Ask Your Question

hayderimran7's profile - activity

2016-03-03 17:35:15 -0500 answered a question why liberty and kilo removes xenapi cinder driver?

Please see this for the related answer http://lists.openstack.org/pipermail/openstack/2015-February/011503.html (http://lists.openstack.org/pipermail/...)

2016-03-03 12:10:06 -0500 answered a question openstack load testing

i think OP wants to have running vms- rally does cleanup after running the task. so it removes all vms after the task is completed. i overcame this by removing the cleanup part from code . lets say you want to run NovaServers.boot_server task https://github.com/openstack/rally/blob/0.3.0/rally/plugins/openstack/scenarios/nova/servers.py#L260 (https://github.com/openstack/rally/bl...) , in the cleaup change "nova" to "" , that way after rally completes the task, you will see the vms are running .

2016-03-03 11:53:31 -0500 received badge  Supporter (source)
2016-03-03 11:52:26 -0500 answered a question Is the true that mulitple users in a same Project have the same privilege to manage any of the instances or volumes created in the Project, e.g. delete or edit an instance or a volume that was created by other people in the same project?

No, it depends upon the role assigned to each user in that project. A project is just like a grouping entity for organizing users, however user privileges depend upon the role assigned to them.
And the power of role is determined by how you define your policy.json for each service. for instance Member role may list all images but is not allowed to create an image.

2016-03-03 01:07:51 -0500 answered a question Tempest test NeutronResourcesTestJSON fails with "KeyError: 'physical_resource_id'"

Hi i figured this out why it was happening. This is because when you run tempest test tempest.api.orchestration.test_neutron_resources with the default given template neutron_basic.yaml, that template uses waitcondition api of heat that you define in your heat.conf in mine it was : heat_waitcondition_server_url = https://10.240.134.179:8000/v1/waitcondition (https://10.240.134.179:8000/v1/waitco...)

you have to make sure you can curl this api from inside your vm , any nova vm . if you cannot curl , you simply will get timeout on that wait condition, because the curl keeps failing to connect to /v1/waitcondition api, and unless that waitcondition is fulfilled, your stack status will never be in COMPLETE state.

now the template creates a wait condition, wait condition is fulfilled when the vm lauched through the stack sends a signal. the signal is using the following command defined in neutron_basic.yaml:

        #!/bin/sh -v
        SIGNAL_DATA='{"Status": "SUCCESS", "Reason": "SmokeServerNeutron created", "Data": "Application has completed configuration.", "UniqueId": "00000"}'
        while ! curl --fail -X PUT -H 'Content-Type:' --data-binary "$SIGNAL_DATA" \
        'wait_handle' ; do sleep 3; done

this is the command runs inside your vm, how i know this, when the test is running, you can go to horizon, click orchestration, click on tempest stack and click on the server > log. i could see the error:

curl: failed to curl localhost by simply removing the following part from neutron_basics.yaml, your test shall pass :)

     REMOVE FOLLOWING PART FROM NEUTRON_BASIC.YAML      
     user_data_format: RAW
      user_data:
      str_replace:
      template: |
        #!/bin/sh -v
        SIGNAL_DATA='{"Status": "SUCCESS", "Reason": "SmokeServerNeutron created", "Data": "Application has completed configuration.", "UniqueId": "00000"}'
        while ! curl --fail -X PUT -H 'Content-Type:' --data-binary "$SIGNAL_DATA" \
        'wait_handle' ; do sleep 3; done
      params:
        wait_handle: {get_resource: WaitHandleNeutron}
     WaitHandleNeutron:
         type: AWS::CloudFormation::WaitConditionHandle
     WaitCondition:
         type: AWS::CloudFormation::WaitCondition
depends_on: Server
properties:
  Handle: {get_resource: WaitHandleNeutron}
  Timeout: {get_param: timeout}

After you removed this , rerun and it shall pass :)

2015-09-25 00:06:43 -0500 received badge  Famous Question (source)
2015-07-12 15:12:54 -0500 received badge  Notable Question (source)
2015-07-11 23:08:13 -0500 received badge  Popular Question (source)
2015-07-10 11:12:34 -0500 asked a question Why is glance a separate service not part of Nova, since only Nova needs it?

I'm curious to know why glance is a separate service, with endpoints and service users since its only needed by Nova, why not just make it a part of Nova then ?
whats the technical motivation of having a separate image registry service? when compared to AWS, they dont have a separate image service for vms..
i would really like and appreciate the details on it :)

2015-07-10 11:12:34 -0500 asked a question Why isn't glance just a part of Nova, since its only service needing it?

I'm curious to know why glance is a separate service, with endpoints and service users since its only needed by Nova, why not just make it a part of Nova then ?
whats the technical motivation of having a separate image registry service? when compared to AWS, they dont have a separate image service for vms..
i would really like and appreciate the details on it :)