Check chain hash for channel announcement and update

[freddiekrugerrand]

Fixes #2228

[dunxen]

LGTM with one small nit.

[dunxen]

nit: Could we just include why it's the wrong chain hash here? I got a little confused with it in isolation but remembered we were dealing with testnet in these gossip tests.

"testnet vs mainnet" or something similar.

[codecov-commenter]

Codecov Report

Patch coverage: 77.77% and project coverage change: -0.01 ⚠️

Comparison is base (c182567) 91.56% compared to head (0a43123) 91.56%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2230      +/-   ##
==========================================
- Coverage   91.56%   91.56%   -0.01%     
==========================================
  Files         104      104              
  Lines       51749    51767      +18     
  Branches    51749    51767      +18     
==========================================
+ Hits        47384    47398      +14     
- Misses       4365     4369       +4     

Impacted Files	Coverage Δ
lightning/src/routing/gossip.rs	89.77% <77.77%> (-0.15%)	⬇️

... and 5 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[benthecarman]

This could also be done for rapid gossip sync updates

[tnull]

Thanks for taking this on!

[tnull]

This could also be done for rapid gossip sync updates

Catching unmatching RGS updates are the primary intention, but I think they are already covered by checking in update_channel_intern?

[freddiekrugerrand]

This could also be done for rapid gossip sync updates

Took a look at processing.rs to make sure RGS is fully covered:

• update_channel_unsigned: calls update_channel_intern which has check
• add_channel_from_partial_announcement: no chain_hash provided

Wondering whether we should add a top level check here in RGS so that we can exit early if we have the wrong hash? Would need to add get_genesis_hash (or check_genesis_hash) to NetworkGraph.

[TheBlueMatt]

Nice!

In a followup, we should also (finally) add support for the init genesis block hash field by fetching the genesis block hash from the ChannelMessageHandler (ie ChannelManager) in PeerManager, then checking that value against what we receive in init message(s) and send it in ours.

[TheBlueMatt]

Wondering whether we should add a top level check here in RGS so that we can exit early if we have the wrong hash? Would need to add get_genesis_hash (or check_genesis_hash) to NetworkGraph.

Yes, yes we should.