Differential D1199

Activate replay protection at the time the next HF is scheduled.
ClosedPublic
Actions

Authored by deadalnix on Mar 15 2018, 00:27.

Details

Reviewers

jasonbcox

Group Reviewers

Restricted Project

Maniphest Tasks

T301: Automatic Future Hardfork Replay Protection

Commits

rSTAGINGdb0e07afa96e: Activate replay protection at the time the next HF is scheduled.
rABCdb0e07afa96e: Activate replay protection at the time the next HF is scheduled.

Summary

This code will remain in place until the next HF is implemented, which ensure that there will be replay protection between the chains in case a minority survives by default. Obviously, the minority can chose to also do a hard fork and remove that replay protection, but then it is up to the minority chain promoter to convince actor in the space.

Test Plan

Added an integration test which grills the replay protection activation and deactivation (in case of reorg)

Diff Detail

Repository

rABC Bitcoin ABC

Lint

Lint Not Applicable

Unit

Tests Not Applicable

Event Timeline

deadalnix created this revision.Mar 15 2018, 00:27

Herald added a reviewer: Restricted Project. · View Herald TranscriptMar 15 2018, 00:27

Build Bitcoin-ABC :: Staging :: Bitcoin-ABC Staging started.

Build Bitcoin-ABC :: Staging :: Bitcoin-ABC Staging passed.

Harbormaster completed remote builds in B2076: Diff 3188.Mar 15 2018, 00:50

schancel added a task: T301: Automatic Future Hardfork Replay Protection.Mar 15 2018, 14:24

schancel edited the summary of this revision. (Show Details)

jasonbcox added a subscriber: jasonbcox.Mar 15 2018, 21:19

jasonbcox added inline comments.

test/functional/abc-p2p-fullblocktest.py
52 ↗	(On Diff #3188)	Just for my understanding, this is to make the value so far in the future that it won't be hit by the test, right?
test/functional/abc-replay-protection.py
148–149 ↗	(On Diff #3188)	I noticed a fair number of modified methods like this in other tests. Not an action item for this diff, but we should consider consolidating some of the common features from these methods into blocktools.py in the future.

The change looks good to me. However, I didn't have time to do a detailed review of the entire test. From a high-level the test looks good. I encourage another reviewer to look over the test.

This revision is now accepted and ready to land.Mar 15 2018, 21:22

deadalnix added inline comments.Mar 15 2018, 21:38

test/functional/abc-p2p-fullblocktest.py
52 ↗	(On Diff #3188)	We don't want the actual value to be reached, that'd screw up all the test assumptions.
test/functional/abc-replay-protection.py
148–149 ↗	(On Diff #3188)	Probably. That being said, code duplication is not as big of a deal in tests as it is in regular code. See https://evgenii.com/blog/code-duplication-in-unit-tests/ to understand why.