Kibana: Kibana4 Discover throws error: "An error occured with your request. Reset your inputs and try again"

Created on 3 Mar 2015  路  14Comments  路  Source: elastic/kibana

Hi, I just upgraded to kibana-4.0.0-linux-x64 from beta3.

Now the Discover page is not working correctly anymore for me. I get a red warning "Discover: An error occured with your request. Reset your inputs and try again". But I have no inputs set, nothing is selected, I don't know what to change. When I press "new search" I immediately get the error.

The Dashboard seems okay, even though I had to remove some Visualizations which suddenly threw errors after the upgrade.

The Discover page should work when I start with a "New search", the error message should be more descriptive, have a "more info" button with error details, maybe even directly enable to "reset inputs" to get a plain state which works.

kibana4error

feedback_needed

Most helpful comment

Why this issue closed?
clear cache not help :(

wtf is wrong with this piece of shit

All 14 comments

Upgrades from Beta 3 are not supported. Please remove the .kibana index. You may also need to remove everything in the URL after the ?

Removing from URL doesn't help, Kibana seems to automatically add those parameters by default.

I removed the .kibana index with curl -XDELETE 'http://localhost:9200/.kibana/'

Now the Discover works again - and I have to recreate my Dashboard with 20 Visualizations.

This is occurring intermittently for me on a fresh install of 4.0.0 with ES 1.4.4

There is no .kibana index to destroy, and the issue re-occurs on deleting the kibana-int index.

I have the same issue with Kibana 4.0.0 and ES 1.4.4.
I've deleted the .kibana index but it doesn't help.

I had the same issue in a local pre-production-setup.

$bin/kibana --version
4.0.2

$curl 'localhost:9200/'                                                                          [10:55:20]
{
  "status" : 200,
  "name" : "logstash-els-vagrant",
  "cluster_name" : "els-vagrant",
  "version" : {
    "number" : "1.5.1",
    "build_hash" : "5e38401bc4e4388537a615569ac60925788e1cf4",
    "build_timestamp" : "2015-04-09T13:41:35Z",
    "build_snapshot" : false,
    "lucene_version" : "4.10.4"
  },
  "tagline" : "You Know, for Search"
}

I'v tried changing the kibana_index to a new index to try with "fresh" settings. with no luck.

I finally fixed it by removing all Indices, but this is not a valid solution for people running in production.

It seems that it appears when you change a mapping type for some field. I had to restart ElasticSearch to fix that.

@lifer-da-man Forgot to mention that i tried restarting all services and even ended up rebooting the VM before removing all Indices

EDIT: I was playing around with the time-filter and the problem reappeard :(

The culprit was "request_timeout" in kibana.yml, The value has apparently been changed from seconds to miliseconds.

As a result my requests to elasticsearch timed out after a wooping 60ms. :)

@devnore <3

I'm running into the same problem, with a fresh install of ES 1.5.2 and Kibana 4.0.2, Linux 64bit. Neither of the above options resolved the issue for me. 'request_timeout' seemed to be fine out of the box (set to a value of 300000).

Another thing which I noticed is that 'Discover' didn't display any fields in the sidebar for me. Only if I unchecked the 'Hide Missing Fields' option did I get the fields from my mapping listed.

I also posted the issue (with a bit more details) on stackoverflow here

Ouch - found the error. (Mine, not Kibana's - as usual... ;-) The issue was malformed JSON in my case, which seemed to work in ES, but then not further down the line in Kibana. Fixed the JSON and all is well.

Clearing the elasticsearch cache did it for me:
curl localhost:9200/_cache/clear

This was caused when I terminated the source sending JSON documents into logstash, via redis input

Hi

Clearing the Elasticsearch cache work for me too. Thank you.

curl localhost:9200/_cache/clear

This error occur, when I deleted all the objects, search, dashboard and Visual.

Why this issue closed?
clear cache not help :(

wtf is wrong with this piece of shit

Was this page helpful?
0 / 5 - 0 ratings