Skip to content

[monitoring] Improve multi-tenant log isolation, dashboard scoping, and VM monitoring #2194

Open

Description

@mattia-eleuteri

Description

Several gaps exist in the current monitoring stack for multi-tenant deployments. This issue tracks them together since they affect the same area.

1. No per-tenant log routing (CRITICAL)

monitoring-agents fluent-bit sends all container logs to vlogs-generic.<target>.svc (typically tenant-root).

Tenants that deploy their own monitoring stack get a VLogs instance (vlogs-generic.<tenant>.svc) but it receives no logs — everything goes to the parent tenant.

Current workaround: Manual patch of the fluent-bit ConfigMap with rewrite_tag + per-tenant OUTPUT blocks.

➡️ PR: #2195

2. Infrastructure metadata leaked in logs

kubernetes_node_name is exposed in log metadata, leaking node topology to end-users. Other sensitive fields: kubernetes_host, infrastructure labels.

➡️ PR: #2196

3. All tenants receive infrastructure dashboards

dashboards.list loads VictoriaMetrics, Hubble, LINSTOR, Flux, control-plane, and other infrastructure dashboards for every tenant, even though these are only relevant to the platform operator (tenant-root).

➡️ PR: #2197

4. Missing VM dashboard files

dashboards.list on main references vm/virtual-machines and vm/console-logs but the corresponding JSON files do not exist under dashboards/vm/.

➡️ PR: #2198

5. No KubeVirt VM metrics scraping

No network policy allows tenant vmagent to reach virt-handler on port 8443 for KubeVirt VM metrics.

➡️ PR: #2199

6. vmagent lacks additionalScrapeConfigs

The tenant VMAgent resource does not support additionalScrapeConfigs, preventing custom scrape targets.

➡️ PR: #2200


PR Summary

# PR Scope Component
1 #2195 Per-tenant log routing in fluent-bit monitoring-agents
2 #2196 Strip infrastructure metadata from logs monitoring-agents
3 #2197 Scope infra dashboards to tenant-root only monitoring
4 #2198 Add KubeVirt VM dashboards monitoring
5 #2199 Egress to virt-handler for VM metrics tenant
6 #2200 additionalScrapeConfigs support in vmagent monitoring

All PRs are independent and can be reviewed/merged separately.

Environment

  • Cozystack v1.0.0-rc.2
  • Multi-tenant deployment with child tenants using monitoring

Activity

  1. added
    epicA large development increment that brings definite value to Cozystack users
    area/monitoringIssues or PRs related to the monitoring stack (vlogs, vmstack, grafana, workloadmonitor)
    kind/featureCategorizes issue or PR as related to a new feature
    triage/acceptedIndicates an issue is ready to be actively worked on
    priority/important-soonMust be staffed and worked on either currently, or very soon, ideally in time for the next release
    on Jun 3, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    area/monitoringIssues or PRs related to the monitoring stack (vlogs, vmstack, grafana, workloadmonitor)epicA large development increment that brings definite value to Cozystack userskind/featureCategorizes issue or PR as related to a new featurepriority/important-soonMust be staffed and worked on either currently, or very soon, ideally in time for the next releasetriage/acceptedIndicates an issue is ready to be actively worked on

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions