-
Notifications
You must be signed in to change notification settings - Fork 305
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Chain halt during swap execution on post-upgraded devnet #4340
Comments
|
The bad value is:
|
That hex-encoded key is prefixed with As of 0.73.1, that key is used here: https://github.com/penumbra-zone/penumbra/blob/v0.73.1/crates/core/component/dex/src/state_key.rs#L114-L123 As of |
Steps to reproduce:
|
I am fairly sure that there are no issue with the migration, but that the bug was introduced from my PR #4188 where I replaced: // Write the new lookup index storing `new_a_from_b` for this trading pair.
self.nonverifiable_put_raw(
engine::routable_assets::a_from_b(&pair).to_vec(),
new_a_from_b.to_be_bytes().to_vec(),
); With a proto schema wrapper: let auxiliary_key = engine::routable_assets::lookup_base_liquidity_by_pair(&pair).to_vec();
self.nonverifiable_put(auxiliary_key, new_tally);
tracing::debug!(
?pair,
"base liquidity heuristic marked directed pair as routable"
); This was an oversight on my part, I double checked that the main index was proto encoded, but must have missed that the other one wasn't. |
@zbuc correct me if i'm wrong, but the reason why the state key |
@erwanor the reason the The bug I was fixing related to the |
Building on the smoke-test rewrite to use process-compose, let's script the migration process, so that we can test current HEAD of the monorepo against a prior tagged version, and validate that necessary migrations are in place. One possible approach is to fetch prebuilt binaries from uploaded artifacts on Github. That's fine for `pd`, but doesn't work for running the smoke tests, due to client/server incompatibility. Therefore we'll clone the entire repo in a git-ignored subdir, and build the old binaries there. Heavy, but reliable. Updated to use the concise `pd migrate` UX from #4339. Previously, there were missing AuctionParams, resolved by #4338. Still seeing some proto incompat post-migration, via the test runs, which appears to match the report in #4340.
Building on the smoke-test rewrite to use process-compose, let's script the migration process, so that we can test current HEAD of the monorepo against a prior tagged version, and validate that necessary migrations are in place. One possible approach is to fetch prebuilt binaries from uploaded artifacts on Github. That's fine for `pd`, but doesn't work for running the smoke tests, due to client/server incompatibility. Therefore we'll clone the entire repo in a git-ignored subdir, and build the old binaries there. Heavy, but reliable. Updated to use the concise `pd migrate` UX from #4339. Previously, there were missing AuctionParams, resolved by #4338. Still seeing some proto incompat post-migration, via the test runs, which appears to match the report in #4340.
Building on the smoke-test rewrite to use process-compose, let's script the migration process, so that we can test current HEAD of the monorepo against a prior tagged version, and validate that necessary migrations are in place. One possible approach is to fetch prebuilt binaries from uploaded artifacts on Github. That's fine for `pd`, but doesn't work for running the smoke tests, due to client/server incompatibility. Therefore we'll clone the entire repo in a git-ignored subdir, and build the old binaries there. Heavy, but reliable. Updated to use the concise `pd migrate` UX from #4339. Previously, there were missing AuctionParams, resolved by #4338. Still seeing some proto incompat post-migration, via the test runs, which appears to match the report in #4340.
Building on the smoke-test rewrite to use process-compose, let's script the migration process, so that we can test current HEAD of the monorepo against a prior tagged version, and validate that necessary migrations are in place. One possible approach is to fetch prebuilt binaries from uploaded artifacts on Github. That's fine for `pd`, but doesn't work for running the smoke tests, due to client/server incompatibility. Therefore we'll clone the entire repo in a git-ignored subdir, and build the old binaries there. Heavy, but reliable. Updated to use the concise `pd migrate` UX from #4339. Previously, there were missing AuctionParams, resolved by #4338. Still seeing some proto incompat post-migration, via the test runs, which appears to match the report in #4340.
Building on the smoke-test rewrite to use process-compose, let's script the migration process, so that we can test current HEAD of the monorepo against a prior tagged version, and validate that necessary migrations are in place. One possible approach is to fetch prebuilt binaries from uploaded artifacts on Github. That's fine for `pd`, but doesn't work for running the smoke tests, due to client/server incompatibility. Therefore we'll clone the entire repo in a git-ignored subdir, and build the old binaries there. Heavy, but reliable. Updated to use the concise `pd migrate` UX from #4339. Previously, there were missing AuctionParams, resolved by #4338. Still seeing some proto incompat post-migration, via the test runs, which appears to match the report in #4340.
Building on the smoke-test rewrite to use process-compose, let's script the migration process, so that we can test current HEAD of the monorepo against a prior tagged version, and validate that necessary migrations are in place. One possible approach is to fetch prebuilt binaries from uploaded artifacts on Github. That's fine for `pd`, but doesn't work for running the smoke tests, due to client/server incompatibility. Therefore we'll clone the entire repo in a git-ignored subdir, and build the old binaries there. Heavy, but reliable. Updated to use the concise `pd migrate` UX from #4339. Previously, there were missing AuctionParams, resolved by #4338. Still seeing some proto incompat post-migration, via the test runs, which appears to match the report in #4340.
Describe the bug
On a local devnet on which I performed a chain upgrade (this may or may not be relevant), I saw a chain halt while swapping:
The text was updated successfully, but these errors were encountered: