Beat [2/4]: implement `blockbeat` #8894

yyforyongyu · 2024-07-04T12:57:27Z

Depends on #8893.

NOTE: itest is fixed in the final PR #9260

This PR introduces a minimal version of the new service Blockbeat as described in #7952, to handle block synchronization among different subsystems.

During startup, blockbeat consumers are registered in the BlockbeatDispatcher, a subservice that's responsible for dispatching new blockbeats to its consumers and waiting for its consumers to finish processing the blocks. If any of the consumers fail to process the block under 30s, or encounter an error during block processing, the system will shut down as it's critical to handle blocks.

This PR focuses on implementing blockbeat Consumer interface for ChainArb, UtxoSweeper and TxPublisher, the following PR focuses on finalizing blockbeat processing in ChainArb's subservices - ChannelArbitrator, chainWatcher, and ContractResolver.

Overview

The flow of the blockbeat process is shown below, whenever a new block arrives, it goes through the waterfall like this,

blockbeat dispatcher receives a new block epoch and makes a blockbeat, and sends it to its consumers sequentially.
ChainArb receives the blockbeat, processes it and signals back when done.
UtxoSweeper receives the blockbeat, processes it and signals back when done.
TxPublisher receives the blockbeat, processes it and signals back when done.
This new block is now considered processed by the blockbeat dispatcher.

sequenceDiagram
		autonumber
		participant bb as BlockBeat
		participant cc as ChainArb
		participant us as UtxoSweeper
		participant tp as TxPublisher
		
		note left of bb: 0. received block x,<br>dispatching...
		
    note over bb,cc: 1. send block x to ChainArb,<br>wait for its done signal
		bb->>cc: block x
		rect rgba(165, 0, 85, 0.8)
      critical signal processed
        cc->>bb: processed block
      option Process error or timeout
        bb->>bb: error and exit
      end
    end

    note over bb,us: 2. send block x to UtxoSweeper, wait for its done signal
		bb->>us: block x
		rect rgba(165, 0, 85, 0.8)
      critical signal processed
        us->>bb: processed block
      option Process error or timeout
        bb->>bb: error and exit
      end
    end

    note over bb,tp: 3. send block x to TxPublisher, wait for its done signal
		bb->>tp: block x
		rect rgba(165, 0, 85, 0.8)
      critical signal processed
        tp->>bb: processed block
      option Process error or timeout
        bb->>bb: error and exit
      end
    end

TODO

add unit tests for new code
add readme

coderabbitai · 2024-07-04T12:58:28Z

Important

Review skipped

Auto reviews are limited to specific labels.

🏷️ Labels to auto review (1)

llm-review

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media?

❤️ Share

🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

Review comments: Directly reply to a review comment made by CodeRabbit. Example:
- I pushed a fix in commit <commit_id>, please review it.
- Generate unit testing code for this file.
- Open a follow-up GitHub issue for this discussion.
Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
- @coderabbitai generate unit testing code for this file.
- @coderabbitai modularize this function.
PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
- @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
- @coderabbitai read src/utils.ts and generate unit testing code.
- @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
- @coderabbitai help me debug CodeRabbit configuration file.

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (Invoked using PR comments)

@coderabbitai pause to pause the reviews on a PR.
@coderabbitai resume to resume the paused reviews.
@coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
@coderabbitai full review to do a full review from scratch and review all the files again.
@coderabbitai summary to regenerate the summary of the PR.
@coderabbitai resolve resolve all the CodeRabbit review comments.
@coderabbitai configuration to show the current CodeRabbit configuration for the repository.
@coderabbitai help to get help.

Other keywords and placeholders

Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
Add @coderabbitai anywhere in the PR title to generate the title automatically.

CodeRabbit Configuration File (`.coderabbit.yaml`)

You can programmatically configure CodeRabbit by adding a .coderabbit.yaml file to the root of your repository.
Please see the configuration documentation for more information.
If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

Documentation and Community

Visit our Documentation for detailed information on how to use CodeRabbit.
Join our Discord Community to get help, request features, and share feedback.
Follow us on X/Twitter for updates and announcements.

guggero

Did a first pass to load context.
The sweeper commits towards the end seem to make sense, but I'm missing all the review context from the sweeper PR saga to make a good judgement call.

But I really like the concept of the block beat! This should help us quite a bit in synchronizing all subsystems.

I assume a follow-up PR that refactors all subsystems to use this will come once this lands?

chainio/README.md

chainio/log.go

chainio/blockbeat.go

chainio/dispatcher.go

chainio/interface.go

contractcourt/channel_arbitrator.go

log.go

ellemouton

Really liking this! Left some questions so long (pretty much all just questions for my understanding). Gonna need another round to really grok things

chainio/interface.go

chainio/dispatcher.go

chainio/interface.go

chainio/blockbeat.go

chainio/interface.go

morehouse

A couple high-level concerns:

The timeout might be too short, causing complete node shutdown under transient load when previously recovery would have been possible.
TxPublisher should receive beats before UtxoSweeper, as discussed previously.

chainio/blockbeat.go

chainio/interface.go

chainio/blockbeat.go

morehouse · 2024-07-26T17:46:10Z

contractcourt/chain_arbitrator.go

+	c.Unlock()
+
+	// Iterate all the copied channels and send the blockbeat to them.
+	err := beat.DispatchConcurrent(channels)


I worry about timeouts in extreme cases here. handleBlockbeat has 30 seconds to complete, but every single channel arbitrator needs to process the block in that time. If there's thousands of channels, how long does it actually take to dispatch?

if there's no force close it takes microseconds otherwise it's milliseconds.

lightninglabs-deploy · 2024-10-28T02:25:19Z

@yyforyongyu, remember to re-request review from reviewers when ready

ziggie1984

I think it very close, we still need to decide on the following:

Will we add ctx where possible and also have the benefit of structured logging where possible
~~Maybe fix the lingering processBlock goroutine maybe via a errGroup with a timeout ctx ?~~=> was not taking the buffered errorChan into account

chainio/README.md

ziggie1984 · 2024-11-20T18:53:19Z

chainio/interface.go

+	// NOTE: The consumer must try its best to NOT return an error. If an
+	// error is returned from processing the block, it means the subsystem
+	// cannot react to onchain state changes and lnd will shutdown.
+	ProcessBlock(b Blockbeat) error


So the change in design however will keep piling up the goroutines now when we timeout for whatever reason. So if the consumer does not report in 60 sec one time this goroutine will linger around forever until the consumer closes.

ziggie1984 · 2024-11-20T19:38:58Z

chainio/blockbeat.go

+	}
+
+	// Create a customized logger for the blockbeat.
+	logPrefix := fmt.Sprintf("Height[%6d]:", b.Height())


I think we should go with structured logging from the beginning, this might also faciliated the structure and we do not need a logger object in the blockbeat.

ziggie1984 · 2024-11-20T19:43:00Z

chainio/dispatcher.go

+
+	errChan := make(chan error, 1)
+	go func() {
+		errChan <- c.ProcessBlock(b)


An argument for the ctx is also structured logging.

ziggie1984 · 2024-11-20T19:43:54Z

chainio/dispatcher.go

+		return fmt.Errorf("%s got err in ProcessBlock: %w", c.Name(),
+			err)
+
+	case <-time.After(timeout):


I don't think that is a good idea to call the timeout here rather than cancelling a context, because then to spawned goroutine might linger around if it does not return in time.

Ok I misunderstood, the errorChan is buffered so there is no way the spawned ProcessBlock worker get's stuck in case we prematurely exit because of a timeout.

yeah - ideally we want everything to be attached to the overall LND parent context in some way so that a cancel cancel things like this.

but i think implementing this in a contained way is easier said than done. I think we may need a big LND overhall that makes sure that the same parent context is shared by all subprocesses before we can do this properly

ziggie1984 · 2024-11-20T19:48:17Z

chainio/dispatcher.go

+// Otherwise, when living in the same queue, the subsystems are notified of the
+// new blocks sequentially, which means it's critical to understand the
+// relationship of these systems to properly handle the order.
+type BlockbeatDispatcher struct {


you mean that the NotifyBlockProcessed should be a callback when calling processblock ? I think that makes sense.

chainio/dispatcher.go

sweep/sweeper.go

ziggie1984

LGTM 🎊

As discussed via other means, the structured logging will not be part of this series and needs a greater change for the whole LND design.

ellemouton

Really great work & fantastic commit structure 🥇

chainio/README.md

ellemouton · 2024-11-21T08:49:07Z

chainio/interface.go

+	// NOTE: The consumer must try its best to NOT return an error. If an
+	// error is returned from processing the block, it means the subsystem
+	// cannot react to onchain state changes and lnd will shutdown.
+	ProcessBlock(b Blockbeat) error


Are you suggesting we should have the ability to cancel the state change made from calling the methods of a subsystem?

we can have this discussion offline too cause i dont think we need to do this in this pr.
But yes: i think any API interface method should take a context so that if the caller is quiting, then the method they are calling into doesnt just hang while it is busy.

ellemouton · 2024-11-21T08:50:16Z

chainio/blockbeat.go

+	}
+
+	// Create a customized logger for the blockbeat.
+	logPrefix := fmt.Sprintf("Height[%6d]:", b.Height())


it's tricky cause we need to do proper context usage first which is hard to contain to just a single system...

ellemouton · 2024-11-21T08:51:29Z

chainio/blockbeat.go

+	}
+
+	// Create a customized logger for the blockbeat.
+	logPrefix := fmt.Sprintf("Height[%6d]:", b.Height())


does it mean we don't need the prefix logger anymore

yeah we can eventually remove the prefix logger. I think we just need to get used to the k=v pairs at the end & of the log line that we can filter on instead of the prefix. Prefix i think should just be the subsystem.

chainio/dispatcher.go

ellemouton · 2024-11-21T08:56:53Z

chainio/dispatcher.go

+		return fmt.Errorf("%s got err in ProcessBlock: %w", c.Name(),
+			err)
+
+	case <-time.After(timeout):


yeah - ideally we want everything to be attached to the overall LND parent context in some way so that a cancel cancel things like this.

but i think implementing this in a contained way is easier said than done. I think we may need a big LND overhall that makes sure that the same parent context is shared by all subprocesses before we can do this properly

ellemouton · 2024-11-21T08:59:45Z

chainio/dispatcher.go

+// Otherwise, when living in the same queue, the subsystems are notified of the
+// new blocks sequentially, which means it's critical to understand the
+// relationship of these systems to properly handle the order.
+type BlockbeatDispatcher struct {


yeah, could be a Processed method on BlockBeat

This commit inits the package `chainio` and defines the interface `Blockbeat` and `Consumer`. The `Consumer` must be implemented by other subsystems if it requires block epoch subscription.

In this commit, a minimal implementation of `Blockbeat` is added to synchronize block heights, which will be used in `ChainArb`, `Sweeper`, and `TxPublisher` so blocks are processed sequentially among them.

This commit adds two methods to handle dispatching beats. These are exported methods so other systems can send beats to their managed subinstances.

This commit adds a blockbeat dispatcher which handles sending new blocks to all subscribed consumers.

This commit implements `Consumer` on `TxPublisher`, `UtxoSweeper`, `ChainArbitrator` and `ChannelArbitrator`.

This commit removes the independent block subscriptions in `UtxoSweeper` and `TxPublisher`. These subsystems now listen to the `BlockbeatChan` for new blocks.

This commit removes the hack introduced in #4851. Previously we had this issue because the chain notifier was stopped before the sweeper, which was changed a while back and we now always stop the chain notifier last. In addition, since we no longer subscribe to the block epoch chan directly, this issue can no longer happen.

The sweeper can handle the waiting so there's no need to wait for blocks inside the resolvers. By offering the inputs prior to their mature heights also guarantees the inputs with the same deadline are aggregated.

This commit removes the block subscriptions used in `ChainArbitrator` and replaced them with the blockbeat managed by `BlockbeatDispatcher`.

This commit removes the block subscriptions used in `ChannelArbitrator`, replaced them with the blockbeat managed by `BlockbeatDispatcher`.

This `immediate` flag was added as a hack so during a restart, the pending resolvers would offer the inputs to the sweeper and ask it to sweep them immediately. This is no longer need due to `blockbeat`, as now during restart, a block is always sent to all subsystems via the flow `ChainArb` -> `ChannelArb` -> resolvers -> sweeper. Thus, when there are pending inputs offered, they will be processed by the sweeper immediately.

To avoid calling GetBestBlock again.

This is needed so the consumers have an initial state about the current block.

In this commit we start to break up the starting process into smaller pieces, which is needed in the following commit to initialize blockbeat consumers.

Refactor the `Start` method to fix the linter error: ``` contractcourt/chain_arbitrator.go:568: Function 'Start' is too long (242 > 200) (funlen) ```

yyforyongyu added the utxo sweeping label Jul 4, 2024

yyforyongyu added this to the v0.18.2 milestone Jul 4, 2024

yyforyongyu self-assigned this Jul 4, 2024

guggero self-requested a review July 9, 2024 14:49

yyforyongyu force-pushed the yy-prepare-blockbeat branch from a0393a0 to 9d6f8e7 Compare July 11, 2024 09:46

saubyk requested a review from ellemouton July 11, 2024 15:18

guggero reviewed Jul 15, 2024

View reviewed changes

saubyk added the P1 MUST be fixed or reviewed label Jul 15, 2024

yyforyongyu force-pushed the yy-prepare-blockbeat branch 2 times, most recently from 16e2353 to 7183080 Compare July 15, 2024 18:02

ellemouton reviewed Jul 17, 2024

View reviewed changes

yyforyongyu force-pushed the yy-prepare-blockbeat branch 2 times, most recently from 755c7f3 to 9ec032a Compare July 18, 2024 07:32

yyforyongyu force-pushed the yy-blockbeat branch from 1b9497c to 842d88f Compare July 18, 2024 07:33

yyforyongyu changed the title ~~Beat [2/3]: implement blockbeat~~ Beat [2/4]: implement blockbeat Jul 18, 2024

saubyk modified the milestones: v0.18.3, v0.19.0 Jul 18, 2024

morehouse reviewed Jul 26, 2024

View reviewed changes

yyforyongyu force-pushed the yy-blockbeat branch 2 times, most recently from 7b05026 to d76a26e Compare October 15, 2024 09:18

yyforyongyu force-pushed the yy-prepare-blockbeat branch from 9ec032a to f6c5de6 Compare October 15, 2024 09:18

yyforyongyu force-pushed the yy-prepare-blockbeat branch from f6c5de6 to e4c6bd4 Compare October 28, 2024 10:30

yyforyongyu force-pushed the yy-blockbeat branch 2 times, most recently from 4e00941 to 7b1b444 Compare October 29, 2024 14:03

yyforyongyu force-pushed the yy-prepare-blockbeat branch from e4c6bd4 to d7d2036 Compare October 29, 2024 14:16

yyforyongyu force-pushed the yy-blockbeat branch from 7b1b444 to d900f98 Compare October 29, 2024 14:17

yyforyongyu mentioned this pull request Oct 29, 2024

Beat [3/4]: prepare resolvers to handle the blockbeat #8922

Merged

yyforyongyu force-pushed the yy-blockbeat branch from d900f98 to 0ad97ec Compare October 29, 2024 18:28

yyforyongyu force-pushed the yy-blockbeat branch from e40d7aa to 736edda Compare November 19, 2024 23:27

yyforyongyu force-pushed the yy-feature-blockbeat branch from dd1d406 to b6c7165 Compare November 20, 2024 01:25

ziggie1984 reviewed Nov 20, 2024

View reviewed changes

ziggie1984 approved these changes Nov 21, 2024

View reviewed changes

ellemouton approved these changes Nov 21, 2024

View reviewed changes

yyforyongyu force-pushed the yy-blockbeat branch from 736edda to 1e17c5c Compare November 21, 2024 12:47

yyforyongyu force-pushed the yy-feature-blockbeat branch from b6c7165 to 616399e Compare November 21, 2024 12:48

yyforyongyu and others added 19 commits November 21, 2024 21:00

chainio: introduce chainio to handle block synchronization

0f8a524

This commit inits the package `chainio` and defines the interface `Blockbeat` and `Consumer`. The `Consumer` must be implemented by other subsystems if it requires block epoch subscription.

chainio: implement Blockbeat

e263ece

In this commit, a minimal implementation of `Blockbeat` is added to synchronize block heights, which will be used in `ChainArb`, `Sweeper`, and `TxPublisher` so blocks are processed sequentially among them.

chainio: add helper methods to dispatch beats

373b795

This commit adds two methods to handle dispatching beats. These are exported methods so other systems can send beats to their managed subinstances.

chainio: add BlockbeatDispatcher to dispatch blockbeats

797c10e

This commit adds a blockbeat dispatcher which handles sending new blocks to all subscribed consumers.

chainio: add partial implementation of Consumer interface

5d91b5e

multi: implement Consumer on subsystems

269e37c

This commit implements `Consumer` on `TxPublisher`, `UtxoSweeper`, `ChainArbitrator` and `ChannelArbitrator`.

sweep: remove block subscription in UtxoSweeper and TxPublisher

09ad818

This commit removes the independent block subscriptions in `UtxoSweeper` and `TxPublisher`. These subsystems now listen to the `BlockbeatChan` for new blocks.

contractcourt: remove waitForHeight in resolvers

eed75c7

The sweeper can handle the waiting so there's no need to wait for blocks inside the resolvers. By offering the inputs prior to their mature heights also guarantees the inputs with the same deadline are aggregated.

contractcourt: remove block subscription in chain arbitrator

627065c

This commit removes the block subscriptions used in `ChainArbitrator` and replaced them with the blockbeat managed by `BlockbeatDispatcher`.

contractcourt: remove block subscription in channel arbitrator

bb494da

This commit removes the block subscriptions used in `ChannelArbitrator`, replaced them with the blockbeat managed by `BlockbeatDispatcher`.

contractcourt: start channel arbitrator with blockbeat

6daea9c

To avoid calling GetBestBlock again.

multi: start consumers with a starting blockbeat

48dd2ea

This is needed so the consumers have an initial state about the current block.

lnd: add new method startLowLevelServices

197f1e8

In this commit we start to break up the starting process into smaller pieces, which is needed in the following commit to initialize blockbeat consumers.

lnd: start blockbeatDispatcher and register consumers

07872fc

contractcourt: fix linter funlen

8bd6153

Refactor the `Start` method to fix the linter error: ``` contractcourt/chain_arbitrator.go:568: Function 'Start' is too long (242 > 200) (funlen) ```

multi: improve loggings

b63855d

chainio: use errgroup to limit num of goroutines

9ab2e53

yyforyongyu force-pushed the yy-blockbeat branch from 1e17c5c to 9ab2e53 Compare November 21, 2024 13:00

yyforyongyu merged commit 7295ae3 into yy-feature-blockbeat Nov 21, 2024
15 of 18 checks passed

yyforyongyu deleted the yy-blockbeat branch November 21, 2024 14:41

yyforyongyu mentioned this pull request Nov 27, 2024

Implement blockbeat #9315

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Beat [2/4]: implement `blockbeat` #8894

Beat [2/4]: implement `blockbeat` #8894

yyforyongyu commented Jul 4, 2024 •

edited

Loading

coderabbitai bot commented Jul 4, 2024 •

edited

Loading

Review skipped

Chat

CodeRabbit Commands (Invoked using PR comments)

Other keywords and placeholders

CodeRabbit Configuration File (`.coderabbit.yaml`)

Documentation and Community

guggero left a comment

ellemouton left a comment

morehouse left a comment •

edited

Loading

morehouse Jul 26, 2024

yyforyongyu Nov 12, 2024

lightninglabs-deploy commented Oct 28, 2024

ziggie1984 left a comment •

edited

Loading

ziggie1984 Nov 20, 2024

ziggie1984 Nov 20, 2024

ziggie1984 Nov 20, 2024

ziggie1984 Nov 20, 2024

ziggie1984 Nov 21, 2024

ellemouton Nov 21, 2024

ziggie1984 Nov 20, 2024

ziggie1984 left a comment

ellemouton left a comment

ellemouton Nov 21, 2024

ellemouton Nov 21, 2024

ellemouton Nov 21, 2024

ellemouton Nov 21, 2024

ellemouton Nov 21, 2024

Beat [2/4]: implement blockbeat #8894

Beat [2/4]: implement blockbeat #8894

Conversation

yyforyongyu commented Jul 4, 2024 • edited Loading

Overview

coderabbitai bot commented Jul 4, 2024 • edited Loading

Review skipped

Chat

CodeRabbit Commands (Invoked using PR comments)

Other keywords and placeholders

CodeRabbit Configuration File (.coderabbit.yaml)

Documentation and Community

guggero left a comment

Choose a reason for hiding this comment

ellemouton left a comment

Choose a reason for hiding this comment

morehouse left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lightninglabs-deploy commented Oct 28, 2024

ziggie1984 left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ziggie1984 left a comment

Choose a reason for hiding this comment

ellemouton left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Beat [2/4]: implement `blockbeat` #8894

Beat [2/4]: implement `blockbeat` #8894

yyforyongyu commented Jul 4, 2024 •

edited

Loading

coderabbitai bot commented Jul 4, 2024 •

edited

Loading

CodeRabbit Configuration File (`.coderabbit.yaml`)

morehouse left a comment •

edited

Loading

ziggie1984 left a comment •

edited

Loading