rabbit_feature_flags: Protect concurrent reloads of the registry #10027
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Outdated suggestions cannot be applied.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.
Why
When a node joins another node, it resets its feature flags registry (the registry is unloaded) and it copies the feature flags states from the remote cluster.
Before this patch, there was a small window where a concurrent use of the registry right between these two steps would reload a registry from the default/empty states, which does not reflect the intent.
This could happen because another node is running peer discovery and queries the cluster membership of the node joining a cluster.
How
We acquire the registry reload lock around the reset+copy and in
rabbit_ff_registry_wrapper
.