[Openstack] Corrupted neutron database after havana to icehouse upgrade

Marco Marino marino.mrc at gmail.com
Fri May 8 09:39:54 UTC 2015


Hi, after an upgrade from havana to icehouse i think neutron database is
corrupted. I'm using neutron + gre with ml2 and openvswitch and i have 2
networks with the same segmentation id (the external shared network and a
private tenant network!)

This is the current state of the db:

mysql> select * from networks;
+----------------------------------+--------------------------------------+------------+--------+----------------+--------+
| tenant_id                        | id                                   |
name       | status | admin_state_up | shared |
+----------------------------------+--------------------------------------+------------+--------+----------------+--------+
| cc0f763bc9aa4ed89abef084580d45a7 | a4c9c004-b0f5-46d5-9e02-ee29ef5ca0b7 |
test-net   | ACTIVE |              1 |      0 |
| 71ec4fc643564bb5aa9f0650126856c7 | b03f34e6-988b-4164-8b3d-463cda887e67 |
cvg-on-net | ACTIVE |              1 |      0 |
| 71ec4fc643564bb5aa9f0650126856c7 | b5e35f6f-240f-4fd4-8684-3a3e35c18264 |
cvg-net    | ACTIVE |              1 |      0 |
| 71ec4fc643564bb5aa9f0650126856c7 | e2633c4e-e9ad-4162-9d6a-774b7acd50f3 |
Ext_Net    | ACTIVE |              1 |      1 |
+----------------------------------+--------------------------------------+------------+--------+----------------+--------+
4 rows in set (0.00 sec)

mysql> select * from ml2_network_segments;
+--------------------------------------+--------------------------------------+--------------+------------------+-----------------+
| id                                   | network_id
  | network_type | physical_network | segmentation_id |
+--------------------------------------+--------------------------------------+--------------+------------------+-----------------+
| 3d8bd871-f026-470e-8687-c808c41f11a6 |
21ee3bbe-def0-4150-b8c6-0249b07fab88 | gre          | NULL             |
            3 |
| a3a2c27f-02b5-4990-b5c5-583a85f9f0d9 |
e2633c4e-e9ad-4162-9d6a-774b7acd50f3 | gre          | NULL             |
            2 |
| 9cd69ce4-27c8-4178-b3c0-bda426821964 |
f56d46df-0c88-4e7b-9e1f-a0291891c96b | gre          | NULL             |
            1 |
| c88f996d-e838-462b-805b-abace9a6d67c |
f7b39864-b611-4118-af05-65c7c8cf13a9 | gre          | NULL             |
            1 |
| 1bf39cdd-ccd8-4c61-aef2-1598f8614fda |
b5e35f6f-240f-4fd4-8684-3a3e35c18264 | gre          | NULL             |
            2 |
| 3a0ff483-c0d5-45fc-884a-bd348971ce23 |
013422f3-5972-4b25-81ca-3bde55997bfe | gre          | NULL             |
            3 |
| 27a76433-ef5e-4bbd-9f31-37579a7372d2 |
b03f34e6-988b-4164-8b3d-463cda887e67 | gre          | NULL             |
            3 |
| eecdb5b6-cbf2-4663-8f4a-f99ba8b7f3dc |
a4c9c004-b0f5-46d5-9e02-ee29ef5ca0b7 | gre          | NULL             |
            1 |
+--------------------------------------+--------------------------------------+--------------+------------------+-----------------+
8 rows in set (0.00 sec)

mysql> select * from ml2_gre_allocations where gre_id < 7;
+--------+-----------+
| gre_id | allocated |
+--------+-----------+
|      1 |         0 |
|      2 |         1 |
|      3 |         1 |
|      4 |         0 |
|      5 |         0 |
|      6 |         0 |
+--------+-----------+
6 rows in set (0.00 sec)


First, note that networks are not deleted from the ml2_network_segments
table. How is this possible? Furthermore, i see that only gre_id 2 and 3
are allocated, but i have an active network ("test-net") with
segmentation_id = 1 Another important point is that the external (shared)
network "Ext_Net" has the same segmentation_id (=2) of the "cvg_net". Why?
When i create a new private network (eg: gre_id = 1), i see that in the
ml2_gre_allocations the flag "allocated" goes to 1 and the network works
(dhcp), but after few minutes the "allocated" flag goes to 0, the gre
tunnel disappear from the compute and the network node and hence the
network doesn't works.
How can i adjust this situation? Please help me. I don't want to drop the
entire neutron database because i don't want lose all security groups....
Thank you
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20150508/0f04475a/attachment.html>


More information about the Openstack mailing list