Increase Maximum Block Size to 1MB

Change log • 2023-09-30: initial post Summary GitHub ASA-2023-002: Default for `BlockParams.MaxBytes` consensus parameter may... ## Amulet Security Advisory for CometBFT: ASA-2023-002 **Component**: CometBFT **Criticality:** Low **Affected versions:** All **Affected users:** Validators, Chain Builders + Maintainers ... Details GitHub ASA-2023-002: Default for `BlockParams.MaxBytes` consensus parameter may... ## Amulet Security Advisory for CometBFT: ASA-2023-002 **Component**: CometBFT **Criticality:** Low **Affected versions:** All **Affected users:** Validators, Chain Builders + Maintainers ... Parameter and new value ``` { "title": "Reduce Maximum Block Size", "description": "This Proposal reduces the maximum block size pursuant to: https://github.com/cometbft/cometbft/security/advisories/GHSA-hq58-p9mv-338c", "changes": [ { "subspace": "baseapp", "key": "BlockParams", "value": { "max_bytes": "1048576" } } ], "deposit": "100000000uatom" } ``` Forum post link x.com…

Excerpt (1187 of 1982 characters). Read the whole post on the forum ↗

Out of curiosity because I am not with my laptop, what is the current maximum size? Can it be anything?

22 megabytes, which is much too large for our needs

Can you explain why 1MB is the perfect size for the hub?

I’m in the camp of ‘spam doesn’t exist on blockchains’, there are only transactions of various subjective values.

If we think there are transactions that are of null or negative value for the chain, the problem is the fee mechanism, not blocksize.

But 22MB does indeed feel way too large for a blockchain with 6 seconds block interval.

No, not really.

I’m not even sure there’s a perfect size but let me tell you what that might be:

1mb that can go up to 5mb if users pay way more fees

Or even

500kb that can go up to 5mb if users pay way more fees, let’s say the full 5mb should be 1000x the cost of the first 500kb.

That way there’s almost always room for more transactions but the more you store, the more you pay and the more stress you put on the network, the more you pay.

1mb just seemed to be a much better and safer approach than 200kb

In relation to what was disclosed yesterday from IcyCro - what would be the suggested course of action for this proposal?

It seems that passing it would INCREASE block size rather than decrease. Some can argue that this is good for future-proofing block sizes if and when more activity happens on the Hub however, for the purpose of this proposal I think it would be misleading to pass it given the current proposal description no?

Jerry Chong