Hello,

I also use head with shield and in order to connect to the cluster, in the text box where you're setting the http address, I use the following syntax: http://username:password@localhost:9200.

Bruno

I'm having a similar issue - it was working fine, and all of the sudden i'm continually prompted by shield for the admin username and password I've set up (and is working!)

FYI - I did try bbonnin's idea, with no luck

Same problem here

+1 same problem

Have you guys tried setting authentication parameters as described here? https://github.com/mobz/elasticsearch-head#url-parameters

Well, i tried again without any url parameters... and it is working now. 👍

I don't know why though

still not working for me :(

There is another solution if you are on aws.

Remove the shield plugin and create a security group for your elastic instances that only allows your ip and your other instances.

Isn't there any proper solution yet?

I did what https://github.com/mobz/elasticsearch-head#url-parameters says but not sure if it's a correct way.

Mine is

http://localhost:9200/_plugin/head/index.html?base_uri=http://localhost:9200&auth_user=some_username&auth_password=some_password

and it doesn't work.

I have shield installed. Without shield there's no issue, but with it, I get cluster health: not connected.

Check your elasticsearch.yml. * for http.cors.allow-origin is not allowed for credentialed requests.

Same for me, It was working before I install shields, but now I only have : cluster health: not connected. I am using ES 2.3

@kevin-lee hi,I have shield installed and es-head not connected, you solution it yet?my es version is 2.4.3

When you are using authentication, Ensure that you add http.cors.allow-headers: Authorization. (https://github.com/mobz/elasticsearch-head#basic-authentication)
Then specify auth_user and auth_password as url parameter
Check this guideline: https://github.com/mobz/elasticsearch-head#x-pack

@philipskokoh I did it，but still failed . my elasticsearch.yml config http.cors.enabled: true .http.cors.allow-origin: "*". http.cors.allow-headers: Authorization network.host:0.0.0.0 and url is http://xxx.xx.xx.x:9200/_plugin/head/index.html?base_url=http://xxx.xx.xx.x:9200/?auth_user=elastic&auth_password=changeme

Hi @fgdgtz183,
I think there's no default elastic user in shield for elasticsearch 2.4.x. You need to add your first shield user through command line. Follow this guideline:
https://www.elastic.co/guide/en/shield/current/enable-basic-auth.html

Assume your new shield user is es_admin, and the password is es_admin_password,
if you can get response for curl -XGET 'http://es_admin:es_admin_password@localhost:9200', then you should be able to access your cluster for elasticsearch-head. I just tested this case on elasticsearch-2.4.0.

I also notice that _stats on this elasticsearch version may return unexpected error message (HTTP response 404), if you have no indices in your cluster. So if you cluster is new and you haven't created a new index, create one new index using
curl -XPUT 'http://es_admin:es_admin_password@localhost:9200/test'

hi @philipskokoh
I created an admin user in shield,curl -XGET 'http://es_admin:es_admin_password@localhost:9200，it is working，but es-head not.

At first ,i was using es-head as a plugin of Elasticsearch and open http://localhost:9200/_plugin/head/，It needs to enter a user name and password,but the head plugin can't connect to the nodes，then i remove the head plugin，i tried it running with built in server and specify auth_user and auth_password as url parameters ：
http://localhost:9100/?auth_user=es_admin&auth_password=es_admin_password /，still not working for me .

This is my setting to replicate your issue:

1. config/elasticsearch.yml

http.cors.enabled: true
http.cors.allow-origin: '*'
http.cors.allow-headers: Authorization

2. Shield installed, and then I add a user called es_admin with password es_admin.

3. Elasticsearch is running as standalone server. I connect to the cluster using via this url:
   localhost:9100/?base_uri=http://localhost:9200&auth_user=es_admin&auth_password=es_admin

@philipskokoh
it is working now,beacause my cluster is new and no any indices. thanks for your notice. :)