detectMultilineErrors on multiple output cause "Duplicated plugin id multiline-detect-except"

[ahmet2mir]

Hello OpenShift!

Describe the bug

When enabling detectMultilineErrors on multiple pipeline item of ClusterLogForwarder we have an issue on Fluentd config

Example

    pipelines:

    - detectMultilineErrors: true
      inputRefs:
      - tenant1-input
      labels:
        cluster: c01e1
        namespace: tenant1
        type: application
      name: tenant1-pipeline
      outputRefs:
      - tenant1-output
      parse: json

    - detectMultilineErrors: true
      inputRefs:
      - tenant2-input
      labels:
        cluster: c01e1
        namespace: tenant2
        type: application
      name: tenant2-pipeline
      outputRefs:
      - tenant2-output
      parse: json

Environment

• OpenShift 4.9

Logs

$ oc  logs -n openshift-logging collector-dp7wg collector
Setting each total_size_limit for 4 buffers to 4811272243 bytes
Setting queued_chunks_limit_size for each buffer to 573
Setting chunk_limit_size for each buffer to 8388608
2022-03-14 19:14:29 +0000 [warn]: '@' is the system reserved prefix. It works in the nested configuration for now but it will be rejected: @timestamp
<Fluent::Config::Section {"@label":"@TENANT1_PIPELINE","tag":"","matches":["<Fluent::Config::Section {\"labels\":{},\"namespaces\":[\"tenant1\"],\"hosts\":[],\"container_names\":[],\"negate\":false}>"]}>
<Fluent::Config::Section {"@label":"@TENANT2_PIPELINE","tag":"","matches":["<Fluent::Config::Section {\"labels\":{},\"namespaces\":[\"tenant2\"],\"hosts\":[],\"container_names\":[],\"negate\":false}>"]}>
<Fluent::Config::Section {"@label":"@_APPLICATION_ALL","tag":"","matches":["<Fluent::Config::Section {\"labels\":{},\"namespaces\":[],\"hosts\":[],\"container_names\":[],\"negate\":false}>"]}>
2022-03-14 19:14:29 +0000 [error]: config error file="/etc/fluent/fluent.conf" error_class=Fluent::ConfigError error="Duplicated plugin id `multiline-detect-except`. Check whole configuration and fix it."

Expected behavior

Able to use detectMultilineErrors on multiple outputs.

Actual behavior

Using on multiple output crash loggin pods.

To Reproduce
Add detectMultilineErrors on multiple item.

Additional context

Fluentd generate

# Copying pipeline tenant1-pipeline to outputs
<label @TENANT1_PIPELINE>
...
  <match kubernetes.**>
    @id multiline-detect-except
    @type detect_exceptions
    remove_tag_prefix 'kubernetes'
    message message
    force_line_breaks true
    multiline_flush_interval .2
  </match>
...
</label>

# Copying pipeline tenant2-pipeline to outputs
<label @TENANT2_PIPELINE>
...
  <match kubernetes.**>
    @id multiline-detect-except
    @type detect_exceptions
    remove_tag_prefix 'kubernetes'
    message message
    force_line_breaks true
    multiline_flush_interval .2
  </match>
...
</label>

[jcantrill]

Fixed by #1333

[ahmet2mir]

Damn, I searched multiline-detect-except and detectMultilineErrors but found nothing (even on sourcegraph).
Thanks @jcantrill