Performance_analyzer: Config

Versions (relevant - OpenSearch/Dashboard/Server OS/Browser):
Open search helm vesion= 2.20.0
Dashboard version= 2.18.0
Opensearch version= 2.14.0

Describe the issue:
As I am using the latest version 2.14.0 by default performance analyzer plugin is available. How should I enable to collect the metrics. When I execute the API call ## GET localhost:9200/_plugins/_performanceanalyzer/metrics/units. I got the response like {
“error”: “no handler found for uri [/localhost:9200/_plugins/_performanceanalyzer/metrics/units] and method [GET]”
}

I saw the document, there is no straight forward congif explained for helm chart. should I do below settings. and this setup has to be done on both master, data, and dashboards?

config:
opensearch.yml: |
plugins.performance_analyzer.enabled: true

Under node I should use this config?

extraEnvs:
- name: OPENSEARCH_JAVA_OPTS
value: “-Dcom.amazon.opendistro.elasticsearch.performanceanalyzer.enabled=true”
extraVolumes:
- name: performance-analyzer-logs
emptyDir:
medium: Memory
sizeLimit: 1Gi
extraVolumeMounts:
- name: performance-analyzer-logs
mountPath: /dev/shm
extraInitContainers:
- name: install-performance-analyzer-agent
image: busybox:1.31.1
command: [‘sh’, ‘-c’, ‘mkdir -p /dev/shm && chown -R 1000:1000 /dev/shm’]

Configuration:

My current config.

clusterName: “opensearch-cluster”
nodeGroup: “data”

roles:

  • ingest
  • data
  • remote_cluster_client

replicas: 2

masterService: “opensearch-cluster-master”

extraEnvs:

  • name: OPENSEARCH_INITIAL_ADMIN_PASSWORD
    value: ###########
  • name: AWS_ROLE_ARN
    value: ${DATA_AWS_ROLE_ARN}

rbac:
create: true
serviceAccountAnnotations:
eks.amazonaws.com/role-arn: ${DATA_AWS_ROLE_ARN}
serviceAccountName: ${data_service_account_name}
automountServiceAccountToken: true

persistence:
labels:
# Add default labels for the volumeClaimTemplate of the StatefulSet
enabled: false
accessModes:
- ReadWriteOnce
size: 10Gi

plugins:
enabled: true
installList:

  • repository-s3

extraVolumes:

  • name: aws-iam-token
    projected:
    sources:
    - serviceAccountToken:
    audience: sts.amazonaws.com
    expirationSeconds: 86400
    path: token

extraVolumeMounts:

  • name: I am-token
    mountPath: /usr/share/opensearch/config/token

config:
opensearch.yml: |
cluster.name: opensearch-cluster
network.host: 0.0.0.0

plugins.security.disabled: true
plugins.security.ssl.http.clientauth_mode: NONE
plugins.security.ssl.http.enabled: false
plugins.security.ssl.transport.enabled: false
plugins.security.ssl.transport.enforce_hostname_verification: false

plugins.security.allow_default_init_securityindex: true
plugins.security.allow_unsafe_democertificates: true
plugins.security.enable_snapshot_restore_privilege: true
plugins.security.check_snapshot_restore_write_privileges: true

plugins.security.protected_indices.enabled: true
plugins.security.protected_indices.roles: ['all_access']
plugins.security.protected_indices.indices: []

plugins.security.restapi.roles_enabled: ["all_access", "security_rest_api_access"]

opensearch.experimental.feature.extensions.enabled: true

node.max_local_storage_nodes: 2

cluster.allocator.existing_shards_allocator.batch_enabled: true
cluster.routing.allocation.enable: all

snapshot.max_concurrent_operations: 1

s3.client.default.identity_token_file: /usr/share/opensearch/config/aws-iam-token/token
s3.client.default.region: us-east-1

Relevant Logs or Screenshots:

Hi,

Please give me an update on this?

Can someone please address this ?

Can someone reply.

I have enabled the plugin by using this cmd "
curl -XPOST localhost:9200/_plugins/_performanceanalyzer/cluster/config -H ‘Content-Type: application/json’ -d ‘{“enabled”: true}’"

I saw some logs are written under the “/dev/shm/performanceanalyzer/” as below.

post that what I have to do? How should I create the dashboard to see the metrics?

Should I install the performance analyzer agent and that I will run on the 9600 port? I am using the helm chart I have already given the details above.

Hi, did you resolve this problem?

@pdcuong27 @learnawsnaresh There is an issue with Performance Analyzer and it has already been reported in GitHub. This affects all versions starting from 2.12.0

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.