Delete attachment on remove_export failure

When deleting an attachment, if the driver's remove_export or the
detach_volume method call fails in the cinder driver, then the
attachment status is changed to error_detaching but the REST API call
doesn't fail.

The end result is:
- Volume status is "available"
- Volume attach_status is "detached"
- There is a volume_attachment record for the volume
- The volume may still be exported in the backend

The volume still being exported in the storage array is not a problem,
since the next attach-detach cycle will give it another opportunity for
it to succeed, and we also do the export on volume deletion.

So in the end leaving the attachment in error_detaching status doesn't
have any use and creates confusion.

This patch removes the attachment record when on an attachment delete
request if the error happens on remove_export or detach_volume calls.

This doesn't change how the REST API attachment delete operation
behaves, the change is that there will not be a leftover attachment
record with the volume in available and detached status.

Closes-Bug: #1935057
Change-Id: I442a42b0c098775935a799876ad8efbe141829ad
This commit is contained in:
Gorka Eguileor 2021-07-20 17:15:54 +02:00
parent 68d4944577
commit 3aa00b0878
3 changed files with 35 additions and 10 deletions

View File

@ -145,6 +145,31 @@ class AttachmentManagerTestCase(test.TestCase):
self.context,
attachment_ref.id)
def test_attachment_delete_remove_export_fail(self):
"""attachment_delete removes attachment on remove_export failure."""
self.mock_object(self.manager.driver, 'remove_export',
side_effect=Exception)
# Report that the connection is not shared
self.mock_object(self.manager, '_connection_terminate',
return_value=False)
vref = tests_utils.create_volume(self.context, status='in-use',
attach_status='attached')
values = {'volume_id': vref.id, 'volume_host': vref.host,
'attach_status': 'reserved', 'instance_uuid': fake.UUID1}
attach = db.volume_attach(self.context, values)
# Confirm the volume OVO has the attachment before the deletion
vref.refresh()
self.assertEqual(1, len(vref.volume_attachment))
self.manager.attachment_delete(self.context, attach.id, vref)
# Attachment has been removed from the DB
self.assertRaises(exception.VolumeAttachmentNotFound,
db.volume_attachment_get, self.context, attach.id)
# Attachment has been removed from the volume OVO attachment list
self.assertEqual(0, len(vref.volume_attachment))
def test_attachment_delete_multiple_attachments(self):
volume_params = {'status': 'available'}
vref = tests_utils.create_volume(self.context, **volume_params)

View File

@ -4949,16 +4949,10 @@ class VolumeManager(manager.CleanableManager,
if has_shared_connection is not None and not has_shared_connection:
self.driver.remove_export(context.elevated(), vref)
except Exception:
# FIXME(jdg): Obviously our volume object is going to need some
# changes to deal with multi-attach and figuring out how to
# represent a single failed attach out of multiple attachments
# TODO(jdg): object method here
attachment.attach_status = \
fields.VolumeAttachStatus.ERROR_DETACHING
attachment.save()
else:
vref.finish_detach(attachment_id)
# Failures on detach_volume and remove_export are not considered
# failures in terms of detaching the volume.
pass
vref.finish_detach(attachment_id)
self._notify_about_volume_usage(context, vref, "detach.end")
# Replication group API (Tiramisu)

View File

@ -0,0 +1,6 @@
---
fixes:
- |
`Bug #1935057 <https://bugs.launchpad.net/cinder/+bug/1935057>`_: Fixed
sometimes on a detach volume may end in available and detached yet have an
attachment in error_detaching.