openstack/openstack-ansible-os_neutron
Code Issues Proposed changes

Cap the maximum number of rpc workers

Browse Source 
Currently, there is only one process for all
rabbitmq processing. This is not enough and
causes systems to get clogged up.

Change-Id: Ie4823ed85582d3056c7837e3a8a4e7a53fac653c
This commit is contained in:
Guilherme Steinmüller 2019-06-10 20:22:21 +00:00
parent b045ba329a
commit 164cc66f7d
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
defaults/main.yml
View File

@ -264,7 +264,8 @@ neutron_oslomsg_amqp1_enabled: "{{ neutron_oslomsg_rpc_transport == 'amqp' }}"
neutron_rpc_thread_pool_size: 64
neutron_rpc_conn_pool_size: 30
neutron_rpc_response_timeout: 60
neutron_rpc_workers: 1
neutron_rpc_workers_max: 16
neutron_rpc_workers: "{{ [[ansible_processor_vcpus|default(2) // 2, 1] | max, neutron_rpc_workers_max] | min }}"
###
### Identity (Keystone) integration