Skip to content

Commit c1849ae

Browse files
markgoddardmarkgoddard
authored
Merge pull request #1036 from stackhpc/document-cloudkitty
Add docs for configuring CloudKitty
2 parents b356c49 + a94ebf6 commit c1849ae

File tree

2 files changed

+142
-0
lines changed

2 files changed

+142
-0
lines changed

‎doc/source/configuration/cloudkitty.rst

Lines changed: 141 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,141 @@
1+
==========
2+
CloudKitty
3+
==========
4+
5+
Configuring in kayobe-config
6+
============================
7+
8+
By default, CloudKitty uses Gnocchi and Ceilometer as the collector and fetcher
9+
backends. Unless the system has a specific reason not to, we recommend instead
10+
using Prometheus as the backend for both. The following instructions explain
11+
how to do this. Also, see the `Kolla Ansible docs on CloudKitty
12+
<https://docs.openstack.org/kolla-ansible/latest/reference/rating/cloudkitty-guide.html>`__
13+
for more details.
14+
15+
Enable CloudKitty and disable InfluxDB, as we are using OpenSearch as the
16+
storage backend. Set the following in ``kolla.yml``:
17+
18+
.. code-block:: yaml
19+
20+
kolla_enable_cloudkitty: true
21+
# Explicitly disable influxdb as we are using OpenSearch as the CloudKitty backend
22+
kolla_enable_influxdb: false
23+
24+
Set Prometheus as the backend for both the collector and fetcher, and
25+
Elasticsearch as the storage backend. Note that our fork of CloudKitty is
26+
patched so that the CloudKitty Elasticsearch V2 storage backend will also work
27+
with an OpenSearch cluster. Proper support for the V2 OpenSearch storage
28+
backend is still pending in Kolla-Ansible `here
29+
<https://review.opendev.org/c/openstack/kolla-ansible/+/898555>`__. Set the
30+
following in ``kolla/globals.yml``:
31+
32+
.. code-block:: yaml
33+
34+
cloudkitty_collector_backend: prometheus
35+
cloudkitty_fetcher_backend: prometheus
36+
cloudkitty_storage_backend: elasticsearch
37+
38+
If you have TLS enabled, you will also need to set the cafile for Prometheus
39+
and Elasticsearch. Set the following in ``kolla/globals.yml``.
40+
41+
.. code-block::
42+
43+
{% raw %}
44+
cloudkitty_prometheus_cafile: "{{ openstack_cacert }}"
45+
cloudkitty_elasticsearch_cafile: "{{ openstack_cacert }}"
46+
{% endraw %}
47+
48+
The default collection period is one hour, which is likely too long for most
49+
systems as CloudKitty charges by the **entire** collection period if any usage
50+
is seen within this timeframe. This is regardless of actual usage, meaning that
51+
even one minute will be charged as a full hour's usage. As a result, it is
52+
recommended to adjust the collection interval, ``period`` (in units of
53+
seconds), appropriately (e.g. ten minutes). Furthermore, when using Prometheus
54+
as the collector, you need to change the ``scope_key`` to match the metrics
55+
provided by the Prometheus OpenStack Exporter. Both of these can be achieved by
56+
setting the following in ``kolla/config/cloudkitty.conf``:
57+
58+
.. code-block:: console
59+
60+
[collect]
61+
scope_key = tenant_id
62+
period = 600
63+
64+
You will need to configure which metrics CloudKitty should track. The following
65+
example, set in ``kolla/config/cloudkitty/metrics.yml``, will track for VM flavors and
66+
the total utilised volume.
67+
68+
.. code-block:: yaml
69+
70+
metrics:
71+
openstack_nova_server_status:
72+
alt_name: instance
73+
groupby:
74+
- uuid
75+
- user_id
76+
- tenant_id
77+
metadata:
78+
- flavor_id
79+
- name
80+
mutate: MAP
81+
mutate_map:
82+
0.0: 1.0 # ACTIVE
83+
11.0: 1.0 # SHUTOFF
84+
12.0: 1.0 # SUSPENDED
85+
16.0: 1.0 # PAUSED
86+
unit: instance
87+
openstack_cinder_limits_volume_used_gb:
88+
alt_name: storage
89+
unit: GiB
90+
groupby:
91+
- tenant_id
92+
93+
If your system had Monasca deployed in the past, you likely have some
94+
relabelled attributes in the Prometheus OpenStack exporter. To account for
95+
this, you should either remove the custom relabelling (in
96+
``kolla/config/prometheus.yml``) or change your ``metrics.yml`` to use the
97+
correct attributes.
98+
99+
Post-configuration with openstack-config
100+
========================================
101+
102+
This is an example `openstack-config
103+
<https://github.com/stackhpc/openstack-config>`__ setup to create mappings for
104+
the metrics configured above. Note that the costs are scaled for the ten minute
105+
collection period, e.g. a flavor with 1 VCPU will cost 1 unit per hour.
106+
107+
.. code-block:: yaml
108+
109+
# Map flavors based on VCPUs
110+
openstack_ratings_hashmap_field_mappings:
111+
- service: instance
112+
name: flavor_id
113+
mappings:
114+
- value: '1' # tiny compute flavor (1 vcpu) with an OpenStack flavor ID of 1
115+
cost: 0.1666666666666666
116+
type: flat
117+
- value: '2' # small compute flavor (2 vcpus) with an OpenStack flavor ID of 2
118+
cost: 0.3333333333333333
119+
type: flat
120+
- value: '3' # medium compute flavor (3 vcpus) with an OpenStack flavor ID of 3
121+
cost: 0.5
122+
type: flat
123+
- value: '4' # large compute flavor (4 vcpus) with an OpenStack flavor ID of 4
124+
cost: 0.6666666666666666
125+
type: flat
126+
- value: '5' # xlarge compute flavor (8 vcpus) with an OpenStack flavor ID of 5
127+
cost: 1.3333333333333333
128+
type: flat
129+
- value: '6' # tiny 2 compute flavor (2 vcpus) with an OpenStack flavor ID of 6
130+
cost: 0.3333333333333333
131+
type: flat
132+
133+
# Map volumes based on GB
134+
openstack_ratings_hashmap_service_mappings:
135+
- service: storage
136+
cost: 0.16666666666666666
137+
type: flat
138+
139+
See the `OpenStack CloudKitty Ratings role
140+
<https://github.com/stackhpc/ansible-collection-openstack/tree/main/roles/os_ratings>`__
141+
for more details.

‎doc/source/configuration/index.rst

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -20,3 +20,4 @@ the various features provided.
2020
magnum-capi
2121
ci-cd
2222
security-hardening
23+
cloudkitty

0 commit comments

Comments
 (0)