Control amount of metricd workers
As of today we didn't manage amount of gnocchi-metricd that equal to amount of CPU on host. So things can go off regarding CPU and memory consumption. For better control of ressources we add a variable to control the number of workers. Change-Id: Idcaec649a4de03f5714c61fda694ada45da41dbc
This commit is contained in:
parent
c845819cae
commit
2be06894c8
|
@ -139,6 +139,9 @@ gnocchi_wsgi_threads: 1
|
|||
gnocchi_wsgi_processes_max: 16
|
||||
gnocchi_wsgi_processes: "{{ [[(ansible_facts['processor_vcpus']//ansible_facts['processor_threads_per_core'])|default(1), 1] | max * 2, gnocchi_wsgi_processes_max] | min }}"
|
||||
|
||||
gnocchi_metricd_workers_max: 16
|
||||
gnocchi_metricd_workers: "{{ [[(ansible_facts['processor_vcpus']//ansible_facts['processor_threads_per_core'])|default(1), 1] | max * 2, gnocchi_metricd_workers_max] | min }}"
|
||||
|
||||
gnocchi_uwsgi_conf_overrides: {}
|
||||
gnocchi_api_init_overrides: {}
|
||||
gnocchi_metricd_init_overrides: {}
|
||||
|
|
|
@ -0,0 +1,7 @@
|
|||
---
|
||||
features:
|
||||
- |
|
||||
Implemented variable ``gnocchi_metricd_workers`` that is designed to
|
||||
controll amount of gnocchi-metricd workers spawned. By default it is
|
||||
equal to number of CPU cores, but no more than 16 workers.
|
||||
|
|
@ -124,7 +124,7 @@ redis_url = {{ gnocchi_incoming_redis_url }}
|
|||
# Number of workers for Gnocchi metric daemons. By default the available number
|
||||
# of CPU is used. (integer value)
|
||||
# Minimum value: 1
|
||||
#workers = 1
|
||||
workers = {{ gnocchi_metricd_workers }}
|
||||
|
||||
[statsd]
|
||||
# The listen IP for statsd (string value)
|
||||
|
|
Loading…
Reference in New Issue