Add MongoDB OpenTelemetry content pack

[shmsr]

Summary

• Adds a new mongodb_otel content pack (type: content) with Kibana dashboards, alerting rule templates, and SLO templates for monitoring MongoDB via the OpenTelemetry mongodbreceiver.
• Includes 3 dashboards (Overview, Operations, Capacity), 8 alert rules, and 3 SLO templates.
• All assets consume metrics from the metrics-mongodbreceiver.otel-* data stream — no optional receiver metrics need to be enabled.

Dashboards

[MongoDB OTel] Overview

Operation throughput, average latency, WiredTiger cache hit ratio, and connection status.

[MongoDB OTel] Operations

Operation rate by type, operation time trends, global lock contention, and network I/O.

[MongoDB OTel] Capacity

Connection utilization, memory usage, storage and index sizes by database, and cursor/session counts.

Alert rules

Alert	Trigger	Severity
Connection exhaustion	Available connections < 10	Critical
Cursor timeouts	Any increase in cursor timeout count	High
Global lock contention	Lock held > 500 ms/sec avg	High
High cache miss rate	Cache miss rate > 5%	High
High cursor count	Open cursors > 1000	Medium
High memory usage	Resident memory > 4 GB	High
High operation latency	Avg latency > 100 ms	High
High session count	Active sessions > 500	Medium

SLO templates

SLO	Target	Window
Avg operation latency	99.5%	30-day rolling
Cache hit ratio	99.5%	30-day rolling
Connection availability	99.5%	30-day rolling

Test plan

• elastic-package build succeeds for packages/mongodb_otel
• elastic-package validate passes (format version, manifest, dashboards)
• Dashboards render correctly in Kibana with sample mongodbreceiver data
• Alert rules and SLO templates load without errors

[github-actions]

✅ Vale Linting Results

No issues found on modified lines!

---

The Vale linter checks documentation changes against the Elastic Docs style guide.

To use Vale locally or report issues, refer to Elastic style guide for Vale.

[ishleenk17]

@shmsr

• Global Lock and operations time by type can have a better Y axis naming.
• Connection utilization is 0%. I see 2 active connections. So should it show some prercentage ?
• Add stack version note for SLO templates

[strawgate]

I ran kb-dashboard 0.4.0 upgrade validation on this PR in an isolated worktree.

Results

• packages/mongodb_otel/_dev/shared
  • upgrade --fail-on-change exit: 3
  • upgrade --write exit: 0
  • post-upgrade compile --format elastic-integrations exit: 0

Upgrade and post-upgrade compile succeeded with 0.4.0.
Please apply these commands in your branch and commit both YAML + regenerated JSON artifacts.

Suggested commands:

uvx --from kb-dashboard-cli==0.4.0 kb-dashboard upgrade --input-dir packages/mongodb_otel/_dev/shared --write
uvx --from kb-dashboard-cli==0.4.0 kb-dashboard compile --input-dir packages/mongodb_otel/_dev/shared --output-dir packages/mongodb_otel/kibana/dashboard --format elastic-integrations

[ishleenk17]

@shmsr

• Global Lock and operations time by type can have a better Y axis naming.
• Connection utilization is 0%. I see 2 active connections. So should it show some prercentage ?
• Add stack version note for SLO templates

These issues aren't addressed yet

[shmsr]

@shmsr

• Global Lock and operations time by type can have a better Y axis naming.
• Connection utilization is 0%. I see 2 active connections. So should it show some prercentage ?
• Add stack version note for SLO templates

These issues aren't addressed yet

Apologies for the delay. Now I have address the issues. PTAL.

[macroscopeapp]

🟠 High alerting_rule_template/mongodb_otel-high-session-count.json:14

The params object for .es-query with esqlQuery searchType is missing required fields (size, threshold, thresholdComparator, termField, termSize, excludeHitsFromPreviousRun) that all other OTel integration alerting templates consistently include. Without threshold and thresholdComparator, the rule cannot determine when to fire alerts — it will likely fail to load or not trigger as expected. The standard pattern for ESQL queries with inline filtering is threshold: [0] and thresholdComparator: ">" so the alert fires when any rows are returned.

🤖 Copy this AI Prompt to have your agent fix this:

In file packages/mongodb_otel/kibana/alerting_rule_template/mongodb_otel-high-session-count.json around line 14:

The `params` object for `.es-query` with `esqlQuery` searchType is missing required fields (`size`, `threshold`, `thresholdComparator`, `termField`, `termSize`, `excludeHitsFromPreviousRun`) that all other OTel integration alerting templates consistently include. Without `threshold` and `thresholdComparator`, the rule cannot determine when to fire alerts — it will likely fail to load or not trigger as expected. The standard pattern for ESQL queries with inline filtering is `threshold: [0]` and `thresholdComparator: ">"` so the alert fires when any rows are returned.

[elasticmachine]

💚 Build Succeeded

• Buildkite Build
• Commit: d3f541d

History

• 💚 Build #40930 succeeded 742e070
• 💚 Build #40920 succeeded 99a7b36
• 💚 Build #39653 succeeded 130d802

cc @shmsr

[elastic-vault-github-plugin-prod]

Package mongodb_otel - 0.1.0 containing this change is available at https://epr.elastic.co/package/mongodb_otel/0.1.0/