skills.rest
cockroachdb
cockroachdbOFFICIAL
Explore all skills by this author →
WebsiteGitHub

mma-investigator

Analyze Multi-Metric Allocator behavior in CockroachDB clusters using Datadog metrics and logs.

32.4k|4.1k|Updated Feb 6, 2014
Search Tags:#distributed systems#performance tuning#datadog#load balancing#allocator#cockroachdb#mma
One-click install
npx skills add https://github.com/cockroachdb/cockroach --skill mma-investigator
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mma-investigator
Source: https://github.com/cockroachdb/cockroach/tree/main/pkg/kv/kvserver/allocator/mmaprototype/.claude/skills/mma-investigator
Command: npx skills add https://github.com/cockroachdb/cockroach --skill mma-investigator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps on-call engineers diagnose and understand load imbalances within CockroachDB clusters, explaining why the Multi-Metric Allocator (MMA) did or did not take specific rebalancing actions.

Core Features & Use Cases

  • Cluster Balance Assessment: Analyzes resource distribution (CPU, disk, network) across cluster nodes and stores.
  • Rebalancing Timeline: Reconstructs the sequence of rebalancing events, identifying triggers and actions.
  • MMA Performance Analysis: Evaluates MMA's success/failure rates and identifies reasons for inaction.
  • Use Case: An engineer notices high CPU on a few nodes. They use this Skill to understand if MMA is aware of the imbalance, if it attempted rebalancing, and why it might have failed or been ineffective, providing clear evidence from metrics and logs.

Quick Start

Investigate MMA behavior on the 'prod-us1' cluster for the last 4 hours.

Frequently Asked Questions about mma-investigator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
Why does CockroachDB load balancing fail to rebalance high CPU usage across nodes?▼

CockroachDB load balancing via the Multi-Metric Allocator may fail to rebalance CPU due to replica constraints, leaseholder preferences, or insufficient available capacity. This Skill analyzes Datadog metrics and logs to identify the exact reasons MMA did not take action.

How do I diagnose write bandwidth imbalances in a CockroachDB cluster?▼

To diagnose write bandwidth imbalances in a CockroachDB cluster, you need to analyze resource distribution across nodes and stores. This Skill queries Datadog metrics to evaluate MMA rebalancing decisions and reconstructs the timeline of events causing the imbalance.

Does this MMA investigator tool require Datadog integration for cluster performance tuning?▼

Yes, this MMA investigator requires Datadog integration to access the necessary metrics and logs. It queries Datadog directly to assess CPU, disk, network, and replica distributions across CockroachDB cluster nodes.

What is the best way to check why the CockroachDB allocator is not moving replicas?▼

The best way to check why the CockroachDB allocator is not moving replicas is to evaluate MMA performance analysis data. This Skill identifies specific reasons for inaction by examining rebalancing successes, failures, and triggers within a specified time window.

Can I use Datadog metrics to trace CockroachDB lease distribution issues?▼

Yes, you can use Datadog metrics to trace CockroachDB lease distribution issues. This Skill provides expert analysis of lease distributions alongside CPU, disk, and write bandwidth metrics to explain MMA's rebalancing behavior.

How do I investigate CockroachDB load imbalances over a specific time window?▼

To investigate CockroachDB load imbalances over a specific time window, you provide the cluster name and desired duration. This Skill queries Datadog to reconstruct the rebalancing timeline and evaluates MMA actions during that exact period.

Related Skills

Explore More Software Engineering Skills→
cockroachdbcockroachdb
3

configuring-log-export

Configure CockroachDB Cloud log and metric export to external monitoring services.

Official
Advanced
jsamuelsen11jsamuelsen11

sharding-strategies

Plan MongoDB sharding strategies for shard-key design, zone patterns, and chunk management.

Community
Advanced
incidentfoxincidentfox
654

jaeger-analysis

Analyze Jaeger traces to identify latency issues and errors across services.

Official
Intermediate
cockroachlabscockroachlabs
18

analyzing-range-distribution

Diagnose range distribution and fragmentation in CockroachDB using SHOW RANGES queries.

Official
Advanced