Skip to content

[9.4] Improve failure handling for shards (#156515) #156819

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.

Sign up for GitHub

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

Merged
elasticsearchmachine merged 3 commits into from
Sep 1, 2026
Merged

[9.4] Improve failure handling for shards (#156515) #156819

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d942f1c
Improve failure handling for shards (#156515)
smalyshev Aug 14, 2026
3c257e7
Merge branch '9.4' into backport/9.4/pr-156515
smalyshev Aug 17, 2026
36eb88c
Merge branch '9.4' into backport/9.4/pr-156515
smalyshev Sep 1, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 19 additions & 3 deletions server/src/main/java/org/elasticsearch/action/search/SearchQueryThenFetchAsyncAction.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -574,7 +574,12 @@ public void handleResponse(NodeQueryResponse response) {
onShardResult(q);
}
case null, default -> {
assert false : "impossible [" + response.results[i] + "]";
var e = new IllegalStateException("data node returned unexpected result for shard [" + s.shardId + "]");
logger.error(
"data node produced unexpected result[" + response.results[i] + "] for shard [" + s.shardId + "]",
e
);
onShardFailure(shardIdx, target, shardIterators[shardIdx], e);
}
}
}
Expand Down Expand Up @@ -950,7 +955,7 @@ private void writeSuccessfulResponse(RecyclerBytesStreamOutput out) throws IOExc
for (int i = 0; i < resultCount; i++) {
var result = queryPhaseResultConsumer.results.get(i);
if (result == null) {
NodeQueryResponse.writePerShardException(out, failures.remove(i));
NodeQueryResponse.writePerShardException(out, shardFailureOrUnknown(i));
} else {
// free context id and remove it from the result right away in case we don't need it anymore
maybeFreeContext(result, relevantShardIndices, namedWriteableRegistry);
Expand All @@ -968,7 +973,7 @@ private void writeReductionFailureResponse(RecyclerBytesStreamOutput out, Except
for (int i = 0; i < resultCount; i++) {
var result = queryPhaseResultConsumer.results.get(i);
if (result == null) {
NodeQueryResponse.writePerShardException(out, failures.remove(i));
NodeQueryResponse.writePerShardException(out, shardFailureOrUnknown(i));
} else {
NodeQueryResponse.writePerShardResult(out, result);
}
Expand All @@ -978,6 +983,17 @@ private void writeReductionFailureResponse(RecyclerBytesStreamOutput out, Except
releaseAllResultsContexts();
}

private Exception shardFailureOrUnknown(int localIndex) {
Exception failure = failures.remove(localIndex);
if (failure == null) {
logger.error("data node produced null failure for shard [{}]", localIndex);
failure = new IllegalStateException(
"shard [" + searchRequest.shards.get(localIndex).shardId + "] neither succeeded nor failed"
);
}
return failure;
}

/**
* This code is strictly for _snapshot_ backwards compatibility. The feature flag
* {@link SearchService#BATCHED_QUERY_PHASE_FEATURE_FLAG} was not turned on when the transport version
Expand Down
Loading