How to Silence Noisy Logs Using Adaptive Logs Drop Rules: A Practical Guide

By ✦ min read

Introduction

If your organization is drowning in low-value logs—like health check pings, forgotten debug statements, or verbose info messages from rarely used services—you’re not alone. These logs inflate your observability bill and clutter your dashboards. Centralized platform teams often struggle to drop them without wading through tedious infrastructure changes. With Grafana Cloud’s new Adaptive Logs drop rules (now in public preview), you can define custom rules to discard worthless logs before they’re even ingested, reducing noise and saving money instantly. This guide walks you through creating and applying these rules step by step.

How to Silence Noisy Logs Using Adaptive Logs Drop Rules: A Practical Guide

What You Need

A Grafana Cloud account with access to Adaptive Logs (public preview).
Permissions to create drop rules (typically Admin or Editor role).
Knowledge of your log labels (e.g., service, namespace, level).
Understanding of which log lines are noise in your environment.
Optional: Familiarity with log sampling concepts (drop percentage).

Step-by-Step Instructions

Step 1: Navigate to Adaptive Logs

Log in to your Grafana Cloud instance. From the main menu, go to Logs > Adaptive Logs. If you don’t see the option, ensure your stack has the feature enabled (contact support if needed). The Adaptive Logs dashboard shows current volume, exemptions, existing drop rules, and pattern recommendations.

Step 2: Create a New Drop Rule

Click the Drop Rules tab, then select Create Drop Rule. You’ll be prompted to define the rule’s name, priority, and criteria. Give it a descriptive name like “Drop health check logs from API” so you can identify it later.

Step 3: Define the Rule Criteria

Using the rule builder, set the conditions that identify noisy logs. You can combine:

Log level: Choose DEBUG, INFO, WARN, ERROR. For example, drop all DEBUG logs from any service.
Log labels: Select a label (e.g., service = health-checker).
Line content: Use a regex or string match (e.g., heartbeat or /health).
Drop percentage: Decide what fraction to discard. 100% drops everything matching the rule; lower values sample the logs (e.g., 90% keeps 10% for troubleshooting).

For instance, to target a chatty batch job, you’d set service = batch-processor, level = INFO, and drop percentage = 90%.

Step 4: Set Rule Priority

Drop rules are evaluated in priority order (lower number = higher priority). If multiple rules match a log line, only the first applies. Place more specific rules at higher priority. For example, a rule dropping all health checks (priority 1) should come before a broad rule dropping DEBUG logs (priority 2). Use the Priority field to assign an integer.

Step 5: Test Your Rule (Optional but Recommended)

Before enabling the rule, you can preview its impact. Grafana Cloud shows estimated volume reduction based on recent logs. Check that you’re not accidentally dropping important logs, like error messages from a critical service. Adjust criteria if the preview indicates too much or too little filtering.

Step 6: Activate the Drop Rule

Once satisfied, toggle the rule to Active. The rule will start affecting incoming logs immediately. Logs that match are discarded before write, so you’ll see reduced ingestion in your bill and dashboards within minutes.

Step 7: Monitor and Iterate

Return to the Adaptive Logs dashboard to view the volume drop. Check the Drop Rules tab for stats on how many logs each rule has dropped. If you notice false positives, modify the criteria or adjust the drop percentage. You can also reorder rules by editing their priority.

Tips for Success

Start conservatively: Use a low drop percentage (e.g., 50%) for new rules to avoid removing logs you still need.
Combine with exemptions: Protect critical logs by creating exemptions before drop rules. Exemptions pass through untouched, ensuring you never lose important data.
Leverage pattern recommendations: Adaptive Logs also offers optimization suggestions based on repetitive patterns. Use drop rules to handle known noise, and let recommendations handle the rest.
Document your rules: Keep a record of what each rule targets and why. This helps your team understand the cost-saving measures.
Review periodically: As services change, so does the log profile. Regularly revisit your drop rules to ensure they still make sense.

By following these steps, you can quickly suppress noisy logs and reduce your Grafana Cloud Logs bill—all without touching application code or infrastructure configurations.

Tags: