cinder cannot attach FC volume

asked 2019-05-27 10:34:48 -0600

joey-opens gravatar image

Hey,

I have an issue while tried to integrate our 3par20800 to OpenStack environment, I edit the configuration file (cinder.conf) so I will able to start and it works fine, I can see inside cinder log there is output that shows the driver loaded successfully. " 2019-05-26 05:38:02.049 124315 INFO cinder.volume.manager [req-a0ba97f0-f557-42e9-bdc7-561a01cdaba6 - - - - -] Initializing RPC dependent components of volume driver HPE3PARFCDriver (4.0.5) 2019-05-26 05:38:02.506 124315 INFO paramiko.transport [-] Connected (version 2.0, client OpenSSH_6.6.1p1) 2019-05-26 05:38:02.747 124315 INFO paramiko.transport [-] Authentication (password) successful! 2019-05-26 05:38:04.193 124315 INFO cinder.volume.manager [req-a0ba97f0-f557-42e9-bdc7-561a01cdaba6 - - - - -] Driver post RPC initialization completed successfully. "

also, I able to create an empty volume via cinder, but, I had an issue once I trying to attach the volume to an instance. in SAN side all zones are configured I also verify that I can create a host and export a volume too, the os on the guest host identify the disk, but when I try to attach the volume from OpenStack side I saw those error inside the nova-compute.log.

"2019-05-26 03:53:55.447 102619 INFO os_brick.initiator.connectors.fibre_channel [-] Fibre Channel volume device not yet found. Will rescan & retry. Try number: 0. 2019-05-26 03:53:57.443 102619 INFO os_brick.initiator.connectors.fibre_channel [-] Fibre Channel volume device not yet found. Will rescan & retry. Try number: 1. 2019-05-26 03:53:59.440 102619 INFO os_brick.initiator.connectors.fibre_channel [-] Fibre Channel volume device not yet found. Will rescan & retry. Try number: 2. 2019-05-26 03:54:01.445 102619 INFO os_brick.initiator.connectors.fibre_channel [-] Fibre Channel volume device not yet found. Will rescan & retry. Try number: 3. 2019-05-26 03:54:03.450 102619 INFO os_brick.initiator.connectors.fibre_channel [-] Fibre Channel volume device not yet found. Will rescan & retry. Try number: 4. 2019-05-26 03:54:05.446 102619 ERROR os_brick.initiator.connectors.fibre_channel [-] Fibre Channel volume device not found. 2019-05-26 03:54:05.446 102619 ERROR oslo.service.loopingcall [-] Fixed interval looping call 'os_brick.initiator.connectors.fibre_channel._wait_for_device_discovery' failed: NoFibreChannelVolumeDeviceFound: Unable to find a Fibre Channel volume device. 2019-05-26 03:54:05.448 102619 ERROR nova.virt.block_device [req-002957eb-4d6d-4ffb-9388-1d414ef702f9 69ba2fbe42c34935a3ad954a21ab5b9f 8ee4d8ed00b542c39906d531db0940af - default default] [instance: c6b7378f-e978-4e80-aadd-6bd49a472c09] Driver failed to attach volume 115439b5-0977-49ad-9ac8-0d40dda3bf62 at /dev/sdb: NoFibreChannelVolumeDeviceFound: Unable to find a Fibre Channel volume device."

the thing is that while the volume in attaching state the 3par volume is export to the compute node ( I also saw when I running "lsblk") but after this error above he back to the available state and the export is gone. Does anyone encourage with that issue or can assist me? I running OpenStack queens(17.1.9) on Ubuntu 16, deploy by OpenStack-ansible. cinder-vol service is on the compute node

edit retag flag offensive close merge delete