image/resources/knfsd-metrics-agent/config/common.yaml

# Copyright 2022 Google LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#      https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

receivers:
  connections:
    collection_interval: 1m

  mounts:
    collection_interval: 1m
    # metrics:
    #   nfs.mount.write_bytes:
    #     enabled: false
    #   nfs.mount.operation.requests:
    #     enabled: false
    #   nfs.mount.operation.errors:
    #     enabled: false
    #   nfs.mount.write_exe:
    #     enabled: false
    #   nfs.mount.write_rtt:
    #     enabled: false
    #   nfs.mount.ops_per_second:
    #     enabled: false
    #   nfs.mount.rpc_backlog:
    #     enabled: false
    #   nfs.mount.read_bytes:
    #     enabled: false
    #   nfs.mount.operation.sent_bytes:
    #     enabled: false
    #   nfs.mount.operation.received_bytes:
    #     enabled: false
    #   nfs.mount.operation.major_timeouts:
    #     enabled: false
    #   nfs.mount.read_exe:
    #     enabled: false
    #   nfs.mount.read_rtt:
    #     enabled: false

  exports:
    collection_interval: 1m
    # metrics:
    #   nfs.exports.total_operations:
    #     enabled: false
    #   nfs.exports.total_read_bytes:
    #     enabled: false
    #   nfs.exports.total_write_bytes:
    #     enabled: false

  slabinfo:
    collection_interval: 1m
    # metrics:
    #   slab.dentry_cache.objsize:
    #     enabled: false
    #   slab.nfs_inode_cache.active_objects:
    #     enabled: false
    #   slab.nfs_inode_cache.objsize:
    #     enabled: false
    #   slab.dentry_cache.active_objects:
    #     enabled: false

  # This may be useful for diagnostics to indicate if the proxy is culling.
  # However, this can be expensive as it requires scanning all the cached blocks
  # in FS-Cache, so be conservative on how often this information is scraped.
  # By default this is not included in the pipeline.
  oldestfile:
    collection_interval: 10m

processors:
  resourcedetection:
    detectors: [gce]

  # NOTE: the open telemetry collector will interpolate environment variables
  # Thus $1 will be replaced with the value of the environment variable named "1"
  # This is escaped by using two dollars, thus use $$ in regexp.
  metricstransform:
    transforms:
    - action: update
      include: nfs.connections
      new_name: nfs_connections

    - action: update
      include: nfs.mount.read_rtt
      new_name: nfsiostat_mount_read_rtt

    - action: update
      include: nfs.mount.read_exe
      new_name: nfsiostat_mount_read_exe

    - action: update
      include: nfs.mount.write_rtt
      new_name: nfsiostat_mount_write_rtt

    - action: update
      include: nfs.mount.write_exe
      new_name: nfsiostat_mount_write_exe

    - action: update
      include: nfs.mount.ops_per_second
      new_name: nfsiostat_ops_per_second

    - action: update
      include: nfs.mount.rpc_backlog
      new_name: nfsiostat_rpc_backlog

    - action: update
      include: nfs.mount.read_bytes
      new_name: mount/read_bytes

    - action: update
      include: nfs.mount.write_bytes
      new_name: mount/write_bytes

    - action: update
      include: nfs.mount.operation.requests
      new_name: mount/operation/requests

    - action: update
      include: nfs.mount.operation.sent_bytes
      new_name: mount/operation/sent_bytes

    - action: update
      include: nfs.mount.operation.received_bytes
      new_name: mount/operation/received_bytes

    - action: update
      include: nfs.mount.operation.major_timeouts
      new_name: mount/operation/major_timeouts

    - action: update
      include: nfs.mount.operation.errors
      new_name: mount/operation/errors

    - action: update
      include: nfs.exports.total_operations
      new_name: exports/total_operations

    - action: update
      include: nfs.exports.total_read_bytes
      new_name: exports/total_read_bytes

    - action: update
      include: nfs.exports.total_write_bytes
      new_name: exports/total_write_bytes

    - action: update
      include: slab.dentry_cache.active_objects
      new_name: dentry_cache_active_objects

    - action: update
      include: slab.dentry_cache.objsize
      new_name: dentry_cache_objsize

    - action: update
      include: slab.nfs_inode_cache.objsize
      new_name: nfs_inode_cache_objsize

    - action: update
      include: slab.nfs_inode_cache.active_objects
      new_name: nfs_inode_cache_active_objects

    - action: update
      include: fscache.oldest_file
      new_name: fscache_oldest_file

    # fsid daemon metrics, these are reported via the oltp receiver
    - action: update
      include: fsid.operation.count
      new_name: fsid/operation/count

    - action: update
      include: fsid.operation.duration
      new_name: fsid/operation/duration

    - action: update
      include: fsid.request.count
      new_name: fsid/request/count

    - action: update
      include: fsid.request.duration
      new_name: fsid/request/duration

    - action: update
      include: fsid.request.retries
      new_name: fsid/request/retries

    - action: update
      include: fsid.sql.query.count
      new_name: fsid/sql/query/count

    - action: update
      include: fsid.sql.query.duration
      new_name: fsid/sql/query/duration

    # prefix all metrics with custom.googleapis.com/knfsd/
    - action: update
      include: ^(.*)$$
      match_type: regexp
      new_name: custom.googleapis.com/knfsd/$$1

exporters:
  # Useful when developing or debugging metrics. When running the agent from the
  # command line this will write the metrics to the terminal.
  logging:
    loglevel: debug

  googlecloud:
    user_agent: knfsd-metrics-agent
    metric:
      prefix: ""
      # Setting this to false will automatically create the metric descriptors.
      # However, it will also reset the display name "custom.googleapi.com/knfsd/..."
      # Disabling this and creating the metric descriptors using Terraform so
      # that we can have nice display names.
      skip_create_descriptor: true

  prometheus:
    endpoint: "localhost:9090"
    send_timestamps: false
    resource_to_telemetry_conversion:
      enabled: true

service:
  telemetry:
    logs:
      level: info

    metrics:
      # disable telemetry, scraping these might confuse the data as it would
      # generate similar metrics to the Ops Agent
      level: none
      # if telemetry is enabled, run it on a different port as Ops Agent uses 8888
      address: ':8889'