zkEVM/get-started/state-sync-snapshots/

[giscus[bot]]

zkEVM/get-started/state-sync-snapshots/

https://docs.polygon.technology/zkEVM/get-started/state-sync-snapshots/

[bensacc]

are there supposed to be instructions here for hashdb import?

[kmurphypolygon]

Hi @bensacc thank you for the question. So if you change the script example to sync with the hashdb or bridgedb option, does it work?

(I'm in the process of updating this doc because it's not so clear, I agree) ...

[letsbangout]

Respectfully, this is a slow and dumb way to populate an SQL database which is going to do nothing but grow over time. Based on 'time' of approx. 65 minutes to execute this on the state db (a 10 GB snapshot), I extrapolate that a restore of the hashdb snapshot should take about 23 hours, all things being equal. Would it not be easier and much quicker to ship a new set of database table files and restore them without psql running, rather than piping raw SQL though 'psql'?

[letsbangout]

It turns out my estimate of time was off by quite a factor. I'm minutes away from the 48 hour mark, and my psql import is still running. The underlying machine is no slouch, it's an AMD 7402P's with 16 vCPUs dedicated to zkevm. To make matters worse, I'm concerned I may end up with a hole in my data or some kind of checksum issue, as I had the 'psql' import running out of a terminal window that I had to interrupt with Ctrl-C and then background the process, i.e. 'bg %1'. For reference, the 'bor' database for Polygon Mainnet is about 4.5TB and can be extracted in 8 to 10 hours. By my math, that makes this about 80x slower than Mainnet, when considering the quantity of data. Please chart a course to fix this somehow.

[kmurphypolygon]

Hi @letsbangout thank you for raising your justified concerns. I'm happy to tell you that we are currently in the process of improving this process and the doc too. Please could you check back in a few weeks time after we've deployed some new Erigon components on the cardona testnet.

[kmurphypolygon]

The doc is updated now.

[timweri]

I still don't see the instructions on how to use the bridgedb and hashdb files.

[kmurphypolygon]

Hi, thanks @timweri. Yes, these are still in progress. Please check back soon, thank you. For info @BrianSeong99

[achiko]

The testnet snapshot download URLs are broken:

<Error>
	<script id="argent-x-extension" data-extension-id="dlcobpjiigpikoobohmabehhmhfoodbb"/>
<script/>
	<Code>AccessDenied</Code>
	<Message>Access Denied</Message>
	<RequestId>2WC7ZFBE5ZMXW2MR</RequestId>
	<HostId>5CX7Hwen1wH8Ylc4Ir/gAG/48X8PL3VNj6HvF4/fkYJimzHMw/Quj2ys5kOkAz4S6Fi9Pw6hR5c=</HostId>
</Error>

• https://zkevm-pub.s3.eu-west-1.amazonaws.com/testnet-bridgedb.sql.gz (312MB)
• https://zkevm-pub.s3.eu-west-1.amazonaws.com/testnet-statedb.sql.gz (1.9GB)
• https://zkevm-pub.s3.eu-west-1.amazonaws.com/testnet-hashdb.sql.gz (49GB)

[EmpieichO]

Hi @achiko

Please note the recently added warning at the beginning of this doc.