Slow Syncronization of Chain Index to Mainnet

[edu-d-rai]

Summary

I'm trying to sync the chain index to mainnet, and as suggested I've adjusted the Append transaction queue size to a 100 thousand, while this worked until block 6210741, from block 6210741 to 6518519 it took 2 days, so if the progress continues at the same rate, and does not slow down further, it will take 6 days to finish syncing.

Does anyone have a snapshot of the chain-index DB for mainnet?

Steps to reproduce the behavior

1. Start Cardano Node
2. Start Chain Index
3. Observe Chain Index Progress

Actual Result

Slow synchronization

Expected Result

Shouldn't take more than a day.

Describe the approach you would take to fix this

-

System info

Plutus App version https://github.com/input-output-hk/plutus-apps/commit/e4062bca213f233cdf9822833b07aa69dff6d22a (tag of April 6th, 2022) Cardano Node version 1.34.1 AWS instance with 32Gb of ram, and 8 cores.

[koslambrou]

6 days of syncing is expected with the current solution.

If you want faster syncing:

• you should set the resume point at a later slot that is actually relevant for your use case
• you can also clone plutus-chain-index and change the event handler to filter transactions you're not interested in
• you can clone plutus-chain-index-core and remove the indexers you're not interested in

We're in the process of rebuilding a more modularized chain-index, thus items 2 and 3 should not be needed. But that would take some time.

Does anyone have a snapshot of the chain-index DB for mainnet?

The team doesn't have such a snapshot.