Adds extra logging to validate hypothesis that we've found a bug in
GetConfirmedBlocks, rather than a general issue.
Change-Id: Ia3acfd7fa393091a3892c453433255cfbe38e4a0
We encountered a bug on mainnet where GetConfirmedBlocks with
commitment level Finalized would fail to return a valid block.
Remove the optimization and simply fetch every slot.
Change-Id: Idce939c6c5f3303a5046efdfd31833fab3d465ba
Optional metadata for debugging. Allows tying a received observation
to an on-chain transaction even if the node missed the observation.
Change-Id: Ie50c10c8ab7f65469555674d90645f91092fb2e3
In cases where we observed a VAA, there is no possibility of gossip DoS.
Increase the timeout to 24 hours to facilitate manual interventions
(like submission of governance VAAs or node restarts/catchup).
Keep the existing five minute timeout for observation-less VAAs.
Change-Id: Ic626108190bd60cf812daadbe191b31cc48c7296
Chances are that more requests will fail until the scheduled
recovery date, so we can't hardcode them.
Change-Id: Ief45c1f7a455827b32e154fae5c8eed366dd3ff7
Surfaced by the retry code:
https://gist.github.com/leoluk/b5d05ed27269b077b834eda771a50058
Accessing s.logger from a goroutine will cause a data race. We didn't
previously encounter this since supervisor would wait for Run() to
return before rescheduling it.
Change-Id: I56a7503081485e58975103d0e25e0c2baf19ca08
rustfmt appears to be a little more complicated since it wants to
download dependencies and needs nightly Rust.
Change-Id: Ia348def30a6459ae2ab6c29a8c3a413216f5eb4b
Ensure there's no core vs. token bridge ambiguity.
Breaking changes to the CLI:
* "guardiand bridge" CLI is now "guardiand node"
* --solanaBridgeAddress is now --solanaContract
* --bridgeKey is now --guardianKey
The Heartbeat proto message had one of its fields renamed from
BridgeAddress to ContractAddress, but this won't break the wire
format and the only consumer appears to be the CLI.
Change includes a "go mod tidy" - it insisted.
Change-Id: Id8b312827737f07f2d5f3944ebce469d946e7f51