Kibana version: 5.2.2
Elasticsearch version: 5.2.2
Server OS version: CentOS 7.3
Browser version: Chrome 57.0.2987.98
Browser OS version: Win 8
Original install method (e.g. download page, yum, from source, etc.): kibana-5.2.2-x86_64.rpm
Description of the problem including expected versus actual behavior:
After upgrading from kibana 5.1 to kibana 5.2, kibana becomes unusable (only the left bar appears, nothing in the main view).
Steps to reproduce:
Errors in browser console (if relevant):
POST http://log1/es_admin/.kibana/index-pattern/_search?stored_fields= 400 (Bad Request)
Error: Bad Request
at respond (kibana.bundle.js?v=14695:14)
at checkRespForFailure (kibana.bundle.js?v=14695:14)
at kibana.bundle.js?v=14695:1
at processQueue (commons.bundle.js?v=14695:38)
at commons.bundle.js?v=14695:38
at Scope.$eval (commons.bundle.js?v=14695:39)
at Scope.$digest (commons.bundle.js?v=14695:39)
at Scope.$apply (commons.bundle.js?v=14695:39)
at done (commons.bundle.js?v=14695:37)
at completeRequest (commons.bundle.js?v=14695:37)
Provide logs and/or server output (if relevant):

Thanks for filing @jnzz. Can you share the response for the request under the network tab?

(anonymous) @ VM212:1
(anonymous) @ commons.bundle.js?v=14695:37
sendReq @ commons.bundle.js?v=14695:37
serverRequest @ commons.bundle.js?v=14695:37
processQueue @ commons.bundle.js?v=14695:38
(anonymous) @ commons.bundle.js?v=14695:38
$eval @ commons.bundle.js?v=14695:39
$digest @ commons.bundle.js?v=14695:39
(anonymous) @ commons.bundle.js?v=14695:39
completeOutstandingRequest @ commons.bundle.js?v=14695:36
(anonymous) @ commons.bundle.js?v=14695:36
Sorry, can you click on the red request name and then click on the response tab, it will look something like this:

The response was:
No handler found for uri [/es_admin/.kibana/index-pattern/_search?stored_fields=] and method [POST]
But I found the error - and it's my old nginx reverse proxy settings for Kibana.
Sorry for bothering you.
Awesome, no problem.
I had the same trouble and it took me a surprisingly long time to isolate the cause - which is my problem. It looks like there is a new request from kibana to /es_admin in version 5.2.2
For anyone (lots of us, I imagine) running a proxy this will break the proxy interface. Not the responsibility of Elastic, but it would be great to know about these kinds of changes for future updates.
A request would be for there to be a note in future release documents outlining new or changed UI folder names so that proxy admins can adjust as needed. This was a 2- minute fix for me, but it took me hours to identify (because it didn't cross my mind that there would be a new call like this).
What was the error wrt "But I found the error - and it's my old nginx reverse proxy settings for Kibana."?
I have the same problem & am using a nginx reverse proxy......
nginx configuration should looks like
location / {
proxy_pass http://localhost:5601;
proxy_http_version 1.1;
}
and no other location-s
Most helpful comment
I had the same trouble and it took me a surprisingly long time to isolate the cause - which is my problem. It looks like there is a new request from kibana to /es_admin in version 5.2.2
For anyone (lots of us, I imagine) running a proxy this will break the proxy interface. Not the responsibility of Elastic, but it would be great to know about these kinds of changes for future updates.
A request would be for there to be a note in future release documents outlining new or changed UI folder names so that proxy admins can adjust as needed. This was a 2- minute fix for me, but it took me hours to identify (because it didn't cross my mind that there would be a new call like this).