-
Notifications
You must be signed in to change notification settings - Fork 8.3k
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.
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
[4.x] Kibana issuing multiple _msearches causing slow performance #8128
Comments
Fixing this should be a pretty big performance win. |
This is not a bug, it is by design. One big PS: The reason you get one request when adding _score is due to the relative nature of scoring. |
Funny, I would like report the opposite, to me multi-search is more expensive than running individual widget searches (I can see in browser dev console, dashboard _msearch took 2.000ms where as running each widget queries individually took no more than a few millisecondes). |
I instantly ran into this after upgrading from ES 1 / kibana 2. My dashboards with 400k in the old system load in 100ms, while kibana 6.x takes 700-800ms just to run discovery on 5k records. This is all because of _msearch in my opinion. Horrible end point. |
Kibana initially sends multiple _msearch requests each with single search and single unique index name, instead of calling _msearch once with an array of multiple indices.
Kibana version: 4.5
Elasticsearch version: 2.3
Server OS version: Mac OS X 10.11.6
Browser version: Firefox v44
Browser OS version: Linux 64-bit
Original install method (e.g. download page, yum, from source, etc.): Downloads page
Description of the problem including expected versus actual behavior: Initial load of Discover page, after broadening the time frame to require multiple indices in the search,
Steps to reproduce:
/elasticsearch/_msearch
requests sent to Elasticsearch. ← Here is the performance problem._score
in the field list, sort by the_score
column, and refresh the page/elasticsearch/_msearch
requests sent to Elasticsearch._score
column, and refresh the page/elasticsearch/_msearch
requests sent to Elasticsearch. ← This is how the request should have been crafted after first setting the time range to "Last 7 days" in step 3.The text was updated successfully, but these errors were encountered: