Uploaded image for project: 'Observability Documentation'
  1. Observability Documentation
  2. OBSDOCS-2505

Missing documentation for input and output tuning options

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Undefined Undefined
    • None
    • Logging 6.2, Logging 6.3
    • Logging
    • False
    • Hide

      None

      Show
      None
    • False
    • Important

      Not present tuning options for input and outputs [0][1]. In specific, it's spoken about the sections available in Logging 5 [2] and [3] and not present in Logging v6

      [0]

      $ oc explain clf.spec.inputs.application.tuning.rateLimitPerContainer
      GROUP:      observability.openshift.io
      KIND:       ClusterLogForwarder
      VERSION:    v1
      
      FIELD: rateLimitPerContainer <Object>
      
      
      DESCRIPTION:
          RateLimitPerContainer is the limit applied to each container
          by this input. This limit is applied per collector deployment.
          
      FIELDS:
        maxRecordsPerSecond	<integer> -required-
          MaxRecordsPerSecond is the maximum number of log records
          allowed per input/output in a pipeline
      

      [1] 

      $ oc explain clusterlogforwarders.observability.openshift.io.spec.outputs.*.tuning
      GROUP:      observability.openshift.io
      KIND:       ClusterLogForwarder
      VERSION:    v1FIELD: tuning <Object>
      DESCRIPTION:
          Tuning specs tuning for the output
          
      FIELDS:
        compression    <string>
        enum: none, gzip, snappy, zlib
          Compression causes data to be compressed before sending over the network.  deliveryMode    <string>
        enum: AtLeastOnce, AtMostOnce
          DeliveryMode sets the delivery mode for log forwarding.
          This optional setting. When it is left unset, the system defaults to using
          an in-memory buffer.
          In-memory buffers offer the highest performance due to low latency, but they
          have two limitations:
          they will consume memory, and they do not provide durability — buffered data
          is lost on process termination or failure.  maxRetryDuration    <integer>
          MaxRetryDuration is the maximum time to wait between retry attempts after a
          delivery failure.  maxWrite    <Object>
          MaxWrite limits the maximum payload in terms of bytes of a single "send" to
          the output.  minRetryDuration    <integer>
          MinRetryDuration is the minimum time to wait between attempts to retry after
          delivery a failure. 

      [2] https://docs.redhat.com/en/documentation/openshift_container_platform/4.14/html/logging/performance-and-reliability-tuning#logging-flow-control-mechanisms
      [3] https://docs.redhat.com/en/documentation/openshift_container_platform/4.14/html/logging/log-collection-and-forwarding-2#logging-delivery-tuning_configuring-log-forwarding

              amehenda@redhat.com Ashwin Mehendale
              rhn-support-kahara Kazuhisa Hara
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated: