Glutton update script #2681

NachoPal · 2023-06-02T14:17:59Z

Script to update Gluttons Parachains. It allows to update Compute (percentage of the left over ref_time to consume during on_idle) and Storage (percentage of left over proof_size to consume during on_idle) storage items for a range of selected parachain ids.

Necessary to learn what are the limits of the Relay chain and subsequently the max number of Parachains it can handle.

sandreim

LGTM, only a few suggestions.

scripts/glutton/update/src/main.rs

scripts/glutton/update/Cargo.toml

rphmeier · 2023-06-02T20:11:40Z

What does it mean to "update" a glutton parachain? Needs more description.

bkchr · 2023-06-04T20:19:18Z

I have the strong feeling that this could be a 10 line polkadot js script?

NachoPal · 2023-06-05T10:11:23Z

What does it mean to "update" a glutton parachain? Needs more description.

Improved description

* Fix docs Signed-off-by: Alexandru Vasile <alexandru.vasile@parity.io> * Update client/relay-chain-rpc-interface/src/reconnecting_ws_client.rs Co-authored-by: Davide Galassi <davxy@datawok.net> * Update cargo.lock with substrate Signed-off-by: Alexandru Vasile <alexandru.vasile@parity.io> * Update Cargo.lock for polkadot Signed-off-by: Alexandru Vasile <alexandru.vasile@parity.io> --------- Signed-off-by: Alexandru Vasile <alexandru.vasile@parity.io> Co-authored-by: Davide Galassi <davxy@datawok.net> Co-authored-by: parity-processbot <>

NachoPal · 2023-06-05T14:15:16Z

I have the strong feeling that this could be a 10 line polkadot js script?

Yeah, it would be definitely shorter, and would have taken me also less time. I considered both options, but Joe preferred me to use subxt.

Now that is done, I would love it to see it merged so we can start deploying the Glutton Parachains as soon as possible.

gilescope

Like the use of subxt!

gilescope · 2023-06-07T09:06:21Z

My only concern is the Cargo.lock changes. At the moment subxt is bringing in another version of substrate. Can/should we persuade it to use the same version of substrate as cumulus does?

scripts/glutton/update/Cargo.toml

gilescope · 2023-06-07T09:55:07Z

scripts/glutton/update/Cargo.toml

+[dependencies]
+codec = { package = "parity-scale-codec", version = "3.4.0", default-features = false }
+subxt = { version = "0.29.0", features = ["unstable-metadata"] }
+sp-keyring = "23.0.0"


I think we can nix this dep, don't think it's needed. Seems to compile on my machine without it.

Suggested change

sp-keyring = "23.0.0"

jsdw · 2023-06-07T11:25:48Z

scripts/glutton/update/src/main.rs

+		.runtime_api()
+		.at_latest()
+		.await?
+		.call(runtime_api_call)


If you want to avoid the unstable-metadata thing you could also use a "raw" call to the same thing (assuming nonce is a u32, which is true in substrate/polkadot and I'd expect most other chains):

let next_nonce: u32 = client .runtime_api() .at_latest() .await? .call_raw("AccountNonceApi_account_nonce", Some(&account.encode())) .await?;

I'm actually going to move Subxt to do this too, and will likely expose a way to get this in the next version

jsdw · 2023-06-07T11:27:44Z

scripts/glutton/update/src/main.rs

+		.unwrap();
+
+	// Submit and watch the tx
+	let in_block = signed_tx.submit_and_watch().await.unwrap().wait_for_in_block().await.unwrap();


Just to note that .wait_for_in_block() doesn't guarantee that the block will be finalized; use .wait_for_finalized() for that, or even .wait_for_finalized_success() to also verify that the tx was successfully executed (ie an ExtrinsicSuccess event was produced from its inclusion)

jsdw · 2023-06-07T11:30:30Z

~~My only concern is the Cargo.lock changes. At the moment subxt is bringing in another version of substrate. Can/should we persuade it to use the same version of substrate as cumulus does?~~

~~Just to reiterate I said in DM; subxt won't bring in any substrate crates (except sp-core-hashing which is tiny) unless used with the "substrate-compat" feature flag so this hsouldn't matter :)~~

ggwpez · 2023-06-07T14:05:54Z

scripts/glutton/update/Cargo.lock

@@ -0,0 +1,4403 @@
+# This file is automatically @generated by Cargo.


@bkchr do you have any opinion on multiple Cargo.lock files per repo?
I still think its not a good idea and then should rather be its own repo.
The monorepo migration will not profit from this either.

If we do not add it to the main workspace, what is the issue for having other Cargo.lock files on their own package?

The dependabot does not update it (so the deps will be out dated) and our CI tooling like fmt, clippy will not work.

I do not mind at all to move it to its own repo. It is just a tool to help to achieve a higher goal which is Glutton Parachains management. It is not mandatory to include it in Cumulus.

But if it is true that adding subxt to the main workspace could potentially lead to multiple versions of the same crate, and at the same time, we do not want to have multiple Cargo.lock files, then it means we are banning subxt from Cumulus/monorepo.

The dependabot does not update it (so the deps will be out dated) and our CI tooling like fmt, clippy will not work.

Yes, I understand that, but there are other scripts using Polkadot Js with a yarn.lock, and those are not checked or updated either by the CI as far as I know. So I do not know why if we accept polkadot js we can not tolerate subxt getting outdated. I would understand it if it were part of the client/runtime.

I do not mind at all to move it to its own repo

I would personally favour this.

jsdw · 2023-06-07T16:17:03Z

My only concern is the Cargo.lock changes. At the moment subxt is bringing in another version of substrate. Can/should we persuade it to use the same version of substrate as cumulus does?

Just to reiterate I said in DM; subxt won't bring in any substrate crates (except sp-core-hashing which is tiny) unless used with the "substrate-compat" feature flag so this hsouldn't matter :)

Actually this isn't true; the substrate-compat feature is enabled by default, and you need the substrate crates currently to sign transactions, and Subxt needs to pull these from crates.io, so there's not currently a way to avoid the dupe deps at the mo!

I am working on adding some native signing support to Subxt so that you can do basic transaction signing like here without pulling in a load of substrate crates.

Alternately I'm hoping for a future where substrate crates are published regularly again, and polkadot/cumulus depend on the crates and not github master; it'll be easier to keep things like Subxt in line with the latest deps then :)

NachoPal · 2023-06-08T10:59:58Z

Closing in favor of: paritytech/parachain-utils#1

NachoPal added 14 commits May 30, 2023 17:02

backup

f34a59d

update glutton example works

1af213f

backup

a63bee2

update glutton example works

4615b8e

works for multi para

fbec97d

clean up dependancies

b5349bf

fix conflicts

c6afda3

fix cargo

2fa5be5

update cargo.lock

9d61bd5

ready for cli

c779b00

working

0301f0c

fmt

0750a64

relay chain arg added

c2d2982

fmt

14c0462

NachoPal added B0-silent Changes should not be mentioned in any release notes T7-system_parachains This PR/Issue is related to System Parachains. A0-please_review Pull request needs code review. C1-low PR touches the given topic and has a low impact on builders. labels Jun 2, 2023

sandreim approved these changes Jun 2, 2023

View reviewed changes

scripts/glutton/update/src/main.rs Show resolved Hide resolved

scripts/glutton/update/src/main.rs Show resolved Hide resolved

scripts/glutton/update/src/main.rs Show resolved Hide resolved

ggwpez reviewed Jun 2, 2023

View reviewed changes

scripts/glutton/update/Cargo.toml Outdated Show resolved Hide resolved

NachoPal and others added 6 commits June 5, 2023 12:32

docs improve

f6f7956

moved to main workspace

6e52e5f

fix conflicts

eee7040

update cargo

b8c7e46

fix conflicts

c617c7a

gilescope approved these changes Jun 7, 2023

View reviewed changes

jsdw reviewed Jun 7, 2023

View reviewed changes

scripts/glutton/update/Cargo.toml Show resolved Hide resolved

gilescope reviewed Jun 7, 2023

View reviewed changes

jsdw reviewed Jun 7, 2023

View reviewed changes

ggwpez reviewed Jun 7, 2023

View reviewed changes

NachoPal closed this Jun 8, 2023

NachoPal deleted the nacho/glutton-update-script branch June 8, 2023 11:01

		@@ -0,0 +1,4403 @@
		# This file is automatically @generated by Cargo.

Glutton update script #2681

Glutton update script #2681

Uh oh!

Conversation

NachoPal commented Jun 2, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sandreim left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

rphmeier commented Jun 2, 2023

Uh oh!

bkchr commented Jun 4, 2023

Uh oh!

NachoPal commented Jun 5, 2023

Uh oh!

NachoPal commented Jun 5, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

gilescope left a comment

Choose a reason for hiding this comment

Uh oh!

gilescope commented Jun 7, 2023

Uh oh!

Uh oh!

gilescope Jun 7, 2023

Choose a reason for hiding this comment

Uh oh!

jsdw Jun 7, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jsdw Jun 7, 2023

Choose a reason for hiding this comment

Uh oh!

jsdw commented Jun 7, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ggwpez Jun 7, 2023

Choose a reason for hiding this comment

Uh oh!

NachoPal Jun 7, 2023

Choose a reason for hiding this comment

Uh oh!

ggwpez Jun 7, 2023

Choose a reason for hiding this comment

Uh oh!

NachoPal Jun 7, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

NachoPal Jun 7, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bkchr Jun 7, 2023

Choose a reason for hiding this comment

Uh oh!

jsdw commented Jun 7, 2023

Uh oh!

NachoPal commented Jun 8, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

NachoPal commented Jun 2, 2023 •

edited

Loading

NachoPal commented Jun 5, 2023 •

edited

Loading

jsdw Jun 7, 2023 •

edited

Loading

jsdw commented Jun 7, 2023 •

edited

Loading

NachoPal Jun 7, 2023 •

edited

Loading

NachoPal Jun 7, 2023 •

edited

Loading

NachoPal commented Jun 8, 2023 •

edited

Loading