core/bcast: add recaster to rebroadcast registrations every epoch #1008
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
@@ Coverage Diff @@
## main #1008 +/- ##
==========================================
- Coverage 54.25% 53.54% -0.72%
==========================================
Files 117 119 +2
Lines 13130 13292 +162
==========================================
- Hits 7124 7117 -7
- Misses 4983 5146 +163
- Partials 1023 1029 +6
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. |
|
run in ropsten cluster appear to be registering each epoch as intended beacon is keeping validators in its registered pool only strange behaviour was to do with /teku-proposer-config endpoint also get a lot of these in the logs
length should be 100 - drop from 96 to 81 is end of an epoch so some validators will have had to have failed their re-broadcast in 3 consecutive epochs |
|
That teku error seems due to empty "fee_recipient" being returned by teku-proposer-config endpoint. Probably due to it being empty in the cluster lock. wrt "failed to submit validator registration", see the beacon API latency metrics. It seems like the beacon node is timing out after 2s. Strange that the register endpoint is so slow? |
xenowits
reviewed
Aug 22, 2022
| defer r.mu.Unlock() | ||
|
|
||
| tuple, ok := r.tuples[pubkey] | ||
| if ok && tuple.duty.Slot >= duty.Slot { |
xenowits
reviewed
Aug 22, 2022
core/bcast/recast.go
Outdated
| r.mu.Lock() | ||
| for k, v := range r.tuples { | ||
| clonedTuples[k] = v | ||
| clonedSubs = append(clonedSubs, r.subs...) |
There was a problem hiding this comment.
couldn't understand why cloning subs inside the loop
There was a problem hiding this comment.
@ciaranmcveigh5 this might have caused some of the latency issue you saw, since we bombarded BN with duplicate rebroadcasts
xenowits
reviewed
Aug 22, 2022
| s.scheduleSlot(slotCtx, slot) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| func (s *Scheduler) emitCoreSlot(ctx context.Context, slot core.Slot) { |
xenowits
reviewed
Aug 22, 2022
|
|
||
| // FirstInEpoch returns true if this is the first slot in the epoch. | ||
| func (s Slot) FirstInEpoch() bool { | ||
| return s.Slot%s.SlotsPerEpoch == 0 |
There was a problem hiding this comment.
do we assume that s.Slot is always a positive integer? coz this won't work for -1
There was a problem hiding this comment.
-1 would return false and that is correct no?
There was a problem hiding this comment.
yeah, it is correct. the testcase i had in my mind was epoch -1, ie, slots from -32...-1.
There was a problem hiding this comment.
not sure that is a valid test case
xenowits
approved these changes
Aug 22, 2022
corverroos
added
the
merge when ready
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds the recaster component to rebroadcast builder registrations every epoch.
category: feature
ticket: #1009