Polkadot v0.9.37 #1241
Polkadot v0.9.37 #1241
Conversation
|
@NunoAlexandre Some clean up is needed and another pass over the command & service on monday. But it compiles now. |
You rule buddy, thanks ✌🏻 |
|
From what I'm seeing in #1252, I think that the lint, doc, and integration failures here are all legit.@NunoAlexandre If you can get those cleaned up I think we can land this tomorrow? Everything we need to cut the release has landed in |
| @@ -1246,7 +1251,10 @@ impl pallet_xcm_transactor::Config for Runtime { | |||
| type CurrencyId = CurrencyId; | |||
| type CurrencyIdToMultiLocation = xcm::CurrencyIdConvert; | |||
| type DerivativeAddressRegistrationOrigin = EnsureRoot<AccountId>; | |||
| type HrmpEncoder = moonbeam_relay_encoder::westend::WestendEncoder; | |||
There was a problem hiding this comment.
Why do we use moonbeam here?
There was a problem hiding this comment.
Because the HrmpEncoder param type is a type of Moonbeam's pallet_xcm_transactor::Config and it requires knowledge about the way that, in this case, Westend encodes the Hrmp calls; We can also create a custom impl that just returns error with a NotImplement kind of error for example.
There was a problem hiding this comment.
mmh. Can you make a note. Just so that we know when we move something of this to production runtimes.
| if !collator_options.relay_chain_rpc_urls.is_empty() && !cli.relaychain_args.is_empty() { | ||
| warn!("Detected relay chain node arguments together with --relay-chain-rpc-urls. This command starts a minimal Polkadot node that only uses a network-related subset of all relay chain CLI options."); | ||
| } | ||
|
|
There was a problem hiding this comment.
This enables Parachain nodes to run a minimized relay node such that they do not have to fully sync with the relay chain to work, so in general they will use fewer system resources. For details, see Cumulus repo: https://github.com/paritytech/cumulus/blob/master/README.md#external-relay-chain-node
| let _overseer_handle = relay_chain_interface | ||
| .overseer_handle() | ||
| .map_err(|e| sc_service::Error::Application(Box::new(e)))?; | ||
|
|
There was a problem hiding this comment.
What is this unused handle for? The overseer handle usual is needed on the cumulus side of things but we never pass it down to the import queues or so.
There was a problem hiding this comment.
I am not sure. Maybe I copied it over from another project while trying to follow the way they have adapted to these changes in the service or maybe Miguel wrote it when he took over the work around the service. I feel very foreign to the whole service side of the chain; is this something you feel comfortable with? Cause maybe it's a good idea if you get yours hands on it and/or educate me along
There was a problem hiding this comment.
What I would like to get more info on is why we need to add a specific ParachainBlockImport everywhere now. What does the relay in this case pass down? What are the checks the parachain import does differently compared to the relay.
There was a problem hiding this comment.
I'm pretty sure that's the stuff Miguel worked on, I don't know the answer without digging further into this. If you got some time and will please do 🙏
There was a problem hiding this comment.
The ParachainBlockImport handles some of the logic around preventing clients from considering forks that haven't been backed by the relay chain. I'm not sure when it was introduced, but cumulus was updated in paritytech/cumulus#1559 to make it mandatory through the use of a marker trait.
There was a problem hiding this comment.
@NunoAlexandre I have no time to dig into the service and am out the next 8 days, so I am approving this one to not letting it linger around for too long.
Update to Polkadot v0.9.37
A small but noteworthy change is that, from now on, extrinsic are explicitly indexed with
#[pallet::call_index(_)], something we need to start doing for new extrinsic in the future and be mindful of not changing indexes, as doing so may break some integration (mostly an XCM concern but still worth keeping it in mind)❗ Blocked by #1233 (review)