Cosmos-SDK wrong Block.Header.LastResultsHash

[silverstr89]

Summary of Bug

it really blows my mind, don't understand how to resolve it.
i getting error

7:49PM INF committed state app_hash=F839B3EA4DC197AD214F7448EFFA3EB1BB850ABB8AF4B690B441CF706C7BA915 
    
height=13257 module=state num_txs=1 panic: Failed to process committed block (13258:78A87F5E0CAE0391A9A3DBCD961460BBB8BE0B811DDF3D50544B855DF264FB64): wrong Block.Header.LastResultsHash.  

Expected 8506B0BD2A9D9C937A487FAEBFA118081458E7D9A5B6B86BC7417AFA751939C3, 

got CA1650098D49306DC6139DA6A90ED10D9C543C57ACEACA24EB309254B9E03E43

when i trying to connect p2p.

also tried to reset data, and delete it all - nothing helps.

could you tell me to what direction need to think?

Version

v0.44.5

Steps to Reproduce

appd start --p2p.persistent_peers

---

For Admin Use

• Not duplicate issue
• Appropriate labels applied
• Appropriate contributors tagged
• Contributor assigned/self-assigned

[tac0turtle]

hey, thanks for the issue, what app are you using?

[facundomedica]

This looks familiar, what OS are you running on your nodes? I had the same issue when starting a node on my Macbook: #11726

[silverstr89]

@marbar3778 it means my starlink based application

[silverstr89]

@facundomedica mac os , arm64. i'm not sure it related to arm64, because on other device (not mac and not arm64) i see same issue

[facundomedica]

@silverstr89 Could you list your nodes and which archs are each one? (if possible, I assume it's a testnet). Also, what other device has this issue? Does it happens at the same block height?

I'm pretty sure that is related to the issue I linked, so if the nodes are in different architectures, you will definitely see this bug. For a detailed explanation check here: #11726
It will get fixed once this PR gets backported #11737 (which is not being backported to v0.46, so we'll have to wait a while for it to land).

[tac0turtle]

this usually means there is a gas issue in your app causing non deterministic usage of gas