# Elasticsearch non blocking ElasticsearchAsyncSearchAsyncClient and ElasticsearchAsyncClient **URL:** \ **Category:** Elasticsearch\ **Created:** [April 11, 2022, 8:39am UTC](https://discuss.elastic.co/t/elasticsearch-non-blocking-elasticsearchasyncsearchasyncclient-and-elasticsearchasyncclient/302077 "2022-04-11T08:39:35Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![saeed\_shahsavan](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/saeed_shahsavan/32/100782_2.png) [@saeed\_shahsavan](https://discuss.elastic.co/u/saeed_shahsavan)\ **Post date:** [April 11, 2022, 8:39am UTC](https://discuss.elastic.co/t/elasticsearch-non-blocking-elasticsearchasyncsearchasyncclient-and-elasticsearchasyncclient/302077/1 "2022-04-11T08:39:35Z")
Recently I have upgraded my project to use Elasticsearch so I have changed my client to use ElasticsearchAsyncSearchAsyncClient and ElasticsearchAsyncClient unfortunately in both clients my response time increased hugely I have put log in my code to check the reason, I have seen that took time from Elasticsearch did not change but time of get response from asyncClient increased here is my code: note: I use vertx in my project ```auto public Future> search(String routing, int size, Instant startTime, Instant endTime) { Promise> responsePromise = Promise.promise(); final var termQueryBuilder = new TermQuery.Builder(); final var rangeQueryBuilder = new RangeQuery.Builder(); if (startTime != null) { rangeQueryBuilder.field(UrlMessage.NEXT_FETCH_FIELD_NAME).gt(JsonData.of(startTime.toString())); } if (endTime != null) { rangeQueryBuilder.field(UrlMessage.NEXT_FETCH_FIELD_NAME).lte(JsonData.of(endTime.toString())); } String finalIndex = indexSelector.getIndex(routing); boolean isMainIndex = finalIndex.equals(index); if (isMainIndex) { termQueryBuilder.field(ROUTING).value(fv -> fv.stringValue(routing)); } var queryParamsStr = String.format("index:%s and finalIndex:%s, routing:%s, isMainIndex:%b, startTime:%s, " + "endTime:%s", index, finalIndex, routing, isMainIndex, String.valueOf(startTime), String.valueOf(endTime)); var asyncSearchRequest = SearchRequest.of(r -> { log.info("start generate query for "); r.index(finalIndex) .preference(routing) .sort(s -> s.field(new FieldSort.Builder().field(UrlMessage.NEXT_FETCH_FIELD_NAME) .order(SortOrder.Asc).build())) .size(size); if (startTime != null || endTime != null) { r.query(q -> q.range(rangeQueryBuilder.build())); } if (isMainIndex) { r.postFilter(f -> f.term(termQueryBuilder.build())); } return r; } ); var queryStartTime = Instant.now(); asyncClient.search(asyncSearchRequest, Map.class) .thenAccept(a -> { var responseDuration = Duration.between(queryStartTime, Instant.now()).toMillis(); log.info("async search response for query params {} took {} ms and response duration {} ms.", queryParamsStr, a.took(), responseDuration); searchByTimeRangeResponseTimeSummary.get(ERROR_TAG, "ok", "none").record(responseDuration); searchByTimeRangeCounter.get(ERROR_TAG, "ok", "none").increment(); responsePromise.complete(a); }) .exceptionally( e -> { var responseDuration = Duration.between(queryStartTime, Instant.now()).toMillis(); var exceptionClassName = e.getClass().getSimpleName(); searchByTimeRangeResponseTimeSummary.get(ERROR_TAG, "fail", exceptionClassName).record(responseDuration); searchByTimeRangeCounter.get(ERROR_TAG, "fail", exceptionClassName).increment(); log.error("async search response failed after {} ms for query params {} and searchRequest" + " {}", responseDuration, queryParamsStr, asyncSearchRequest, e); responsePromise.fail(e); return null; }); return responsePromise.future(); } ``` this is my output log: ```auto 14:35:37.960 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:isna.ir, isMainIndex:true, startTime:2022-01-22T07:58:12.716887Z, endTime:2022-01-29T07:58:12.716887Z took 18898 ms and response duration 305197 ms. 14:35:38.237 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:misaq.me, isMainIndex:true, startTime:2022-03-17T11:39:28.277284Z, endTime:2022-04-11T10:00:38.850104Z took 8908 ms and response duration 299386 ms. 14:35:38.274 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:zil.ink, isMainIndex:true, startTime:2022-03-21T02:51:21.025142Z, endTime:2022-04-11T10:00:38.851925Z took 8850 ms and response duration 299422 ms. 14:35:38.288 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:ninisite.com, isMainIndex:true, startTime:2022-01-22T09:34:47.882911Z, endTime:2022-01-29T09:34:47.882911Z took 8861 ms and response duration 299435 ms. 14:35:38.358 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:isna.ir, isMainIndex:true, startTime:2022-01-22T07:58:12.716887Z, endTime:2022-01-29T07:58:12.716887Z took 19293 ms and response duration 305594 ms. 14:35:38.552 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:weblite.me, isMainIndex:true, startTime:2022-01-18T15:31:48.449286Z, endTime:2022-01-25T15:31:48.449286Z took 8840 ms and response duration 298696 ms. 14:35:38.558 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:zums.ac.ir, isMainIndex:true, startTime:2022-04-11T04:41:20.407427Z, endTime:2022-04-11T10:00:39.856178Z took 8846 ms and response duration 298702 ms. 14:35:38.632 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:tbzmed.ac.ir, isMainIndex:true, startTime:2022-02-19T08:31:30.216535Z, endTime:2022-04-11T10:00:44.760949Z took 1675 ms and response duration 293871 ms. 14:35:38.646 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:tums.ac.ir, isMainIndex:true, startTime:null, endTime:2022-04-11T10:00:43.846905Z took 2026 ms and response duration 294799 ms. 14:35:38.799 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:sheypoor.com, isMainIndex:true, startTime:2022-01-05T16:46:06.714578Z, endTime:2022-01-12T16:46:06.714578Z took 8950 ms and response duration 298939 ms. 14:35:38.807 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:refah.ir, isMainIndex:true, startTime:null, endTime:2022-04-11T10:00:44.760386Z took 1850 ms and response duration 294047 ms. 14:35:38.813 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:imed.ir, isMainIndex:true, startTime:null, endTime:2022-04-11T10:00:37.755708Z took 12398 ms and response duration 301057 ms. 14:35:38.892 [I/O dispatcher 1] INFO IndexClient#lambda$search$33:322 - async search response for query params index:url and finalIndex:url, routing:shatel.ir, isMainIndex:true, startTime:null, endTime:2022-04-11T10:00:43.847387Z took 2271 ms and response duration 295045 ms. ``` This is my search thread pool cat command: ```auto curl -X GET "server:9200/_cat/thread_pool/search?v=true&h=id,name,active,rejected,completed&pretty" ``` and this is its output: ```auto id name active rejected completed FvxD6jlCQk-3-L7i0-mp2w search 0 108381 4725144 PVOAYFI8Qd6Cd_40vC3FKQ search 0 105740 4955279 4Y-0bQiMRbuGV59J6avgEg search 2 106830 4587353 bS0phHy4QuOsIUhs6CEpCg search 2 106432 4787197 3m23_pKPSnuaXFWMhhU-PA search 2 106943 4815429 thSp4pDqSDuqoTWXn8mAlQ search 0 107789 4401516 GmLlKhoWQvuzhMfPP6SB3g search 0 110134 7184929 ``` I think there will be some concurrency issue ---
**Author:** ![saeed\_shahsavan](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/saeed_shahsavan/32/100782_2.png) [@saeed\_shahsavan](https://discuss.elastic.co/u/saeed_shahsavan)\ **Post date:** [April 14, 2022, 6:57am UTC](https://discuss.elastic.co/t/elasticsearch-non-blocking-elasticsearchasyncsearchasyncclient-and-elasticsearchasyncclient/302077/2 "2022-04-14T06:57:32Z")
FInally we have solved the problem by using vertx webClient: ```auto var mustQuery = new JsonObject().put("must", mustJsonArray); var requestBody = new JsonObject() .put("query", new JsonObject().put("bool", mustQuery)) .put("sort", new JsonArray(List.of( new JsonObject().put("nextFetch", new JsonObject().put("order", "asc"))))); webClient.getAbs(queryString) .sendJsonObject(requestBody) .onFailure(responsePromise::fail) .onSuccess(s -> { var searchResponse = new JsonObject(s.bodyAsString()).mapTo(ElasticResponseMessage.class); log.info(searchResponse); responsePromise.complete(searchResponse); }); ``` I think there is a mistake in my implementation. I think I made a mistake using Elasticsearch non blocking client ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [May 12, 2022, 6:58am UTC](https://discuss.elastic.co/t/elasticsearch-non-blocking-elasticsearchasyncsearchasyncclient-and-elasticsearchasyncclient/302077/3 "2022-05-12T06:58:06Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.