DIY cloudbreak docker login

[Dhiraj Peechara]

Hi,

    I am trying to install cloud break docker on EC2 for self hosting. I am unable to login and below are the logs ::

15:18:20.433 [http-nio-8080-exec-3] INFO  o.s.b.a.a.l.AuditListener - AuditEvent [timestamp=Wed Nov 05 15:18:20 UTC 2014, principal=<MY_EMAIL>, type=AUTHENTICATION_SUCCESS, data={details=org.springframework.security.web.authentication.WebAuthenticationDetails@3bcc: RemoteIpAddress: 173.80.71.220; SessionId: 3F5907301AB0A7B7A6D15D67FF9E09BA}]

15:18:21.032 [http-nio-8080-exec-4] ERROR o.a.c.c.C.[.[.[.[dispatcherServlet] - Servlet.service() for servlet [dispatcherServlet] in context with path [] threw exception

java.io.FileNotFoundException: class path resource [favicon.ico] cannot be resolved in the file system for resolving its last-modified timestamp

at org.springframework.core.io.AbstractResource.lastModified(AbstractResource.java:155) ~[spring-core-4.0.3.RELEASE.jar!/:4.0.3.RELEASE]

at org.springframework.core.io.AbstractFileResolvingResource.lastModified(AbstractFileResolvingResource.java:169) ~[spring-core-4.0.3.RELEASE.jar!/:4.0.3.RELEASE]

while the browser is redirect to error page.

Whitelabel Error Page

This application has no explicit mapping for /error, so you are seeing this as a fallback.

I am on git branch : 0.1.21

Thank you

[Krisztian Horvath]

Hi,

The screen you copied here is the backend. If you try the same URL + /health you’ll see something like this: {"status":"ok","database":"PostgreSQL","hello":1}. Are you trying to deploy the oauth based cloudbreak?

Br,

Krisz

--
You received this message because you are subscribed to the Google Groups "SequenceIQ Cloudbreak" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloudbreak+...@googlegroups.com.
To post to this group, send email to cloud...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cloudbreak/5c785f75-c9df-4302-9322-5f1dcaa87f22%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Dhiraj Peechara]

My goal is to set up multi-node-yarn cloud using self hosted cloudbreak in ec2. Once I have this set up I need to use periscope to be able to dynamically scale the cluster. 

To have periscope to be able to dynamically scale I believe I will need sultanas as well ?

no I am not trying to set up oauth based cloudbreak. I am trying to set up self hosted cloud beak in EC2 with instructions from QuickStart and installation. I do see {"status":"ok","database":"PostgreSQL","hello":1} when I go to /health

Thank you

[Krisztian Horvath]

Hi,

Did you issue the following commands?

git clone https://github.com/sequenceiq/docker-cloudbreak.git

cd docker-cloudbreak

git checkout 0.1.21

./start_cloudbreak.sh

We switched from 0.1.21 to a newer version, but the documentation didn’t follow this change, I’ll update it. Please run:

git checkout master

git pull

and then try to run the ./start_cloudbreak.sh script which will deploy it automatically for you.

Br,

Krisz

To view this discussion on the web visit https://groups.google.com/d/msgid/cloudbreak/489a74ad-26ad-4107-b438-2e3e03f02542%40googlegroups.com.

[Dhiraj Peechara]

I got the latest from master and still I am getting this

curl 127.0.0.1:8080

{"timestamp":1415207442188,"error":"Not Found","status":404,"message":""}

curl 127.0.0.1:8080/health

{"status":"ok","database":"PostgreSQL","hello":1}

in the logs I see below error

17:14:32.656 [http-nio-8080-exec-4] ERROR o.a.c.c.C.[.[.[.[dispatcherServlet] - Servlet.service() for servlet [dispatcherServlet] in context with path [] threw exception

java.io.FileNotFoundException: class path resource [favicon.ico] cannot be resolved in the file system for resolving its last-modified timestamp

at org.springframework.core.io.AbstractResource.lastModified(AbstractResource.java:155) ~[spring-core-4.0.3.RELEASE.jar!/:4.0.3.RELEASE]

at org.springframework.core.io.AbstractFileResolvingResource.lastModified(AbstractFileResolvingResource.java:169) ~[spring-core-4.0.3.RELEASE.jar!/:4.0.3.RELEASE]

at org.springframework.web.servlet.resource.ResourceHttpRequestHandler.handleRequest(ResourceHttpRequestHandler.java:144) ~[spring-webmvc-4.0.3.RELEASE.jar!/:4.0.3.RELEASE]

at org.springframework.web.servlet.mvc.HttpRequestHandlerAdapter.handle(HttpRequestHandlerAdapter.java:51) ~[spring-webmvc-4.0.3.RELEASE.jar!/:4.0.3.RELEASE]

I see below docker containers running : docker ps

cbee2c2e8a74        sequenceiq/uluwatu:latest      "/uluwatu/start.sh"    6 minutes ago       Up 6 minutes        0.0.0.0:3000->3000/tcp             uluwatu

71268382f0b3        sequenceiq/sultans:latest      "/sultans/start.sh"    6 minutes ago       Up 6 minutes        3001/tcp, 0.0.0.0:8081->8080/tcp   sultans

27b381fb2464        sequenceiq/cloudbreak:latest   "/start_cloudbreak_a   7 minutes ago       Up 7 minutes        0.0.0.0:8080->8080/tcp             cloudbreak

62eaff5f39a6        sequenceiq/uaa:latest          "/tmp/run.sh"          7 minutes ago       Up 7 minutes        8080/tcp                           uaa

fe394df40d60        postgres:latest                "/docker-entrypoint.   7 minutes ago       Up 7 minutes        5432/tcp                           uaadb

f27352a83e67        postgres:latest                "/docker-entrypoint.   8 minutes ago       Up 8 minutes        0.0.0.0:5432->5432/tcp             postgresql

[Krisztian Horvath]

Hi,

127.0.0.1 supposed to be the backend, for the UI (uluwatu container) please go to localhost:3000.

Br,

Krisz

To view this discussion on the web visit https://groups.google.com/d/msgid/cloudbreak/85254011-6e92-428d-83cd-56c8dcf87c4b%40googlegroups.com.

[Dhiraj Peechara]

I am trying to install docker-cloudbreak on an ec2 instance. On redirecting it is picking up the internal IP. Even on changing the ip to real ip in browser i still don't get the login prompt.

URL :: http://<PUBLIC_IP>:3000  redirecting to http://172.17.0.68:8080/oauth/authorize?response_type=code&client_id=uluwatu&scope=openid+cloudbreak.templates+cloudbreak.credentials+cloudbreak.blueprints+cloudbreak.stacks&redirect_uri=http://localhost:3000/authorize

even changing the internal ip to public ip in the browser will not help.

[Marton Sereg]

The start_cloudbreak.sh script is mainly for development purposes, it won't work for a hosted solution out of the box. It is set up to work on localhost, that's why you are redirected to the internal ip. The public ip won't work, you can see in the script yourself that no ports are forwarded from the UAA docker container so it won't be available from outside.

If you want to host it on AWS (or anywhere else), you'll have to make some changes to the start_cloudbreak.sh script:

- you'll have to specify an UAA configuration file like this one: https://github.com/sequenceiq/docker-uaa/blob/master/dev.yml (update the default users and the localhost part of the redirect uri of uluwatu to a publicly available address - e.g.: http://public-aws-address:3000/authorize)

  you can find more about the UAA configuration here: http://blog.sequenceiq.com/blog/2014/10/16/using-uaa-as-an-identity-server/

- you must forward the 8080 port of the UAA docker container so it will be available from outside

- you'll have to set the proper environment variables when starting cloudbreak, uluwatu and sultans to point to the publicly available UAA server.

  e.g: if UAA is running on http://public-aws-address:8080 set ULU_IDENTITY_ADDRESS to this url.

Btw do you have any specific reasons to deploy the infrastructure yourself and not to use the hosted solution?

[Kevin Kuhl]

Hi Marton,

I'm not sure if the OP ever answered you, but I am interested in deploying my own version of this. Reasons are that I want to provide similar functionality to users as part of an offering, however I will likely need custom authorization/authentication for an existing user base, experiment with different ambari versions (my services may require changes to what's currently supported), different stacks deployed with ambari including some proprietary stuff, maybe reimplement/brand UI and to hook into the CloudBreak API from other places in a larger deployed platform we have. In a nutshell, some users may want to deploy their own fork of this (or at least reuse some of the engine thats at the core) for their own offerings.

Hope this provides some detail on why users would want to be deploying their own.

I'm assuming the application stack SequenceIQ has deployed in production does not run on single servers containing multiple docker containers (I could be wrong). If we were to experiment with the code in our own environment we would need the constituent parts and databases hosted separately as well. If I'm wrong on any of these points feel free to correct me :).

Thanks,

Kevin

[Janos Matyas]

Kevin

Part of the coming commercial package we encourage everybody to host their own Cloudbreak. Most of the answers for your question is yes - if you can send us your contact details we can touch-base and discuss these offline.

Janos

[Pratik Shah]

Hi , 

I would be helpful if you share the updated steps to deploy cloudbreak.

I have tested the hosted solution which worked very well and also lead me to explore the DIY solution as well. 

Note : i am using a free trial version on GCP and would like to deploy cloudbreak on it.

Regards

Pratik