Skip to content

feat(lm-logs): default cri tail parser, optional monitor agent, and input validation#223

Merged
lmshubhashri merged 3 commits into
mainfrom
lm-logs/1.1.1
May 19, 2026
Merged

feat(lm-logs): default cri tail parser, optional monitor agent, and input validation#223
lmshubhashri merged 3 commits into
mainfrom
lm-logs/1.1.1

Conversation

@lm-aditya-kadam

@lm-aditya-kadam lm-aditya-kadam commented May 19, 2026

Copy link
Copy Markdown
Contributor

CRI container log parser — kubernetes.containerTailParserType defaults to cri and the generated tail
block uses @type cri when selected, which aligns with containerd/CRI-O log format and avoids
relying on FLUENT_CONTAINER_TAIL_PARSER_TYPE for the common case. multi_format remains available for
legacy or mixed environments.
Monitor agent — fluent.monitorAgent (enabled/bind/port) injects a monitor_agent into
fluent.conf and exposes the chosen port on the DaemonSet when enabled, so operators can scrape or
port-forward to /api/plugins.json and related endpoints.

@lm-aditya-kadam lm-aditya-kadam changed the title feat(lm-logs): CRI tail parser default, optional monitor_agent, and input validation feat(lm-logs): default cri tail parser, optional monitor agent, and input validation May 19, 2026
lm-aditya-kadam and others added 3 commits May 19, 2026 17:54
Use kubernetes.containerTailParserType (default cri) so kubernetes.conf emits a
valid <parse>@type cri block for containerd/CRI-O logs.

Keep multi_format + patterns as optional via Helm value.

Bump chart to 1.1.0-rc02.
* feat(lm-logs): add configurable monitor_agent for buffer metrics

Add fluent.monitorAgent config block to expose Fluentd plugin/buffer metrics
via HTTP when enabled. Conditionally inject <source @type monitor_agent> into
ConfigMap and expose containerPort on DaemonSet when enabled.

Changes:
- Add fluent.monitorAgent config (enabled, bind, port) in values.yaml
- Conditionally include <source> block in configmap.yaml template
- Conditionally expose containerPort in deamonset.yaml template
- Update values.schema.json with monitorAgent field definition
- Bump chart version to 1.1.0-rc04

Feature disabled by default. Enable by setting:
  lm-logs:
    fluent:
      monitorAgent:
        enabled: true
        bind: 0.0.0.0
        port: 24220

* feat(lm-logs): add configurable monitor_agent for buffer metrics

Add fluent.monitorAgent config block to expose Fluentd plugin/buffer metrics
via HTTP when enabled. Conditionally inject <source @type monitor_agent> into
ConfigMap and expose containerPort on DaemonSet when enabled.

Changes:
- Add fluent.monitorAgent config (enabled, bind, port) in values.yaml
- Conditionally include <source> block in configmap.yaml template
- Conditionally expose containerPort in deamonset.yaml template
- Update values.schema.json with monitorAgent field definition
- Bump chart version to 1.1.1-rc01
ct / yamllint require a newline at end of chart.yaml. Without it, chart-testing
fails on rule new-line-at-end-of-file.

Co-authored-by: Cursor <cursoragent@cursor.com>
@lmshubhashri

Copy link
Copy Markdown
Contributor

merging on behalf of @lm-aditya-kadam

@lmshubhashri lmshubhashri merged commit f766d32 into main May 19, 2026
12 checks passed
@lmshubhashri lmshubhashri deleted the lm-logs/1.1.1 branch May 19, 2026 13:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants