Handle MOVED error pointing to same endpoint. #3003
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…on before retrying the request.
barshaul
force-pushed
the
moved_same_endpoint
branch
from
948a601 to
9c32b44
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
Handle MOVED error pointing to same endpoint
Problem
When Redis/Valkey servers are deployed behind DNS records, load balancers, or proxies (as is common in managed environments), a MOVED error may be returned with a target node endpoint that is the same endpoint from which the error originated. Before the proposed change, when StackExchange.Redis received a MOVED error pointing to the same endpoint, it would fail the request and propagate the MOVED exception back to the application.
Issue ref: #2990
Solution
When a MOVED error points to the same endpoint, the client now triggers a reconnection before retrying the command. This allows the DNS record, proxy, or load balancer to route the new connection to a different underlying server host, enabling the retry to succeed.
Why proactive reconnection is necessary
The server closes the connection on its end immediately after sending the MOVED-to-same-endpoint error. However, due to the multiplexed nature of StackExchange.Redis connections, the library wouldn't notice the disconnection when it retries the request—all Write and Flush operations return successfully even though the underlying connection is broken. The disconnection is only detected when attempting to read the response.
At that point, it isn't safe to retry on the connection failure: we don't know if the request was actually sent to the server, so the connection error is raised back to the application, which also cannot determine if it's safe to retry. By initiating a reconnect immediately following the MOVED-to-same-endpoint error, we ensure the retry occurs on a fresh connection to a new server host.
Changes
Core fix:
ResultProcessor.cs: Detect when MOVED endpoint matches the current server endpoint and trigger reconnection by disposing the existing connection.Tests:
MovedToSameEndpointTests.cs: Integration test verifying the reconnect-and-retry behaviorMovedTestServer.cs: Test server helper that simulates MOVED responses pointing to the same endpoint