p39718465
/
zabbix
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '93958d0fb0'
${ noResults }
zabbix/templates/cloud/openstack
History
yzl 93958d0fb0
zabbix6.0 		1 year ago
..
README.md zabbix6.0 1 year ago
template_cloud_openstack_http.yaml zabbix6.0 1 year ago

README.md
Escape

OpenStack Nova by HTTP

Overview

This template is designed for the effortless deployment of OpenStack Nova monitoring by Zabbix via HTTP and doesn't require any external scripts.

Requirements

Zabbix version: 7.0 and higher.

Tested versions

This template has been tested on:

  • OpenStack Yoga release and OpenStack built from sources (27568ea3):

    • Compute API v2.1

Configuration

Zabbix should be configured according to the instructions in the Templates out of the box section.

Setup

This template is not meant to be used independently. A host with the OpenStack by HTTP template will discover the Nova service automatically and create a host prototype with this template assigned to it.

If needed, you can specify an HTTP proxy for the template to use by changing the value of {$OPENSTACK.NOVA.HTTP.PROXY} user macro.

For tenant usage statistics, it is possible to choose a custom time period for which the data will be queried. This can be set with the {$OPENSTACK.NOVA.TENANT.PERIOD} macro value. The value can be one of the following:

  • y - current year until now;

  • m - current month until now (default value);

  • w - current week until now;

  • d - current day until now;

This template discovers servers (instances) present in the project and monitors their statuses, but, depending on different use cases, most likely it is not necessary to monitor all servers. To filter which servers to monitor, set the {$OPENSTACK.SERVER.DISCOVERY.NAME.MATCHES} and {$OPENSTACK.SERVER.DISCOVERY.NAME.NOT_MATCHES} macro values accordingly. This logic also applies to other low-level discovery rules.

OpenStack configuration

For the OpenStack monitoring user to be able to access the API resources used in this template, it is needed to configure the policy file for OpenStack Nova.

On the OpenStack server, open the /etc/nova/policy.json file in your favorite text editor.

In this file, assign the following target resources to the role that the monitoring user uses:

{
  "os_compute_api:servers:index": "role:monitoring",
  "os_compute_api:servers:show": "role:monitoring",
  "os_compute_api:os-services:list": "role:monitoring",
  "os_compute_api:os-hypervisors:list-detail": "role:monitoring",
  "os_compute_api:os-availability-zone:detail": "role:monitoring",
  "os_compute_api:os-simple-tenant-usage:list": "role:monitoring"
}

If some role is already assigned to the target, it is possible to add another role with or, for example, role:firstRole or role:monitoring.

Note that a restart of OpenStack Nova services might be needed for these new changes to be applied.

Macros used

Name Description Default
{$OPENSTACK.NOVA.SERVICE.URL}

API endpoint for Nova Service, e.g., https://local.openstack:8774/v2.1.
{$OPENSTACK.TOKEN}

API token for the monitoring user.
{$OPENSTACK.HTTP.PROXY}

Sets the HTTP proxy for script and HTTP agent items. If this parameter is empty, then no proxy is used.
{$OPENSTACK.NOVA.TENANT.PERIOD}

Period for which tenant usage statistics will be queried. Possible values are:

'y' - current year until now,

'm' - current month until now,

'w' - current week until now,

'd' - current day until now.

 m
{$OPENSTACK.NOVA.INTERVAL.LIMITS}

Interval for absolute limit HTTP agent item query.

 3m
{$OPENSTACK.NOVA.INTERVAL.SERVERS}

Interval for server HTTP agent item queries.

 3m
{$OPENSTACK.NOVA.INTERVAL.SERVICES}

Interval for service HTTP agent item query.

 3m
{$OPENSTACK.NOVA.INTERVAL.HYPERVISOR}

Interval for hypervisor HTTP agent item query.

 3m
{$OPENSTACK.NOVA.INTERVAL.AVAILABILITY_ZONE}

Interval for availability zone HTTP agent item query.

 3m
{$OPENSTACK.NOVA.INTERVAL.TENANTS}

Interval for tenant HTTP agent item query.

 3m
{$OPENSTACK.NOVA.INSTANCES.UTIL.WARN}

Sets the percentage threshold for creating a warning severity event about instances resource count.

 75
{$OPENSTACK.NOVA.INSTANCES.UTIL.HIGH}

Sets the percentage threshold for creating a high severity event about instances resource count.

 90
{$OPENSTACK.NOVA.CPU.UTIL.WARN}

Sets the percentage threshold for creating a warning severity event about vCPU resource usage.

 75
{$OPENSTACK.NOVA.CPU.UTIL.HIGH}

Sets the percentage threshold for creating a high severity event about vCPU resource usage.

 90
{$OPENSTACK.NOVA.RAM.UTIL.WARN}

Sets the percentage threshold for creating a warning severity event about RAM resource usage.

 75
{$OPENSTACK.NOVA.RAM.UTIL.HIGH}

Sets the percentage threshold for creating a high severity event about RAM resource usage.

 90
{$OPENSTACK.SERVER.DISCOVERY.NAME.MATCHES}

Sets the server name regex filter to use in server discovery for including.

 .*
{$OPENSTACK.SERVER.DISCOVERY.NAME.NOT_MATCHES}

Sets the server name regex filter to use in server discovery for excluding.

 CHANGE_IF_NEEDED
{$OPENSTACK.SERVICES.DISCOVERY.HOST.MATCHES}

Sets the host name regex filter to use in compute services discovery for including.

 .*
{$OPENSTACK.SERVICES.DISCOVERY.HOST.NOT_MATCHES}

Sets the host name regex filter to use in compute services discovery for excluding.

 CHANGE_IF_NEEDED
{$OPENSTACK.SERVICES.DISCOVERY.BINARY.MATCHES}

Sets the binary name regex filter to use in compute services discovery for including.

 .*
{$OPENSTACK.SERVICES.DISCOVERY.BINARY.NOT_MATCHES}

Sets the binary name regex filter to use in compute services discovery for excluding.

 CHANGE_IF_NEEDED
{$OPENSTACK.HYPERVISOR.DISCOVERY.HOSTNAME.MATCHES}

Sets the hostname regex filter to use in hypervisor discovery for including.

 .*
{$OPENSTACK.HYPERVISOR.DISCOVERY.HOSTNAME.NOT_MATCHES}

Sets the hostname regex filter to use in hypervisor discovery for excluding.

 CHANGE_IF_NEEDED
{$OPENSTACK.HYPERVISOR.DISCOVERY.TYPE.MATCHES}

Sets the type regex filter to use in hypervisor discovery for including.

 .*
{$OPENSTACK.HYPERVISOR.DISCOVERY.TYPE.NOT_MATCHES}

Sets the type regex filter to use in hypervisor discovery for excluding.

 CHANGE_IF_NEEDED
{$OPENSTACK.HYPERVISOR.DISCOVERY.IP.MATCHES}

Sets the host IP address regex filter to use in hypervisor discovery for including.

 .*
{$OPENSTACK.HYPERVISOR.DISCOVERY.IP.NOT_MATCHES}

Sets the host IP address regex filter to use in hypervisor discovery for excluding.

 CHANGE_IF_NEEDED
{$OPENSTACK.AVAILABILITY_ZONE.DISCOVERY.NAME.MATCHES}

Sets the zone name regex filter to use in availability zone discovery for including.

 .*
{$OPENSTACK.AVAILABILITY_ZONE.DISCOVERY.NAME.NOT_MATCHES}

Sets the zone name regex filter to use in availability zone discovery for excluding.

 CHANGE_IF_NEEDED

Items

Name Description Type Key and additional info
Nova: Get absolute limits

Gets absolute limits for the project.

 HTTP agent openstack.nova.limits.get

Preprocessing

  • JSON Path: $.limits.absolute

    ⛔️Custom on fail: Set error to: Could not get absolute project limits
Nova: Get servers

Gets a list of servers.

 HTTP agent openstack.nova.servers.get

Preprocessing

  • JSON Path: $.servers

    ⛔️Custom on fail: Set error to: Could not get servers list
Nova: Get compute services

Gets a list of compute services and its data.

 HTTP agent openstack.nova.services.get

Preprocessing

  • JSON Path: $.services

    ⛔️Custom on fail: Set error to: Could not get compute services list
Nova: Get hypervisors

Gets a list of hypervisors and its data.

 HTTP agent openstack.nova.hypervisors.get

Preprocessing

  • JSON Path: $.hypervisors

    ⛔️Custom on fail: Set error to: Could not get hypervisors list
Nova: Get availability zones

Gets a list of availability zones and its data.

 HTTP agent openstack.nova.availability_zone.get

Preprocessing

  • JSON Path: $.availabilityZoneInfo

    ⛔️Custom on fail: Set error to: Could not get availability zones list
Nova: Get tenants

Gets a list of tenants and its data.

 Script openstack.nova.tenant.get

Preprocessing

  • JSON Path: $.tenant_usages

    ⛔️Custom on fail: Set error to: Could not get tenant list
Nova: Instances count, current

Number of servers in each tenant.

 Dependent item openstack.nova.limits.instances.current

Preprocessing

  • JSON Path: $.totalInstancesUsed
Nova: Instances count, max

Number of allowed servers for each tenant.

 Dependent item openstack.nova.limits.instances.max

Preprocessing

  • JSON Path: $.maxTotalInstances
Nova: Instances count, free

Number of available servers for each tenant.

 Calculated openstack.nova.limits.instances.free

Preprocessing

  • Discard unchanged with heartbeat: 1h
Nova: vCPUs usage, current

Number of used server cores in each tenant.

 Dependent item openstack.nova.limits.vcpu.current

Preprocessing

  • JSON Path: $.totalCoresUsed
Nova: vCPUs usage, max

Number of allowed server cores for each tenant.

 Dependent item openstack.nova.limits.vcpu.max

Preprocessing

  • JSON Path: $.maxTotalCores
Nova: vCPUs usage, free

Number of available server cores for each tenant.

 Calculated openstack.nova.limits.vcpu.free

Preprocessing

  • Discard unchanged with heartbeat: 1h
Nova: RAM usage, current

Amount of used server RAM.

 Dependent item openstack.nova.limits.ram.current

Preprocessing

  • JSON Path: $.totalRAMUsed

  • Custom multiplier: 1048576
Nova: RAM usage, max

Amount of allowed server RAM.

 Dependent item openstack.nova.limits.ram.max

Preprocessing

  • JSON Path: $.maxTotalRAMSize

  • Custom multiplier: 1048576
Nova: RAM usage, free

Amount of available server RAM.

 Calculated openstack.nova.limits.ram.free

Preprocessing

  • Discard unchanged with heartbeat: 1h

Triggers

Name Description Expression Severity Dependencies and additional info
Nova: Current instances count is too high

Current instances count has exceeded {$OPENSTACK.NOVA.INSTANCES.UTIL.HIGH}% of the max available instances count.

 last(/OpenStack Nova by HTTP/openstack.nova.limits.instances.current) >= ({$OPENSTACK.NOVA.INSTANCES.UTIL.HIGH} / 100 * last(/OpenStack Nova by HTTP/openstack.nova.limits.instances.max)) High
Nova: Current instances count is high

Current instances count has exceeded {$OPENSTACK.NOVA.INSTANCES.UTIL.WARN}% of the max available instances count.

 last(/OpenStack Nova by HTTP/openstack.nova.limits.instances.current) >= ({$OPENSTACK.NOVA.INSTANCES.UTIL.WARN} / 100 * last(/OpenStack Nova by HTTP/openstack.nova.limits.instances.max)) Warning Depends on:
  • Nova: Current instances count is too high
Nova: Current vCPU usage is too high

Current vCPU usage has exceeded {$OPENSTACK.NOVA.CPU.UTIL.HIGH}% of the max available vCPU usage.

 last(/OpenStack Nova by HTTP/openstack.nova.limits.vcpu.current) >= ({$OPENSTACK.NOVA.CPU.UTIL.HIGH} / 100 * last(/OpenStack Nova by HTTP/openstack.nova.limits.vcpu.max)) High
Nova: Current vCPU usage is high

Current vCPU usage has exceeded {$OPENSTACK.NOVA.CPU.UTIL.WARN}% of the max available vCPU usage.

 last(/OpenStack Nova by HTTP/openstack.nova.limits.vcpu.current) >= ({$OPENSTACK.NOVA.CPU.UTIL.WARN} / 100 * last(/OpenStack Nova by HTTP/openstack.nova.limits.vcpu.max)) Warning Depends on:
  • Nova: Current vCPU usage is too high
Nova: Current RAM usage is too high

Current RAM usage has exceeded {$OPENSTACK.NOVA.RAM.UTIL.HIGH}% of the max available RAM usage.

 last(/OpenStack Nova by HTTP/openstack.nova.limits.ram.current) >= ({$OPENSTACK.NOVA.RAM.UTIL.HIGH} / 100 * last(/OpenStack Nova by HTTP/openstack.nova.limits.ram.max)) High
Nova: Current RAM usage is high

Current RAM usage has exceeded {$OPENSTACK.NOVA.RAM.UTIL.WARN}% of the max available RAM usage.

 last(/OpenStack Nova by HTTP/openstack.nova.limits.ram.current) >= ({$OPENSTACK.NOVA.RAM.UTIL.WARN} / 100 * last(/OpenStack Nova by HTTP/openstack.nova.limits.ram.max)) Warning Depends on:
  • Nova: Current RAM usage is too high

LLD rule Nova: Servers discovery

Name Description Type Key and additional info
Nova: Servers discovery

Discovers OpenStack Nova servers.

 Dependent item openstack.nova.server.discovery

Preprocessing

  • Discard unchanged with heartbeat: 1h

Item prototypes for Nova: Servers discovery

Name Description Type Key and additional info
Server [{#SERVER_ID}]:[{#SERVER_NAME}]: Status

Server status.

 HTTP agent openstack.nova.server.status.get[{#SERVER_ID}]

Preprocessing

  • JSON Path: $.server.status

    ⛔️Custom on fail: Set error to: Could not parse the detailed server report

  • JavaScript: The text is too long. Please see the template.

  • Discard unchanged with heartbeat: 1h

Trigger prototypes for Nova: Servers discovery

Name Description Expression Severity Dependencies and additional info
Server [{#SERVER_ID}]:[{#SERVER_NAME}]: Status is "ERROR"

Server is in "ERROR" status.

 last(/OpenStack Nova by HTTP/openstack.nova.server.status.get[{#SERVER_ID}])=5 High Manual close: Yes
Server [{#SERVER_ID}]:[{#SERVER_NAME}]: Status has changed

Status of the server has changed. Acknowledge to close the problem manually.

 last(/OpenStack Nova by HTTP/openstack.nova.server.status.get[{#SERVER_ID}])<>last(/OpenStack Nova by HTTP/openstack.nova.server.status.get[{#SERVER_ID}],#2) and length(last(/OpenStack Nova by HTTP/openstack.nova.server.status.get[{#SERVER_ID}]))>0 Info Manual close: Yes
Depends on:
  • Server [{#SERVER_ID}]:[{#SERVER_NAME}]: Status is "ERROR"

LLD rule Nova: Compute services discovery

Name Description Type Key and additional info
Nova: Compute services discovery

Discovers OpenStack compute services.

 Dependent item openstack.nova.services.discovery

Preprocessing

  • Discard unchanged with heartbeat: 1h

Item prototypes for Nova: Compute services discovery

Name Description Type Key and additional info
Compute service [{#HOST}]:[{#BINARY}]:[{#ID}]: Raw data

Raw data of the service.

 Dependent item openstack.nova.services.raw[{#ID}]

Preprocessing

  • JSON Path: $[?(@.id == "{#ID}")].first()

    ⛔️Custom on fail: Set error to: Could not parse the detailed services report
Compute service [{#HOST}]:[{#BINARY}]:[{#ID}]: State

State of the service.

 Dependent item openstack.nova.services.state[{#ID}]

Preprocessing

  • JSON Path: $.state

    ⛔️Custom on fail: Set error to: Could not parse the detailed services report

  • JavaScript: The text is too long. Please see the template.

  • Discard unchanged with heartbeat: 1h
Compute service [{#HOST}]:[{#BINARY}]:[{#ID}]: Status

Status of the service.

 Dependent item openstack.nova.services.status[{#ID}]

Preprocessing

  • JSON Path: $.status

    ⛔️Custom on fail: Set error to: Could not parse the detailed services report

  • JavaScript: The text is too long. Please see the template.

  • Discard unchanged with heartbeat: 1h
Compute service [{#HOST}]:[{#BINARY}]:[{#ID}]: Disabling reason

Reason for disabling a service.

 Dependent item openstack.nova.services.disabled.reason[{#ID}]

Preprocessing

  • JSON Path: $.disabled_reason

    ⛔️Custom on fail: Set error to: Could not parse the detailed services report

  • Discard unchanged with heartbeat: 1h

Trigger prototypes for Nova: Compute services discovery

Name Description Expression Severity Dependencies and additional info
Compute service [{#HOST}]:[{#BINARY}]:[{#ID}]: State is "down"

State of the service is "down".

 last(/OpenStack Nova by HTTP/openstack.nova.services.state[{#ID}])=0 Warning Manual close: Yes
Compute service [{#HOST}]:[{#BINARY}]:[{#ID}]: Status is "disabled"

Status of the server is disabled. Acknowledge to close the problem manually.

 last(/OpenStack Nova by HTTP/openstack.nova.services.status[{#ID}])=0 and length(last(/OpenStack Nova by HTTP/openstack.nova.services.disabled.reason[{#ID}]))>=0 Info Manual close: Yes

LLD rule Nova: Hypervisor discovery

Name Description Type Key and additional info
Nova: Hypervisor discovery

Discovers OpenStack Nova hypervisors.

 Dependent item openstack.nova.hypervisors.discovery

Preprocessing

  • Discard unchanged with heartbeat: 1h

Item prototypes for Nova: Hypervisor discovery

Name Description Type Key and additional info
Hypervisor [{#ID}]:[{#HOSTNAME}]: Raw data

Raw data of the hypervisor.

 Dependent item openstack.nova.hypervisors.raw[{#ID}]

Preprocessing

  • JSON Path: $[?(@.id == "{#ID}")].first()

    ⛔️Custom on fail: Set error to: Could not parse the detailed hypervisor report
Hypervisor [{#ID}]:[{#HOSTNAME}]: State

State of the hypervisor.

 Dependent item openstack.nova.hypervisors.state[{#ID}]

Preprocessing

  • JSON Path: $.state

    ⛔️Custom on fail: Set error to: Could not parse the detailed hypervisor report

  • JavaScript: The text is too long. Please see the template.

  • Discard unchanged with heartbeat: 1h
Hypervisor [{#ID}]:[{#HOSTNAME}]: Status

Status of the hypervisor.

 Dependent item openstack.nova.hypervisors.status[{#ID}]

Preprocessing

  • JSON Path: $.status

    ⛔️Custom on fail: Set error to: Could not parse the detailed hypervisor report

  • JavaScript: The text is too long. Please see the template.

  • Discard unchanged with heartbeat: 1h
Hypervisor [{#ID}]:[{#HOSTNAME}]: Version

Hypervisor version.

 Dependent item openstack.nova.hypervisors.version[{#ID}]

Preprocessing

  • JSON Path: $.hypervisor_version

    ⛔️Custom on fail: Set error to: Could not parse the detailed hypervisor report

  • Discard unchanged with heartbeat: 1h

Trigger prototypes for Nova: Hypervisor discovery

Name Description Expression Severity Dependencies and additional info
Hypervisor [{#ID}]:[{#HOSTNAME}]: State is "down"

State of the hypervisor is "down".

 last(/OpenStack Nova by HTTP/openstack.nova.hypervisors.state[{#ID}])=0 Warning Manual close: Yes
Hypervisor [{#ID}]:[{#HOSTNAME}]: Status is "disabled"

Status of the hypervisor is disabled.

 last(/OpenStack Nova by HTTP/openstack.nova.hypervisors.status[{#ID}])=0 Info Manual close: Yes
Hypervisor [{#ID}]:[{#HOSTNAME}]: Version has changed

Version of the hypervisor has changed. Acknowledge to close the problem manually.

 last(/OpenStack Nova by HTTP/openstack.nova.hypervisors.version[{#ID}])<>last(/OpenStack Nova by HTTP/openstack.nova.hypervisors.version[{#ID}],#2) and length(last(/OpenStack Nova by HTTP/openstack.nova.hypervisors.version[{#ID}]))>0 Info Manual close: Yes

LLD rule Nova: Availability zones discovery

Name Description Type Key and additional info
Nova: Availability zones discovery

Discovers OpenStack Nova availability zones.

 Dependent item openstack.nova.availability_zone.discovery

Preprocessing

  • Discard unchanged with heartbeat: 1h

Item prototypes for Nova: Availability zones discovery

Name Description Type Key and additional info
Availability zone [{#ZONE_NAME}]: Raw data

Raw data of the availability zone.

 Dependent item openstack.nova.availability_zone.raw[{#ZONE_NAME}]

Preprocessing

  • JSON Path: $[?(@.zoneName == "{#ZONE_NAME}")].first()

    ⛔️Custom on fail: Set error to: Could not parse the detailed availability zone report
Availability zone [{#ZONE_NAME}]: State

Current state of the availability zone.

 Dependent item openstack.nova.availability_zone.state[{#ZONE_NAME}]

Preprocessing

  • JSON Path: $.zoneState.available

    ⛔️Custom on fail: Set error to: Could not parse the detailed availability zone report

  • Boolean to decimal

  • Discard unchanged with heartbeat: 1h
Availability zone [{#ZONE_NAME}]: Host count

Count of hosts and service objects under single availability zone.

 Dependent item openstack.nova.availability_zone.host_count[{#ZONE_NAME}]

Preprocessing

  • JSON Path: $['hosts'].[*].[*].length()

    ⛔️Custom on fail: Set error to: Could not parse the detailed availability zone report

  • Discard unchanged with heartbeat: 1h

Trigger prototypes for Nova: Availability zones discovery

Name Description Expression Severity Dependencies and additional info
Availability zone [{#ZONE_NAME}]: Zone is unavailable

Availability zone is not available.

 last(/OpenStack Nova by HTTP/openstack.nova.availability_zone.state[{#ZONE_NAME}])=0 Warning Manual close: Yes

LLD rule Nova: Tenant discovery

Name Description Type Key and additional info
Nova: Tenant discovery

Discovers tenants and their usage data.

 Dependent item openstack.nova.tenant.discovery

Preprocessing

  • Discard unchanged with heartbeat: 1h

Item prototypes for Nova: Tenant discovery

Name Description Type Key and additional info
Tenant [{#TENANT_ID}]: Raw data

Raw data of the tenant.

 Dependent item openstack.nova.tenant.raw[{#TENANT_ID}]

Preprocessing

  • JSON Path: $[?(@.tenant_id == "{#TENANT_ID}")].first()

    ⛔️Custom on fail: Set error to: Could not parse the tenant report
Tenant [{#TENANT_ID}]: Total hours

Total duration that the servers exist (in hours).

 Dependent item openstack.nova.tenant.total_hours[{#TENANT_ID}]

Preprocessing

  • JSON Path: $.total_hours

    ⛔️Custom on fail: Set error to: Could not parse the detailed tenant report

  • JavaScript: return Math.round(value * 100) / 100;

  • Discard unchanged with heartbeat: 1h
Tenant [{#TENANT_ID}]: Total vCPUs usage

Total vCPU usage hours for the current tenant (project).

Multiplying the number of virtual CPUs of the server by hours the server exists, and then adding that all together for each server.

 Dependent item openstack.nova.tenant.total_vcpu[{#TENANT_ID}]

Preprocessing

  • JSON Path: $.total_vcpus_usage

    ⛔️Custom on fail: Set error to: Could not parse the detailed tenant report

  • JavaScript: return Math.round(value * 100) / 100;

  • Discard unchanged with heartbeat: 1h
Tenant [{#TENANT_ID}]: Total disk usage

Total disk usage hours for the current tenant (project).

Multiplying the server disk size (in GiB) by hours the server exists, and then adding that all together for each server.

 Dependent item openstack.nova.tenant.disk_usage[{#TENANT_ID}]

Preprocessing

  • JSON Path: $.total_local_gb_usage

    ⛔️Custom on fail: Set error to: Could not parse the detailed tenant report

  • JavaScript: return Math.round(value * 100) / 100;

  • Discard unchanged with heartbeat: 1h
Tenant [{#TENANT_ID}]: Total memory usage

Total memory usage hours for the current tenant (project).

Multiplying the server memory size (in MiB) by hours the server exists, and then adding that all together for each server.

 Dependent item openstack.nova.tenant.total_memory_mb_usage[{#TENANT_ID}]

Preprocessing

  • JSON Path: $.total_memory_mb_usage

    ⛔️Custom on fail: Set error to: Could not parse the detailed tenant report

  • JavaScript: return Math.round(value * 100) / 100;

  • Discard unchanged with heartbeat: 1h

OpenStack by HTTP

Overview

This template is designed for the effortless deployment of OpenStack monitoring by Zabbix via HTTP and doesn't require any external scripts.

Requirements

Zabbix version: 7.0 and higher.

Tested versions

This template has been tested on:

  • OpenStack Yoga release and OpenStack built from sources (27568ea3):

    • Identity API v3
    • Compute API v2.1 (for OpenStack Nova by HTTP template)

Configuration

Zabbix should be configured according to the instructions in the Templates out of the box section.

Setup

This is a master template that needs to be assigned to a host, and it will discover all OpenStack services supported by Zabbix automatically.

Before using this template it is recommended to create a separate monitoring user on OpenStack that will have access to specific API resources. Zabbix uses OpenStack application credentials for authorization, as it is a more secure method than a username and password-based authentication.

Below are instructions and examples on how to set up a user on OpenStack that will be used by Zabbix. Examples use the OpenStack CLI (command-line interface) tool, but this can also be done from OpenStack Horizon (web interface).

If using the CLI tool, make sure you have the OpenStack RC file for your project with a user that has rights to create other users, roles, etc., and source it, for example, . zabbix-admin-openrc.sh.

The OpenStack RC file can be obtained from Horizon.

The project that needs to be monitored is assumed to be already present in OpenStack. In the following examples, a project named zabbix is used:

# openstack project list
+----------------------------------+--------------------+
| ID                               | Name               |
+----------------------------------+--------------------+
| 28d6bb25d62b4e7e8c2d59ce056a0334 | service            |
| 4688a19e02324c42a34220e9b6a2407e | admin              |
| bc78db4bb2044148a0abf90be512fa12 | zabbix             |
+----------------------------------+--------------------+
  1. After the project name is noted, a monitoring user needs to be created. This can be done by executing an openstack user create command:
# openstack user create --project zabbix --password-prompt zabbix-monitoring
User Password:
Repeat User Password:
+---------------------+----------------------------------+
| Field               | Value                            |
+---------------------+----------------------------------+
| default_project_id  | bc78db4bb2044148a0abf90be512fa12 |
| domain_id           | default                          |
| enabled             | True                             |
| id                  | abd3eda9a29244568b1801e4825b6d71 |
| name                | zabbix-monitoring                |
| options             | {}                               |
| password_expires_at | None                             |
+---------------------+----------------------------------+
  1. When the monitoring user is created, it needs to be assigned a role. But first, a monitoring-specific role needs to be created:
# openstack role create --description "A role for Zabbix monitoring user" monitoring
+-------------+-----------------------------------+
| Field       | Value                             |
+-------------+-----------------------------------+
| description | A role for Zabbix monitoring user |
| domain_id   | None                              |
| id          | 93577a7f13184cf7af76f7bdecf7f6ee  |
| name        | monitoring                        |
| options     | {}                                |
+-------------+-----------------------------------+
  1. Then assign this newly created role to the monitoring user created in Step 1:
# openstack role add --user zabbix-monitoring --project zabbix monitoring
  1. Verify that the role has been assigned correctly. There should be one role only:
# openstack role assignment list --user zabbix-monitoring --project zabbix --names
+------------+---------------------------+-------+----------------+--------+--------+-----------+
| Role       | User                      | Group | Project        | Domain | System | Inherited |
+------------+---------------------------+-------+----------------+--------+--------+-----------+
| monitoring | zabbix-monitoring@Default |       | zabbix@Default |        |        | False     |
+------------+---------------------------+-------+----------------+--------+--------+-----------+
  1. Get the OpenStack RC file for the monitoring user in this project, source it, and generate application credentials:
# openstack application credential create --description "Application credential for Zabbix monitoring" zabbix-app-cred
  +--------------+----------------------------------------------------------------------------------------+
  | Field        | Value                                                                                  |
  +--------------+----------------------------------------------------------------------------------------+
  | description  | Application credential for Zabbix monitoring                                           |
  | expires_at   | None                                                                                   |
  | id           | c8087b91354249f3b157a50fc5ecfb3c                                                       |
  | name         | zabbix-app-cred                                                                        |
  | project_id   | bc78db4bb2044148a0abf90be512fa12                                                       |
  | roles        | monitoring                                                                             |
  | secret       | E1kC-s8QTWUaIpmexF18GW-FL3TI9-HXoexdExvGsw7uOhb3SEFW1zDa1qTs80Vqn-2xgviIPRuYOCDp2NDVUg |
  | system       | None                                                                                   |
  | unrestricted | False                                                                                  |
  | user_id      | abd3eda9a29244568b1801e4825b6d71                                                       |
  +--------------+----------------------------------------------------------------------------------------+

While creating the application credential, it is also possible to define access rules using the --access-rules flag, which offers even more fine-grained access to various API endpoints. This is optional and up to the user to decide if such rules are needed.

Once the application credential is created, the values of id and secret need to be set as user macro values in Zabbix:

  • value of id in {$APP.CRED.ID} user macro;
  • value of secret in {$APP.CRED.SECRET} user macro.

At this point, the monitoring user will not be able to access any resources on OpenStack, therefore some access rights need to be defined. Access rights are set using policies. Each service has its own policy file, therefore further steps for setting up policies, are mentioned in the template documentation of each supported service, e.g., OpenStack Nova by HTTP.

Macros used

Name Description Default
{$OPENSTACK.KEYSTONE.API.ENDPOINT}

API endpoint for Identity Service, e.g., https://local.openstack:5000.
{$OPENSTACK.AUTH.INTERVAL}

API token regeneration interval, in minutes. By default, OpenStack API tokens expire after 60m.

 50m
{$OPENSTACK.HTTP.PROXY}

Sets the HTTP proxy for the authorization item. Host prototypes will also use this value for HTTP proxy. If this parameter is empty, then no proxy is used.
{$OPENSTACK.APP.CRED.ID}

Application credential ID for monitoring user access.
{$OPENSTACK.APP.CRED.SECRET}

Application credential password for monitoring user access.

Items

Name Description Type Key and additional info
OpenStack: Get access token and service catalog

Authorizes user on the OpenStack Identity service and gets the service catalog.

 Script openstack.identity.auth

LLD rule OpenStack: Nova discovery

Name Description Type Key and additional info
OpenStack: Nova discovery

Discovers OpenStack services from the monitoring user's services catalog.

 Dependent item openstack.services.nova.discovery

Feedback

Please report any issues with the template at https://support.zabbix.com

You can also provide feedback, discuss the template, or ask for help at ZABBIX forums