emc-openstack / vnx-direct-driver Goto Github PK

Hi,

If Openstack creating the VM, but the VM timed out, cinder try to remove LUN from EMC, but it's not work.

2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher Traceback (most recent call last):
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line 134, in _dispatch_and_reply
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher incoming.message))
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line 177, in _dispatch
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher return self._do_dispatch(endpoint, method, ctxt, args)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line 123, in _do_dispatch
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher result = getattr(endpoint, method)(ctxt, *_new_args)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/osprofiler/profiler.py", line 105, in wrapper
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher return f(_args, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/manager.py", line 127, in lvo_inner1
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher return lvo_inner2(inst, context, volume_id, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/openstack/common/lockutils.py", line 233, in inner
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher retval = f(_args, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/manager.py", line 126, in lvo_inner2
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher return f(__args, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/manager.py", line 455, in delete_volume
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher {'status': 'error_deleting'})
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/openstack/common/excutils.py", line 82, in exit
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher six.reraise(self.type, self.value, self.tb)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/manager.py", line 444, in delete_volume
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher self.driver.delete_volume(volume_ref)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/osprofiler/profiler.py", line 105, in wrapper
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher return f(_args, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/drivers/emc/emc_cli_iscsi.py", line 78, in delete_volume
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher self.cli.delete_volume(volume)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/drivers/emc/emc_vnx_cli.py", line 106, in inner
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher ret = func(self, _args, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/drivers/emc/emc_vnx_cli.py", line 1597, in delete_volume
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher self._client.delete_lun(volume['name'])
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/drivers/emc/emc_vnx_cli.py", line 106, in inner
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher ret = func(self, _args, *_kwargs)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/cinder/volume/drivers/emc/emc_vnx_cli.py", line 366, in delete_lun
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher raise EMCVnxCLICmdError(command_delete_lun, rc, out)
2015-02-11 13:54:35.915 1962 TRACE oslo.messaging.rpc.dispatcher EMCVnxCLICmdError: Bad or unexpected response from the storage volume backend API: EMCVnxCLICmdError : ['lun', '-destroy', '-name', u'volume-be841785-bfb5-49fe-bbca-5df735f0823b', '-forceDetach', '-o'](Return Code: 156) (Output: [u"Cannot unbind LUN because it's contained in a Storage Group\n"])

Hello EMC,

older releases of openstack (liberty, mitaka) had configuration parameter san_secondary_ip, which specified IP address to second SP (usually B).

This parameter is not available in openstack newton and ocata and now I am confused if I should configure parameter san_ip multiple times, or if san_ip accepts an array (IP addresses separated by comma).

Either way it would be nice to update documentation about configuring multiple SPs.

I followed this documentation: https://docs.openstack.org/ocata/config-reference/block-storage/drivers/emc-vnx-driver.html

Hi, i found out that if you have a FC group on the VNX already attached to your compute, New volumes will fail to attach to the VMs. This is because a host can only have one storage group associated with it in the VNX and Cinder is looking for (and will create) a group named the FQDN of the attaching compute. So, if you need to have non cinder LUNs attached to your compute you need to pre-create the storage group named the FQDN of your compute. When you do that cinder will see the group already created and associate new LUNs with it properly. It would be quite common for the compute to boot from FC in which case you need to create (or rename) that VNX storage group to make cinder work.

I hope this helps someone else who ran into same issue.

Hi, I have newton and I can add snapcopy=true in metadata for default like in liberty

This driver create a storage group same name of hypervisor, but if Openstack try to live migration a vm then get error, because the another hypervisor doesn't see the lun. I would like to request the option, that I set to storage group name instead of storage group per hypervisor.