a node from scratch, the only things I'd need to make good use of this snapshot are the /snapshots, parlia, and chaindata dirs?

[Nathan (Blaise) Bruer]

Exactly... The node directory is only to make resuming a sync slightly faster, so it doesn't need to spend as much time finding nodes.

…

-------- Original Message --------

On Feb 7, 2023, 12:29 PM, Eric Ferreira wrote: Hmmm. So in this case, if I want to build a node from scratch, the only things I'd need to make good use of this snapshot are the /snapshots, parlia, and chaindata dirs? — Reply to this email directly, [view it on GitHub](#50 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/AAN7CIRBI3MNRYA3OS2ICM3WWKIB3ANCNFSM6AAAAAAUUJQRFM). You are receiving this because you were mentioned.Message ID: ***@***.***>

[Eric Ferreira]

Awesome. Thank you!