-
Notifications
You must be signed in to change notification settings - Fork 2.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Start Channel Arbitrators concurrently #9262
base: master
Are you sure you want to change the base?
Start Channel Arbitrators concurrently #9262
Conversation
Important Review skippedAuto reviews are limited to specific labels. 🏷️ Labels to auto review (1)
Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the 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? 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
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)
Other keywords and placeholders
CodeRabbit Configuration File (
|
98664a1
to
d68ce0c
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice! Thanks a lot for the fix 🎉
I think we might want to make it configurable, just in case.
Here's my proposal to add a config value (hidden, so we don't have to add it to the sample-lnd.conf
file to avoid people messing with it when not necessary):
arb.diff.txt
contractcourt/chain_arbitrator.go
Outdated
@@ -27,6 +28,12 @@ import ( | |||
// ErrChainArbExiting signals that the chain arbitrator is shutting down. | |||
var ErrChainArbExiting = errors.New("ChainArbitrator exiting") | |||
|
|||
const ( | |||
// chainArbTimeout is the timeout for the chain arbitrator to start | |||
// the channel arbitrators for each channel. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We could add some rationale here for why this number is rather high. Something along the lines of: "Because starting arbitrators might depend on resolutions that are done by an external component (e.g. aux component hooks), this might require other lnd subsystems to be fully started. So we need to give everything enough time."
for _, arbitrator := range c.activeChannels { | ||
startState, ok := startStates[arbitrator.cfg.ChanPoint] | ||
if !ok { | ||
stopAndLog() | ||
|
||
// In case we encounter an error we need to cancel the | ||
// context to ensure all goroutines are cleaned up. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: we don't actually clean up the goroutines yet (or I mean we don't interrupt the arb.Start()
yet, we just don't wait for it anymore). So can perhaps add a TODO here that we'll want to eventually add a context to Start()
so things will clean up properly on cancel?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Now we clean them up, I think adding a context ist not really the way to go because looking into the startup functionality it does not look like a context might help here.
Ok I think this approach is not really good because you were raising a really good point here @guggero:
This call does block forever when we launch the |
By easy way, I don't think it is appropriate design, cancels imo make only sense if we know we spawn goroutines in that function but our problem arises because a function call to a normal subsystem does not timeout, so I think we need to improve there. |
what we could do which is the only way it could work is something like this:
basically starting a goroutine in a goroutine. Only then can we really abort after the timeout. |
Ok I mean spawning goroutines might not be the best solution here because we need for each startup 2, but I implemented it nonetheless because maybe it will take a while until we have the hooks for external subsystems configurable with timeouts. |
1579cc6
to
af4df94
Compare
@Roasbeef this works as well and resolves the deadlock. |
When starting the channel arbitrators we make sure they are started concurrently.
The channel arbitrator startups have now a configurable timeout but this config value is hidden.
02ef00d
to
b0ee865
Compare
Followup of #9253. This makes sure we start the channel arbitrators concurrently with a maximum timeout of 5 min.