Uh oh!
There was an error while loading. Please reload this page.
Avoid downloading multiple chains of headers on sync - #2017
Closed
UdjinM6 wants to merge 2 commits into
Closed
Conversation
Reverts "Fix duplicate headers download in initial sync (dashpay#1589)" and all following fixes This reverts commit 169afaf.
codablock
commented
Mar 27, 2018
Wouldn't this result in the same issue as in bitcoin#8054? If a fork happens at exactly lastHeaderHeight+1, you'll miss that fork. If that fork turns out to be the longest chain, you wouldn't be able to catch up. |
UdjinM6
commented
Mar 27, 2018
Author
Good point. Also my patch seems to be needlessly complex. Closing. |
This file contains hidden or 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
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The main idea is to never send
GETHEADERSifHEADERSmessage is at its max size (i.e. the peer we got this from should still have more blocks to announce which means that it's most likely not some recent fork or smth like that but actual historical blocks) and all headers inHEADERSmessage we received are already known. Tested this on testnet and devnet, seems to be working fine and fixing the initial issue completely. Tests also pass with no issues on my local machine and Travis seems to be pretty happy too https://travis-ci.org/UdjinM6/dash/builds/358612242.Note: This is alternative to #1589 (d7a521d reverts the original PR and all subsequent related changes in one go).