Kibaba /_search query causes Chrome tab to hang

[melissachang]

I'm guessing there's a memory leak in Kibana and/or Chrome.

Kibana version:
docker image docker.elastic.co/kibana/kibana-oss:6.2.2

Elasticsearch version:
docker image docker.elastic.co/elasticsearch/elasticsearch-oss:6.2.2

Server OS version:
4.18.10-1rodete2-amd64 #1 SMP Debian 4.18.10-1rodete2 (2018-10-09) x86_64 GNU/Linux

Browser version:
Chrome Version 70.0.3538.110 (Official Build) (64-bit)

Browser OS version:

Original install method (e.g. download page, yum, from source, etc.):
docker

Describe the bug:
In the Dev Tools tab, after kicking off a _search query, the Chrome tab hangs.

Steps to reproduce:

1. I created a nurse_s_health_study index:

health status index                       uuid                   pri rep docs.count docs.deleted store.size pri.store.size
yellow open   nurse_s_health_study        ZeV41DhMRnyjWGcMzfz8JQ   5   1     121701            0      5.6gb          5.6gb

2. I have this query in Kibana dev tools. Here is the screenshot before I run the query:
   
   
   

3. I click the green play button. Here are the screenshots several seconds after clicking the button:
   
   
   

In the Resources tab, I can see the Chrome process memory steadily going up. (Now it hovers between 1.8 and 1.9 GiB).

Expected behavior:
Kibana should show the results.

If I copy the curl command in Kibana and run it in a terminal, the command runs fine.

curl -XGET "http://localhost:9200/nurse_s_health_study/_search" -H 'Content-Type: application/json' -d'
{
  "query": {
    "multi_match": {
      "query": "estr",
      "type": "phrase_prefix"
    }
  },
  "from": 0,
  "size": 1000
}'

If I add > /tmp/results.txt:

nhs ~/data-explorer (mr/search_api): ls -lh /tmp/results.txt 
-rw-r----- 1 melchang primarygroup 180M Dec  6 00:04 /tmp/results.txt

So the results are only 180M.
Screenshots (if relevant):

Errors in browser console (if relevant):
None

Provide logs and/or server output (if relevant):
Strangely, I don't see any Elasticsearch logs.

Any additional context:
You can probably repo with any large index (over 5G).

[elasticmachine]

Pinging @elastic/es-ui

[bmcconaghy]

@melissachang We have made a number of changes to console (particularly updating the JSON editor it uses) since 6.2.2. I'm wondering if you can try upgrading to 6.5.x and see if the problem is reproducible on that version?

[melissachang]

Yes, I just reproed with 6.5.4.

Unfortunately my data is private. I'm trying to recreate a similar public dataset for elastic/elasticsearch#36452. If I'm able to, I'll try to repro this bug as well.

Kibana version:
docker image docker.elastic.co/kibana/kibana-oss:6.5.4

Elasticsearch version:
docker image docker.elastic.co/elasticsearch/elasticsearch-oss:6.5.4

Server OS version:
4.18.10-1rodete2-amd64 #1 SMP Debian 4.18.10-1rodete2 (2018-10-09) x86_64 GNU/Linux

Browser version:
Version 71.0.3578.98 (Official Build) (64-bit)

Browser OS version:

Original install method (e.g. download page, yum, from source, etc.):
docker

Describe the bug:
In the Dev Tools tab, after kicking off a _search query, the Chrome tab hangs.

Steps to reproduce:

1. I created a nurse_s_health_study index:

health status index                       uuid                   pri rep docs.count docs.deleted store.size pri.store.size
yellow open   nurse_s_health_study        jgk7VUENSAa4vrkFiVynkw   5   1     121701            0      5.5gb          5.5gb

1. I click the green play button. Here are the screenshots several minutes after clicking the button:
   
   

This time, the Chrome tab process is taking 1.5 G instead of 1.8 G. (Though my index is slightly different.)

Expected behavior:
Kibana should show the results.

The curl command only takes 1.4s.

~/data-explorer/ui (master): time curl -XGET "http://localhost:9200/nurse_s_health_study/_search" -H 'Content-Type: application/json' -d'
{
    "query": {
    "multi_match": {
      "query": "estr",
      "type": "phrase_prefix"
    }
  },
  "size": 1000
}' > /tmp/results.txt 
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  221M  100  221M  100   119   164M     88  0:00:01  0:00:01 --:--:--  164M

real	0m1.393s
user	0m0.023s
sys	0m0.171s
~/data-explorer/ui (master): ls -lh /tmp/results.txt
-rw-r--r-- 1 melchang primarygroup 222M Dec 20 16:03 /tmp/results.txt

[mibragimov]

@cjcenizal I tested this out locally and could not reproduce it? Should we close this?

[cjcenizal]

Closing this as we haven't been able to repro or heard other reports of this bug. @melissachang Could you please reopen this if you're still encountering the problem? Thank you!