Hello Everyone, I have installed Cloudbreak on Azure VM and provisioned HDP using it which has also enabled Periscope (default) Web UI (Uluwatu). This has helped me setting manual alarms and scaling policy from the UI. However, my aim is to achieve the same through API calls. I was able to connect Cloudbreak using below API method, first generate the access token and then connect to Cloudbreak through API: Token Generation for Cloudbreak Shell (Worked): export TOKEN=$(curl -iX POST -H "accept: application/x-www-form-urlencoded" -d 'credentials={"username":"admin@example.com","password":"password"}' "http://cloudbreakURI:8089/oauth/authorize?response_type=token&client_id=cloudbreak_shell" | grep Location | cut -d'=' -f 3 | cut -d'&' -f 1) API call to Cloudbreak (Worked): curl -k -X GET -H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" https://cloudbreakUI/cb/api/v1/stacks/user Now, I want to do the same thing with Periscope. I have followed the Periscope API "https://periscope.docs.apiary.io" which explains that we need to attach Ambari REST API Endpoint with Periscope. Here I got confused that if Periscope is running by default in Cloudbreak machine and I am able to set alarms in Periscope UI using Ambari alerts then why do I need to attach with a an existing cluster for API calls? Periscope is running in CloudbreakVM: [hduser@cloudbreakVM ~]$ ps -ef | grep periscope hduser 41503 38823 0 06:55 pts/0 00:00:00 grep --color=auto periscope root 66067 58001 0 Nov24 ? 00:00:00 /bin/bash /start_periscope_app.sh root 66075 66067 0 Nov24 ? 00:07:55 java -jar /periscope.jar In one of the github repository, I found below call method to generate the token for Periscope client but it doesn't generate in my case: Token Generation for Periscope client (Doesn't work): curl -iX POST -H "accept: application/x-www-form-urlencoded" -d 'credentials={"username":"admin@example.com","password":"password"}' "http://cloudbreakURI:8089/oauth/authorize?response_type=token&client_id=periscope-client&scope.0=openid&source=login&redirect_uri=http://periscope.client" ERROR: HTTP/1.1 500 Internal Server Error Server: Apache-Coyote/1.1 Cache-Control: no-store X-XSS-Protection: 1; mode=block X-Frame-Options: DENY X-Content-Type-Options: nosniff Content-Language: en Content-Length: 0 Date: Sat, 25 Nov 2017 06:57:19 GMT Connection: close I also tried adding Ambari REST API endpoint as follows but no luck: Attaching Ambari REST API to Periscope (Doesn't work): curl --include \ --request POST \ --header "Content-Type: application/json" \ -H "Authorization: Bearer $TOKEN" \ --data-binary "{ \"host\": \"AmbariURI\", \"port\": \"8080\", \"user\": \"admin\", \"pass\": \"admin\" }" \ 'http://AmbariURI:8080/api/v1/clusters/' ERROR: { "status": 403, "message": "Missing authentication token" } When I try to directly calls from Ambari REST API, it doesn't return any result: curl -k -X GET -H "Content-Type: application/json" --user admin:admin http://AmbariURI:8080/api/v1/clusters/<clustername>/policies curl -k -X GET -H "Content-Type: application/json" --user admin:admin http://AmbariURI:8080/api/v1/clusters/<clustername>/alarms Can someone help in connecting Periscope APIs for autoscaling on a running HDP cluster. Regards, ... View more

Hello Team, I have couple of questions on provisioning datanodes using cloudbreak where I would like to know the approx time in commissioning a data node to the cluster (assuming node is available with all prerequisites (OS) to get started for commissioning) ? And, if I want to provision 50 datanodes, does the parallelism of provisioning those nodes depends on cores available on Ambari Server? Example: 1. Ambari Server with 32 cores 2. Assuming, 20 cores available on Ambari server for this process 3. Requirement: 50 datanodes (autoscaling) Does this mean that only 20 nodes will be provisioned parallel as a first batch (1 core = 1 datanode provisioning) and then 20 followed by 10? So, if 15 mins are taken to provisioning 1 batch (20 nodes in parallel) then 50 nodes would take 45 mins for provisioning? Regards, Sakhuja ... View more