VerOps
Get started
PlatformObserve

Infrastructure and Kubernetes monitoring

See the servers, containers and clusters your services run on, next to the traces and logs they produce. A small Rust agent covers Linux and Windows hosts, a Helm-installed agent covers Kubernetes, and cloud integrations bring in AWS and Azure metrics.

Get started Read the documentation
Host agent
Linux and Windows, written in Rust
Connections
Outbound only, no inbound ports
Kubernetes
Helm chart, read-only RBAC
Cloud
AWS CloudWatch, Azure Monitor

Hosts with the machine agent

The machine agent runs as a systemd unit on Linux and as a native service on Windows. It checks in on a fixed heartbeat and runs the commands the platform hands back.

  • Host health

    CPU and memory on every heartbeat, and capacity and free space for each mounted volume, with SSD or HDD noted.

  • Processes

    Top processes by CPU and memory, with parent and child processes added up, so a server with many worker processes shows where the load really is.

  • Commands on demand

    From VerOps you can read a log file, list processes or run a health check on the host. Results come back on the next heartbeat.

  • Server inventory (optional)

    An inventory module, off by default, reports installed software, users and hardware into the same asset estate that Scout feeds from workstations.

    Documentation

  • Opt-in self-update

    The agent can replace its own binary when you publish a newer release to it. It is off by default and the download is checksum-verified.

    Documentation

  • Containers and hosts from OpenTelemetry

    Hosts and containers that send OpenTelemetry data with host.name or container.name attributes appear on the Infrastructure page too, with or without the agent.

    Documentation

Kubernetes

The Kubernetes agent is installed with a Helm chart. Its cluster role only gets, lists and watches resources.

  • Cluster, nodes and pods

    Nodes, namespaces and pods with their status, restarts and resource use from kubelet stats.

  • Workloads

    Deployments, ReplicaSets, DaemonSets, StatefulSets, Jobs, CronJobs and autoscalers, so you can tell a crash loop from a rollout.

  • Services, storage and events

    Services and endpoints, ingresses, persistent volumes and claims, resource quotas, and the cluster event stream.

  • Guided investigations

    The Kubernetes page walks through common cases such as crash-looping and pending pods.

    Documentation

Cloud integrations

  • AWS

    CloudWatch metrics, connected through cross-account IAM role assumption (recommended) or static access keys.

    Documentation

  • Azure

    Azure Monitor metrics, connected through multi-tenant app consent (recommended) or a service principal.

    Documentation

  • Stored credentials

    Cloud credentials are encrypted at rest with AES-256-GCM.

  • Same dashboards

    Cloud metrics land in the same store as everything else and can be charted and alerted on in the same way.

    Documentation

Getting started

  1. 01

    Add a host

    Under Infrastructure, then Machine agents, create an agent to get its key, then install the Linux or Windows package from Resources.

  2. 02

    Add a cluster

    Install the chart with your API key and a cluster name people will recognise in alerts. It runs one cluster-level pod and one pod per node.

    helm repo add verops https://charts.verops.io
helm install verops-agent verops/verops-k8s-agent \
  -n verops --create-namespace \
  --set apiKey=$VEROPS_API_KEY \
  --set clusterName=prod-east
  3. 03

    Connect a cloud account

    Under Cloud Integrations, follow the AWS or Azure wizard and check the health state once collection starts.

Limits worth knowing

  • The machine agent supports Linux and Windows. macOS is not supported.
  • The machine agent reads log files on request. It does not tail and ship every log continuously; use an OpenTelemetry Collector for that.
  • Docker containers are not discovered by the machine agent. They appear from OpenTelemetry container attributes.
Documentation

Read more in the docs

Try it on your own systems

Create a workspace, or tell us about your environment and we will walk you through it.

Get started Talk to sales