Mackerel

Mackerel is a cloud monitoring service for teams running servers, applications and container workloads. It is best suited to operators who want infrastructure signals and business-level service metrics in one console. Its breadth is compelling, but Standard billing grows with hosts, metrics and trace volume, so larger estates need careful budgeting.

Overview

Installing the Mackerel agent on a server begins host monitoring, after which hosts can be arranged into services and roles. This structure maps technical data to the applications and responsibilities your team actually manages. User-defined service metrics can sit beside host measurements, covering signals such as response time, sales and page views.

Mackerel can combine metrics, monitor predicted values and apply machine-learning anomaly detection. Integrations include Amazon Web Services, Microsoft Azure and Google Cloud. Its container agent treats an Amazon ECS task or Kubernetes pod as a host, extending the same monitoring model to orchestrated workloads. Open-source plugins, APIs and the mkr command-line tool provide practical extension points. Customer Reliability Engineers can also advise on implementation and use.

Key features

Host, service and application context

Host and service metrics in one system help teams correlate infrastructure conditions with application or commercial impact. Services and roles give larger environments a clearer organizing layer than a flat host list.

Prediction and anomaly detection

Predicted-value monitoring is useful when a threshold should follow an expected trend rather than remain fixed. Machine-learning anomaly detection adds another way to surface unusual behavior, although teams should still define which signals matter operationally.

Cloud and container coverage

Native connections for AWS, Azure and Google Cloud suit multi-cloud estates. Treating ECS tasks and Kubernetes pods as hosts keeps container monitoring within the same model, rather than forcing a separate workflow.

Extensibility and support

Plugins, APIs and mkr allow teams to bring in data or automate administration beyond the built-in integrations. Reliability Engineers can guide implementation; Standard support contact is by email and is free during business days from 10:00 to 19:00.

Pricing

Mackerel uses a freemium model, with a 14-day trial and paid usage from 2,180 JPY/mo. The trial supplies Standard-plan functions and allows 100M trace spans, making it a meaningful way to evaluate the full service before committing.

Free

The Free plan costs nothing. It includes five standard and micro hosts in total, 500 metrics, one day of data retention, 10 monitoring rules, one custom dashboard and up to 5M trace spans. This is suitable for a small proof of concept or a tightly bounded service; the one-day retention and single dashboard quickly constrain historical analysis and team-wide views.

Standard

Standard has custom pricing based on usage: 2,180 yen (tax included) per standard host month, 660 yen (tax included) per micro host month, 11 yen (tax included) per metric and 330 yen (tax included) per 1M spans. A minimum usage fee applies. In return, retention extends to 460 days, with 100 monitoring rules, 100 custom dashboards and unlimited users. The separate Standard (micro host) option supports 30 metrics per micro host, which can reduce the footprint for lightly instrumented machines, while host, metric and trace charges still need to be forecast.

Platforms

Mackerel is delivered as a cloud service, with API, Linux, web and Windows access. It supports VM monitoring, service monitoring and agentless monitoring; operating-system coverage includes Linux and Windows.

Who it's for

Mackerel fits infrastructure and platform teams managing cloud, VM and container environments that also need application or business metrics. Standard is the practical choice for organizations requiring long retention, many dashboards, more rules and unlimited users. The Free plan is better for a small deployment, evaluation or low-volume monitoring footprint. Teams seeking predictable, host-unlimited costs or a self-hosted system should look elsewhere.

Pros and cons

• Pros: Host and custom service metrics share one organized view; AWS, Azure, Google Cloud, ECS and Kubernetes coverage supports modern mixed infrastructure; prediction, anomaly detection and extensibility broaden monitoring beyond fixed thresholds; unlimited users on Standard suit larger teams.
• Cons: Free retention is only one day, with five hosts, 10 rules and one dashboard; Standard charges separately for hosts, metrics and trace spans and includes a minimum usage fee; Standard technical support is email-only during stated business hours.

Alternatives

Choose Netdata instead if broad device access, including Android, iOS, macOS and self-hosted deployments, matters; its Community plan is free for up to five active connected nodes and one active custom dashboard per room.

Checkmk is a stronger fit for a free, integration-heavy community deployment, offering more than 2,000 pre-built integrations and roughly 100 hosts with community support.

Pick openITCOCKPIT for a self-hosted infrastructure and application monitoring option with a free Community Edition.

Sensu Go suits teams wanting a free tier for 100 nodes across one site, with a paid Pro option billed annually.

Checkmate is a free self-hosted choice with all features, no monitor limits and 12 alert channels.

Icinga is appropriate when an open-source, subscription-free edition with unlimited hosts and services is the priority.

MetricsHub is an alternative centered on an open-source project built around OpenTelemetry.

Prometheus is the straightforward free, open-source monitoring and alerting toolkit under the Apache 2.0 license.

Verdict

Choose Mackerel when your team needs one cloud service to connect host, application, cloud and container metrics, with anomaly detection and extensibility built in. Look elsewhere if usage-based Standard charges, short Free retention or a preference for self-hosting outweigh that unified operational view.