Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

devstack Queens: Tacker, cannot delete VNF

Have a devstack queens installation of Tacker and associated services (tacker, mistral, heat,

cannot delete a vnf , vnf stays in PENDING_DELETE status

tacker vnf-list
+--------------------------------------+-------------------+---------------------------+----------------+--------------------------------------+--------------------------------------+
| id                                   | name              | mgmt_url                  | status         | vim_id                               | vnfd_id                              |
+--------------------------------------+-------------------+---------------------------+----------------+--------------------------------------+--------------------------------------+
| 44276401-3e4b-40b4-8c43-2d3e06d4d20f | sample-cirros-vnf | {"VDU1": "192.168.120.4"} | PENDING_DELETE | 9386c68d-02db-4e1e-9f58-bf67bdf865be | a8fa22e5-a5b5-4908-8d58-6c0a31fd6fe0 |
+--------------------------------------+-------------------+---------------------------+----------------+--------------------------------------+--------------------------------------+

then

tacker vnf-delete 44276401-3e4b-40b4-8c43-2d3e06d4d20f


Unable to delete the below vnf(s):
 Cannot delete 44276401-3e4b-40b4-8c43-2d3e06d4d20f: VNF 44276401-3e4b-40b4-8c43-2d3e06d4d20f could not be found

I did at some point have to delete a lot of uwsgi runaway processes which were consuming all CPU (and making mgmt via CLI or Horizon impossible). After doing this I could see my created VNFs , networks and images again. All OS services seem to be running OK (openstack service show tacker; show heat etc) or sudo systemctl status devstack@h-* or sudo systemctl status devstack@t* show status active (running)

Is there any workaround (I assume trying to delete the VM associated with VNF will just make things worse for tacker)

devstack Queens: Tacker, cannot delete VNF

Have a devstack queens installation of Tacker and associated services (tacker, mistral, heat,

cannot delete a vnf , vnf stays in PENDING_DELETE status

tacker vnf-list
+--------------------------------------+-------------------+---------------------------+----------------+--------------------------------------+--------------------------------------+
| id                                   | name              | mgmt_url                  | status         | vim_id                               | vnfd_id                              |
+--------------------------------------+-------------------+---------------------------+----------------+--------------------------------------+--------------------------------------+
| 44276401-3e4b-40b4-8c43-2d3e06d4d20f | sample-cirros-vnf | {"VDU1": "192.168.120.4"} | PENDING_DELETE | 9386c68d-02db-4e1e-9f58-bf67bdf865be | a8fa22e5-a5b5-4908-8d58-6c0a31fd6fe0 |
+--------------------------------------+-------------------+---------------------------+----------------+--------------------------------------+--------------------------------------+

then

tacker vnf-delete 44276401-3e4b-40b4-8c43-2d3e06d4d20f


Unable to delete the below vnf(s):
 Cannot delete 44276401-3e4b-40b4-8c43-2d3e06d4d20f: VNF 44276401-3e4b-40b4-8c43-2d3e06d4d20f could not be found

I did at some point have to delete a lot of uwsgi runaway processes which were consuming all CPU (and making mgmt via CLI or Horizon impossible). After doing this I could see my created VNFs , networks and images again. All OS services seem to be running OK (openstack service show tacker; show heat etc) or sudo systemctl status devstack@h-* or sudo systemctl status devstack@t* show status active (running)

Is there any workaround (I assume trying to delete the VM associated with VNF will just make things worse for tacker)

Some strange behaviour around Heat (as Tacker converts Tosca to Heat thought I might be able to delete the Heat Stack?)

openstack stack list

ERROR: <html><head> <title>503 Service Unavailable</title> </head><body> Service Unavailable

The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later.

Apache/2.4.18 (Ubuntu) Server at 192.168.122.67 Port 80
</body></html>

Re: Heat I notice that Horizon has no longer a page for orchestration via Heat (was in Project/Compute if I'm not mistaken). Is this page excluded once Tacker is installed (i.e orchestration MUST happen at higher MANO layer)?