volume extend is not catch exception when in some status

[zhangjinnan]

[stack@localhost devstack]$ cinder create --name test 1
+---------------------------------------+--------------------------------------+
| Property | Value |
+---------------------------------------+--------------------------------------+
| attachments | [] |
| availability_zone | nova |
| bootable | false |
| consistencygroup_id | None |
| created_at | 2015-09-26T03:01:30.000000 |
| description | None |
| encrypted | False |
| id | d8789ef9-a68f-46d9-be4b-32bb3b94c6a3 |
| metadata | {} |
| migration_status | None |
| multiattach | False |
| name | test |
| os-vol-host-attr:host | None |
| os-vol-mig-status-attr:migstat | None |
| os-vol-mig-status-attr:name_id | None |
| os-vol-tenant-attr:tenant_id | 1fb3aabb988e43efb1d1a9a5f89cf2da |
| os-volume-replication:driver_data | None |
| os-volume-replication:extended_status | None |
| replication_status | disabled |
| size | 1 |
| snapshot_id | None |
| source_volid | None |
| status | creating |
| user_id | 89ba6206a112484c96624fe5678f7409 |
| volume_type | lvmdriver-1 |
+---------------------------------------+--------------------------------------+
[stack@localhost devstack]$ cinder list
+--------------------------------------+-----------+------------------+------+------+-------------+----------+-------------+-------------+
| ID | Status | Migration Status | Name | Size | Volume Type | Bootable | Multiattach | Attached to |
+--------------------------------------+-----------+------------------+------+------+-------------+----------+-------------+-------------+
| d8789ef9-a68f-46d9-be4b-32bb3b94c6a3 | available | - | test | 1 | lvmdriver-1 | false | False | |
+--------------------------------------+-----------+------------------+------+------+-------------+----------+-------------+-------------+

[stack@localhost devstack]$ cinder reset-state --state error test
[stack@localhost devstack]$ cinder list
+--------------------------------------+--------+------------------+------+------+-------------+----------+-------------+-------------+
| ID | Status | Migration Status | Name | Size | Volume Type | Bootable | Multiattach | Attached to |
+--------------------------------------+--------+------------------+------+------+-------------+----------+-------------+-------------+
| d8789ef9-a68f-46d9-be4b-32bb3b94c6a3 | error | - | test | 1 | lvmdriver-1 | false | False | |
+--------------------------------------+--------+------------------+------+------+-------------+----------+-------------+-------------+

[stack@localhost devstack]$ cinder extend test 2
ERROR: Invalid volume: Volume d8789ef9-a68f-46d9-be4b-32bb3b94c6a3 status must be available to extend, but current status is: error. (HTTP 400) (Request-ID: req-f6674f0a-50b3-4e8b-888d-b31418255eca)

In this scene, cinder-api log is not any log or exception.

[Huang Zhiteng (zhiteng-huang)]

What's the expected state? From cinder client output, it looks fine to me.

[Gorka Eguileor (gorka)]

Cinder client output is fine, but in extend, unlike with other actions, this doesn't get logged in the API. For example migrate, retype and others log this exact same case (status is not valid) using LOG.error.

[Alex Meade (alex-meade)]

@Gorka Why would cinder-api log this using LOG.error? it's not a server error it was a user mistake. IMO the other operations are logging incorrectly. As an operator, I don't wanna have to go poking around to figure out why there are errors in my logs if it's because a user was attempting something incorrect. Also, I would note that this should be an HTTP 409 not a 400.

[Sean McGinnis (sean-mcginnis)]

Unassigning due to no activity for > 6 months.