Container provider refuses to delete


#1

I have attempted to remove a container provider but it just refuses to delete.

  1. select provider
  2. Configuration -> Remove Container Provider
  3. Delete initiated for 1 Containers Providers from the CFME Database

But then the provider never goes away. It just stays there.


#2

I am also having the same issue for a Cloud Provider (OpenStack)


#3

After clearing out 400000 messages in the miq_queue i can now see that the delete operation is just stuck in a dequeue state.

vmdb_produciton=# select id,priority,method_name,state,queue_name,class_name,server_guid,zone,role,msg_timeout from miq_queue where method_name='destroy';
       id       | priority | method_name |  state  | queue_name |              class_name               | server_guid | zone | role | msg_timeout 
----------------+----------+-------------+---------+------------+---------------------------------------+-------------+------+------+-------------
 10000072312312 |      100 | destroy     | dequeue | generic    | ManageIQ::Providers::ContainerManager |             | OSP  |      |        3600
 10000072312352 |      100 | destroy     | dequeue | generic    | ManageIQ::Providers::CloudManager     |             | OSP  |      |        3600

#4

@itewk

Thanks for you posting. I ran into similar issue, not able to remove VMWare providers.

Following is how I clear out all the rows in miq_queue table. I am not sure what kind of impact will do to the current miq instance. But anyway I am giving up my current all-in-one approach and switch to cluster approach with Postgres DB HA.

[root@miq01 vmdb]# bin/rails db
Password: pgpassword
psql (9.5.4)
Type "help" for help.
vmdb_production=# SELECT count(*) from miq_queue;                               
count
--------
 176166
(1 row)
vmdb_production=#  DELETE FROM miq_queue;
DELETE 176166
vmdb_production=# \q