Anca Zamfir

> that is my [solution](https://github.com/falcons-x/ibc-rs/commit/8e3d2aef433cb2b64ed40ac837c8f34ef420937b) Current design is to fill a Tx with messages until either `max_msg_num` or `max_tx_size` are reached, whichever occurs first. The logic in your solution should...

> Apparently, the v1 mempool keeps [FIFO behavior](https://github.com/tendermint/tendermint/blob/v0.34.x/mempool/v1/mempool.go#L306) when the priorities are the same, so one short-term solution could be to make sure the relayer transactions end up with a...

Thanks for the pointer and notes! > There was discussions about should we set priority based on msg types or gas price/fee, there was no consensus on that, so the...

@mircea-c can you paste the hermes config for osmosis?

and for v0.15? was the diff only `gas_multiplier = 1.1` vs `gas_adjustment = 0.1`

i think `max_gas * price` is not enough to get a reasonable fee when simulation fails. v1.0.0-rc0: `max fee, for use in tx simulation: Fee { amount: "501uosmo", gas_limit: 5000000...

I think the fee used when simulation fails is too low. Maybe it was before although I don't understand the difference in the `max fee` logs if the only change...

@mircea found that there is an osmosis configuration parameter `min-gas-price-for-high-gas-tx = ".0025"` that causes a different/ higher gas price for Tx-es with high gas. The configured `max_gas` in hermes (used...

Based on the above I think these hermes config changes should help short term: ``` max_gas = 999999 gas_multiplier = 1.2 max_msg_num = 5 ```

Due to https://github.com/informalsystems/ibc-rs/issues/2477 I suggest using an even smaller `max_gas` until that is fixed. Maybe `max_gas = 800000`.