Concord
HA Cockpit for Patroni
Switch over with evidence — not with curl under pressure.
Local-first Mac cockpit for Patroni day-2: live topology, Confidence, Safe Ops, and an audit trail you can take to a postmortem.
Not affiliated with Concord CLM (concord.app), Patroni / Zalando, or Walmart Labs Concord.
The cost of blind HA
Patroni estates are still run from terminals, wiki checklists, and Grafana that only observes. Planned switchover becomes tribal knowledge. A lost leader becomes a race against lag you cannot see in one place. When REST is behind SSH or Kubernetes, the picture fragments further — and the blast radius is downtime plus a reputation hit.
What Concord commits to
See the estate, decide with live evidence, execute behind guardrails, leave a record. Not a SQL IDE. Not a full Kubernetes operator. A day-2 HA cockpit for operators who already own Patroni.
- Confidence
- Switchover readiness as ready / caution / blocked — before you confirm a mutate.
- Live evidence
- Trust probes for DCS last-seen, write VIP reachability, and backup age — next to topology, not in a separate console.
- Safe Ops
- Preflight → confirm → progress → audit for switchover, pause, restart, reload. Advanced paths gated for Pro.
- Audit you can export
- Local trail of preview, success, failure, and dry-run — with integrity hash for postmortems and change review.
Built for
- DBA mid-market who prepare switchovers weekly and refuse curl under pressure
- SRE / Platform with mixed VM + Kubernetes Patroni estates
- Air-gap / partner paths that need offline signed license (via Partners)
Not for
- SQL IDE or query workbench buyers
- Full APM / observability platforms
- Cloud-only Patroni control planes you do not operate yourself
- Mongo-first or non-Postgres HA as the primary job
Where Concord sits
Observe → mutate safely → audit, local-first. Curl is the floor; Grafana observes; ClusterControl and Ivory solve different jobs.
| Observe | Mutate | Audit | Local-first | |
|---|---|---|---|---|
| curl / patronictl | Manual | Raw REST | Chat / wiki | Yes |
| Grafana | Strong | No | Separate | Depends |
| ClusterControl | Broad | Agent / fleet | Platform | Server-centric |
| IvorySQL tooling | DB-focused | Varies | Varies | Varies |
| Concord | Topology + Confidence | Safe Ops guardrails | Local exportable | Mac local-first |
Operator-grade proof points
- Direct, SSH jump, and Kubernetes port-forward in one workspace model
- Project board: leader, lag, pause, last check across clusters
- Config snapshots, diff, and PATCH with the same audit discipline as ops
- Community is a real tool for ≤3 clusters — not a demo stub
- Local-first: SQLite catalog, Keychain secrets, usable on VPN and air-gap paths
Built for DBA and SRE
Primary: database administrators who prepare switchovers weekly and cannot afford curl under pressure. Secondary: platform / SRE teams with mixed VM and Kubernetes Patroni. Later: MSP and compliance buyers who need shared audit and policy — Team and Enterprise paths.
Start Community. Unlock Pro when the estate grows.
Basic health and confirmed switchover for small estates stay free. Scale, Kubernetes, advanced ops, alerts, and longer retention are Pro — without dark-pattern countdowns.
Community≤3 clusters · Direct + SSH · observe · switchover / pause · audit 90d
ProUnlimited · K8s · full ops · config edit · alerts · retention+
How teams use Concord
Anonymized day-2 scenarios from design-partner shaped estates — planned switchover, lost leader, air-gap license.
News
Published from the Concord CMS when available.
Showing static fallback — CMS unreachable or empty.