PowerFlex: Troubleshooting OpenStack Cinder Installation
Résumé: This article describes the steps to troubleshoot a ScaleIO Cinder installation for OpenStack.
Instructions
To configure the OpenStack Cinder driver to use PowerFlex, add the following to the /etc/cinder/cinder.conf file.
This configuration is for the Pike version of OpenStack. Other versions may vary.
[Default] enabled_backends = scaleio [scaleio] volume_driver = cinder.volume.drivers.dell_emc.scaleio.driver.ScaleIODriver volume_backend_name = scaleio san_ip = GATEWAY_IP sio_storage_pools = Domain1:Pool1,Domain2:Pool2 san_login = SIO_USER san_password = SIO_PASSWD san_thin_provision = false
Review the /var/log/cinder/volume.log file. You should see lines such as the following:
2018-05-28 14:04:10.659 28321 INFO cinder.volume.drivers.emc.scaleio [req-4584c6b0-9d6b-4745-a63f-45b19fbd0b46 - - - - -] ScaleIO get domain id by name request: https://10.241.215.218:443/api/types/Domain/instances/getByName::H2O_PD1. 2018-05-28 14:04:10.863 28321 INFO cinder.volume.drivers.emc.scaleio [req-4584c6b0-9d6b-4745-a63f-45b19fbd0b46 - - - - -] ScaleIO get pool id by name request: https://10.241.215.218:443/api/types/Pool/instances/getByName::a8341aab00000000,H2O_POOL2.
If you do not, you have a communications issue with the ScaleIO gateway. Verify the connectivity with the gateway by running the following command from the OpenStack Cinder server:
curl -u : https:///api/login --insecure
The command should return an authentication token. If you receive an error, you have a communication issue with the ScaleIO gateway and should resolve that first.
Next, ensure that the Cinder service is set up properly.
From the controller node, log in to the OpenStack system (for example, source ~/demorc_admin), then view the cinder services:
cinder service-list
You should see output like the following:
+------------------+----------------+------+---------+-------+----------------------------+-----------------+ | Binary | Host | Zone | Status | State | Updated_at | Disabled Reason | +------------------+----------------+------+---------+-------+----------------------------+-----------------+ | cinder-scheduler | controller | nova | enabled | up | 2018-05-28T18:19:21.000000 | - | | cinder-volume | cinder@scaleio | nova | enabled | up | 2018-05-28T18:19:24.000000 | - | +------------------+----------------+------+---------+-------+----------------------------+-----------------+
If you do not see a cinder-volume service with @scaleio or if the state of that service is down, you must check that service.
From the server running the cinder-volume, check the service status:
Service openstack-cinder-volume status
If the service is not running, start it and check the status again. If it fails to start, check the /var/log/cinder/volume.log file for errors.
Note: The status of the SED devices is displayed in the output as managed, but unencrypted.