Infrastructure monitoring

See every host, container, and node in one table

Uptrace auto-detects OpenTelemetry host metrics and activates a dashboard with zero configuration. CPU, memory, disk, filesystem, and network from every host, grouped by fleet and alertable like everything else in your stack.

Start monitoring your hosts Schedule a demo

14-day free trial · No credit card required

Live · 6 hosts

1 alert firing

host_nameCPUMEMstatus

web-01

34%

61%

web-02

78%

55%

warn

db-primary

22%

84%

worker-03

93%

48%

crit

cache-01

18%

72%

db-replica

41%

68%

avg CPU 48%avg MEM 65%region us-east-1

Dashboards that build themselves

The moment Uptrace sees hostmetrics data, it activates a pre-built dashboard for CPU, memory, disk, and network, no manual setup, no YAML to write first.

Auto-activating template ships with Uptrace and matches on incoming metric names
Covers CPU utilization and time, memory usage, disk I/O, filesystem, network, and load average
Same mechanism powers templates for PostgreSQL, MySQL, Redis, and Kubernetes
Standard OpenTelemetry hostmetricsreceiver, no proprietary agent required

Every host in one table, not one dashboard per box

Table dashboards group by host_name so you see your whole fleet at a glance, then drill into a single row for that host's full timeseries.

CPU, memory, disk, filesystem, and network from every scraper the collector runs
Per-process metrics when you need them: CPU time and memory per process
Resource detection tags every host with cloud provider, region, and hostname automatically
Same PromQL-compatible query language as every other metric in Uptrace

Alert on any host metric

Host metrics are alertable the same way as spans and logs. Set a threshold once and route it through any of the eight notification channels covered on the alerting page.

Metric monitors fire on threshold crossings for any hostmetrics series
Group by host, region, or cloud provider to split noisy fleets into focused alerts
Recovery notifications sent automatically when a host returns to normal
See the full monitor and channel setup on the alerting page

Tag hosts with their real environment

The OpenTelemetry Resource Detection processor stamps every host with where it actually runs, so a query for "production, us-east" just works.

Cloud metadata: AWS EC2, ECS, and EKS; GCP and GKE; Azure VMs and AKS
Kubernetes pod, node, and namespace attributes for cluster-wide views
Docker container metadata when running inside a container
Filter or group any dashboard, monitor, or alert by these attributes

cloud.provider

aws, gcp, azure, or none for bare metal

cloud.region

Region or availability zone

k8s.pod.name

Pod, node, and namespace for cluster views

host.name

Resolved via DNS, OS, or FQDN lookup

Better together

Each signal is more powerful when correlated with the others. Uptrace stores traces, logs, and metrics in one place, no tool-switching, no data gaps.

Distributed tracing

Find the exact span that broke your request, across every service and database call.

Log management

Search structured logs at any scale and jump from any span to its logs in one click.

Metrics

PromQL-compatible metrics with no cardinality limits and automatic dashboards.

Common questions, clear answers

Do I need a special agent to monitor infrastructure?

No. Uptrace uses the standard OpenTelemetry Collector with the hostmetricsreceiver, the same open-source component used across the OpenTelemetry ecosystem. There is no proprietary agent to install or license.

Which metrics does infrastructure monitoring cover?

CPU utilization and time, memory usage, disk I/O and operations, filesystem usage including NFS, network I/O and connections, load average, paging and swap activity, and process counts. Per-process CPU and memory are also available when the collector runs with elevated permissions.

Do dashboards need to be configured manually?

No. Uptrace ships pre-built dashboard templates and activates them automatically the first time it sees matching metric names, so a host dashboard appears without writing any YAML or clicking through a setup wizard.

Can I monitor hosts across multiple clouds in one view?

Yes. The OpenTelemetry Resource Detection processor tags every host with its cloud provider, region, and other metadata automatically, so you can filter or group a single dashboard across AWS, GCP, Azure, Kubernetes, and bare metal hosts at once.

How is this different from the alerting product?

Infrastructure monitoring is about collecting and visualizing host-level metrics. Alerting is the layer that watches those metrics and any other signal in Uptrace and notifies you when a threshold is crossed. Host metrics work with the same metric monitors and notification channels described on the alerting page.

Ready to see your whole fleet in one table?

Deploy in minutes with Docker. Open source. Fixed pricing. Your data stays on your infrastructure.

Start monitoring your hosts

14-day free trial · No credit card required

Learn more

Documentation Pricing Schedule a demo