CAE-534 - RediSearch: Sticky Cursor in Cluster (Stage 1) #3409
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
uglide
requested changes
Aug 25, 2025
| @@ -1713,17 +1713,26 @@ public RedisFuture<AggregationReply<K, V>> ftAggregate(K index, V query) { | |||
| } | |||
|
|
|||
| @Override | |||
| public RedisFuture<AggregationReply<K, V>> ftCursorread(K index, long cursorId, int count) { | |||
| return dispatch(searchCommandBuilder.ftCursorread(index, cursorId, count)); | |||
| public RedisFuture<AggregationReply<K, V>> ftCursorread(K index, AggregationReply<K, V> aggregateReply, int count) { | |||
There was a problem hiding this comment.
One reason to iterate over the aggregated result with the cursor is that the reply is too big, contains too many results, etc. Using AggregationReply as an argument to ftCursorread defeats the original purpose. A much cleaner approach would be to introduce an abstraction for Cursor that encapsulates the cursorId and nodeId and pass it to the ftCursor commands.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
TL;DR: Make
FT.AGGREGATE … WITHCURSORsticky to the shard that created it by carrying the routing token (nodeId) in theAggregationReplyand routingFT.CURSOR READ/DELaccordingly. No writer changes.What changed
AggregationReply<K,V>@Nullable String nodeId+getNodeId(); package-privatesetNodeId(...).equals/hashCode.API (all
@Experimental)Cursor ops now consume the reply:
ftCursorread(AggregationReply<K,V> reply, int count)ftCursorread(AggregationReply<K,V> reply)ftCursordel (AggregationReply<K,V> reply)Removed/adapted old
(index, cursorId)variants.Cluster behavior
FT.AGGREGATE, ifcursorId > 0, stampreply.setNodeId(executedNode.getNodeId()).ftCursorread/ftCursordelresolvereply.nodeId → endpointviaPartitionsand dispatch to that node.Single node / missing
nodeIdMigration note
AggregationReplyinstead of(index, cursorId).What's next
Stage 2: read-write load-balancing for keyless commands, including
FT.SEARCH/FT.AGGREGATE.