Kibana error: (Unknown provider: sessionTimeoutProvider)

[stefandoorn]

Kibana version: 5.x

Elasticsearch version: 5.x

Server OS version: Centos 7

Browser version: Chrome

Browser OS version: Mac OSX

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

Steps to reproduce:
I installed Elasticsearch using yum following the instructions on the website. Afterwards the same for Kibana. Both work when accessing them through the browser. Then I installed x-pack following the instructions (first for ES, then for Kibana). Restarting the services and reloading Kibana gives me the errors below. Uninstall also breaks Kibana, so I'm kind of stuck now with it.

Errors in browser console (if relevant):
browser:
Fatal Error
Uncaught Error: [$injector:unpr] Unknown provider: sessionTimeoutProvider <- sessionTimeout <- $http <- config
http://errors.angularjs.org/1.4.7/$injector/unpr?p0=sessionTimeoutProvider%20%3C-%20sessionTimeout%20%3C-%20%24http%20%3C-%20config (http://192.168.22.22:5601/bundles/commons.bundle.js?v=14476:36)
Version: 5.0.2
Build: 14476
Error: Uncaught Error: [$injector:unpr] Unknown provider: sessionTimeoutProvider <- sessionTimeout <- $http <- config
http://errors.angularjs.org/1.4.7/$injector/unpr?p0=sessionTimeoutProvider%20%3C-%20sessionTimeout%20%3C-%20%24http%20%3C-%20config (http://192.168.22.22:5601/bundles/commons.bundle.js?v=14476:36)
at window.onerror (http://192.168.22.22:5601/bundles/commons.bundle.js?v=14476:69:13503)

console:
commons.bundle.js?v=14476:55 Error: Uncaught Error: [$injector:unpr] Unknown provider: sessionTimeoutProvider <- sessionTimeout <- $http <- config
http://errors.angularjs.org/1.4.7/$injector/unpr?p0=sessionTimeoutProvider%20%3C-%20sessionTimeout%20%3C-%20%24http%20%3C-%20config (http://192.168.22.22:5601/bundles/commons.bundle.js?v=14476:36)
at window.onerror (commons.bundle.js?v=14476:69)Notifier._showFatal @ commons.bundle.js?v=14476:55
commons.bundle.js?v=14476:36 Uncaught Error: [$injector:unpr] Unknown provider: sessionTimeoutProvider <- sessionTimeout <- $http <- config(…)
commons.bundle.js?v=14476:55 Uncaught Error: Uncaught Error: [$injector:unpr] Unknown provider: sessionTimeoutProvider <- sessionTimeout <- $http <- config(…)
http://192.168.22.22:5601/app/kibana Failed to load resource: the server responded with a status of 503 (Service Unavailable)

Provide logs and/or server output (if relevant):

[epixa]

@stefandoorn Can you delete the optimize/bundles directory as well as the optimize/.babel-cache.json file and restart Kibana? That will trigger a completely fresh rebuild of the Kibana source with x-pack.

Do you get any errors like this in your kibana logs: #8818 ?

Also, I recommend moving this conversation to IRC or the forums in order to get more timely feedback, as we use github for bugs and feature requests.

[nbari]

Had the same issue, just rebooted server and working now.

[epixa]

@nbari Did you need to delete the optimize/bundles directory, or was a restart alone enough to fix it?

[nbari]

First I deleted optimize/bundles and optimize/.babel-cache.json restarted kibana and didn't work, then I just did a reboot and got it working.

Later I just add xpack.security.enabled: false in both elasticsearch.yml and kibana.yml

[tronicum]

I had a nginx with basic auth in front of kibana. having two of those chained, did not work...!
You might add this to the documentation as many people "secure" their kibana with a reverse proxy if they dont use x-pack.

[cptcanuck]

I was having this problem as well. Fresh install of ES and Kibana with x-pack, running with NGINX and basic auth in front.

Doing as @nbari suggested, disabling xpack security, worked.

This seems related to #9226 as well.

[howardjones]

Just been through the same here, with 5.2.2 from the yum repo of each application. Disabling xpack security did work, but I was hoping to look at AD integration and role-based access control, so it's not too helpful!

[mfn]

Can you delete the optimize/bundles directory as well as the optimize/.babel-cache.json file and restart Kibana? That will trigger a completely fresh rebuild of the Kibana source with x-pack.

Worked for me; note: the restart took minutes until kibana node server responded via http.
With no log, no output, nothing this was irritating but at some point I realized that files in optimize/bundles/ accumulated and just waited.

[epixa]

@mfn Yeah, the code transpilation can take a really long time, and it's maddening. This is the biggest thing we're looking to address with the new build system, which would cut that down to nothing and make this whole problem go away.

[hsafe]

Hi all,
I was having identical issue and still not resolved for elk stack 5.4 on centos7 nginx reverse proxy with simple authentication on.
I tried all the steps here: delete the files under optimize/ folder and rebooted and alternatively setting the
xpack.security.enabled: false or disabling simple authentication directive in nginx.
I should state that I waited enough for the kibanna to rebuild its internals as well...
I went as far as from a non-listening kibana on default port to listening kibana...but then lost all together logging and visualization. Now the service is up filebeats at clients are OK and shipping to logstash but no more record of the graphs... I do not want to frantically jump here and there or revert...appreciate to help.
Current status is :xpack.security.enabled: false in kibana and elastic yaml and basic security in nginx is on.
Can someone please help?

[thomasneirynck]

Closing due to inactivity. Can reopen if problem persists.

[navnitt]

What is the final solution to this?

[mfn]

@navnitt back #9369 (comment) in this worked for me: #9369 (comment)

[navnitt]

restarting kibana after deleting optimize is not working for me.

[navnitt]

restarting kibana after deleting optimize is not working for me.

[navnitt]

changed the ownership of optimize after reboot worked.But it is aking for a password and I dint reset any password.

[JoshuaSmeda]

changed the ownership of optimize after reboot worked.But it is aking for a password and I dint reset any password.

The above solution does work. If you still experiencing an issue, please open a new issue. There is no need to troubleshoot your issue on an existing issue when a solution has already been provided.

[BobBlank12]

OK... similar issue. After deleting the bundle and babel-cache, restart of kibana still no go. ps -ef | grep kibana showed another kibana process still running. Make SURE there isn't an additional Kibana instance running or orphaned.

[rishiloyola]

Why did you close this @thomasneirynck? I am still having an issue with this. Please open it to solve the issue.

[Leaf-Lin]

It's still affecting 6.5.0 (with x-pack comes preinstalled).
I can confirm the workaround fix the issue:

1. deleted optimize/bundles and optimize/.babel-cache.json
2. ps -ef | grep kibana to check if there’s another kibana process still running, if there are, kill -9 PID or restart the server. Make sure there's only one kibana running per host!
3. restarted kibana (this will take longer than usual to rebuild)

---

But we'd also like to know why it breaks at the first place. I'm unable to see root cause for 6.5.0 as there's no x-pack installation involved.

[Bargs]

The comment above as well as this comment from a duplicate ticket both report this error appearing suddenly after x-pack had already been installed and working for some time. I think that's worth looking into, so I'm going to re-open this ticket.

[elasticmachine]

Pinging @elastic/kibana-operations

[elasticmachine]

Pinging @elastic/kibana-platform