yahoo-eng-team team mailing list archive
-
yahoo-eng-team team
-
Mailing list archive
-
Message #48873
[Bug 1562842] Re: instance going to scheduler when invalid v4-fixed-ip is provided in nova boot command
@Abhilash:
root cause
----------
The nova conductor log file "n-cond.log" you attached shows the error:
Invalid input received:
Fixed IP 1.0.0.255 is not a valid ip address for network
97adf977-8b62-4996-a800-bbbdaf9c0fd9.
This gets raised in Liberty code at [1]. It's also there in Mitaka and
current master (Newton).
As you already mentioned, this happens after the scheduler chose one
compute host and the spawn of the instance has started.
conclusion
----------
The CLI doesn't offer any client side validation. It returns the server
side created response. Because the spawn of an instance is a long
running task, the "launch" REST API triggers the creation asynchronously.
This means the CLI only returns that the trigger of the creation got
accepted. It doesn't make a statement if this will succeed.
I don't see a benefit in introducing a "pre-validation" before the
scheduler decides where to place an instance. I'm closing the bug report
as "Opinion/wishlist" as it doesn't look like a bug to me. If you think
this is wrong, please reopen it and add a reasoning.
References:
[1] https://github.com/openstack/nova/blob/acb2dc5e27a85b9148599f1c4dd59e317752f125/nova/network/neutronv2/api.py#L367-L370
** Changed in: nova
Status: Incomplete => Opinion
** Changed in: nova
Importance: Undecided => Wishlist
** Tags added: network neutron scheduler
--
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1562842
Title:
instance going to scheduler when invalid v4-fixed-ip is provided in
nova boot command
Status in OpenStack Compute (nova):
Opinion
Bug description:
When providing invalid v4-fixed-ip in nova boot command, the instance
will queue with the scheduler but fail to boot. Instead what should
happen is, there should be check at client side or server side and
error message should be thrown so that VM does not get queued with the
scheduler.
neutron net-list
+--------------------------------------+---------+----------------------------------------------------------+
| id | name | subnets |
+--------------------------------------+---------+----------------------------------------------------------+
| 97adf977-8b62-4996-a800-bbbdaf9c0fd9 | public | 35364205-3b7d-47db-8bf3-e1590416d9f1 2001:db8::/64 |
| | | ddacd805-5e07-4e63-a827-fd094c61e84a 172.24.4.0/24 |
| f9976ad3-33fb-44a4-b25f-e8356de9e7d2 | private | 6427f306-2646-479e-b47b-5a115f020d1c 10.0.0.0/24 |
| | | 4de2d8be-0c1a-4b9c-aabf-3a896a7c67c0 fd07:4910:1fa6::/64 |
+--------------------------------------+---------+----------------------------------------------------------+
nova boot abi_1 --image a258964a-4250-4b6d-9fe9-492ac2b3d8da --flavor m1.tiny --nic net-id=97adf977-8b62-4996-a800-bbbdaf9c0fd9,v6-fixed-ip=1.0.0.255
(note: 1.0.0.255 is broadcast ip for the network, should be invalid.)
+--------------------------------------+----------------------------------------------------------------+
| Property | Value |
+--------------------------------------+----------------------------------------------------------------+
| OS-DCF:diskConfig | MANUAL |
| OS-EXT-AZ:availability_zone | |
| OS-EXT-SRV-ATTR:host | - |
| OS-EXT-SRV-ATTR:hostname | abi-1 |
| OS-EXT-SRV-ATTR:hypervisor_hostname | - |
| OS-EXT-SRV-ATTR:instance_name | instance-0000004e |
| OS-EXT-SRV-ATTR:kernel_id | e2a5fdda-7422-4e41-b59a-fa2dad82583e |
| OS-EXT-SRV-ATTR:launch_index | 0 |
| OS-EXT-SRV-ATTR:ramdisk_id | 4ea6bc99-2c01-459b-9ce1-b6bc6a51ad79 |
| OS-EXT-SRV-ATTR:reservation_id | r-wt958hij |
| OS-EXT-SRV-ATTR:root_device_name | - |
| OS-EXT-SRV-ATTR:user_data | - |
| OS-EXT-STS:power_state | 0 |
| OS-EXT-STS:task_state | scheduling |
| OS-EXT-STS:vm_state | building |
| OS-SRV-USG:launched_at | - |
| OS-SRV-USG:terminated_at | - |
| accessIPv4 | |
| accessIPv6 | |
| adminPass | FtPf4XB3exjh |
| config_drive | |
| created | 2016-03-28T12:33:13Z |
| flavor | m1.tiny (1) |
| hostId | |
| id | fb037e34-7372-45ee-88b6-7dc1d228fe46 |
| image | cirros-0.3.4-x86_64-uec (a258964a-4250-4b6d-9fe9-492ac2b3d8da) |
| key_name | - |
| metadata | {} |
| name | abi_1 |
| os-extended-volumes:volumes_attached | [] |
| progress | 0 |
| security_groups | default |
| status | BUILD |
| tenant_id | 9f87ed00b37d434abd0e422d9f0b9a9b |
| updated | 2016-03-28T12:33:13Z |
| user_id | 19250807be0c4c42b970a02ccf3126d2 |
+--------------------------------------+----------------------------------------------------------------+
nova list
+--------------------------------------+------------+--------+------------+-------------+----------------------------------+
| ID | Name | Status | Task State | Power State | Networks |
+--------------------------------------+------------+--------+------------+-------------+----------------------------------+
| 5fefe57f-77cb-4779-9727-fa7d82cf857b | abhilash_1 | ACTIVE | - | Running | public=2001:db8::2d, 172.24.4.44 |
| 28bd1ecf-393d-4073-9f00-62c07fe2ad49 | abi_2 | ERROR | - | NOSTATE | |
+--------------------------------------+------------+--------+------------+-------------+----------------------------------+
This abi_2 is in error state, which should instead through message at boot command only.
To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1562842/+subscriptions
References
