---
title: "Dasha - performance dashboard"
url: https://daily.dev/posts/dasha---performance-dashboard-cm6mu78wl
source_url: https://www.postgresql.org/about/news/dasha-performance-dashboard-3360
type: article
source: "PostgreSQL"
published: 2026-08-12T07:35:55.393Z
updated: 2026-08-12T07:36:21.747Z
tags: ["open-source", "postgresql", "mcp"]
reading_time: 3
upvotes: 1
comments: 0
language: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Dasha - performance dashboard

**[PostgreSQL](https://daily.dev/sources/postgres)** · 3 min read · 1 upvotes · 0 comments

## Summary

Dasha is an open source performance dashboard for PostgreSQL fleets that requires no agent or extension installation on database hosts. It connects via a read-only role and monitors multiple clusters and their replicas simultaneously. Key features include a composite 0-100 health score across eight categories, query analysis with pg_stat_statements snapshots, automatic snapshot capture on activity spikes, index analysis with drop-safety verdicts, table/maintenance tracking, lock and connection monitoring, OIDC/RBAC access control, and an MCP connector exposing 26 tools and 5 prompts to AI assistants. It supports PostgreSQL 14–18 and is licensed under GPL v3.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.postgresql.org/about/news/dasha-performance-dashboard-3360>

## Questions this post answers

### How can I monitor multiple PostgreSQL clusters and their replicas without installing any agent on the database hosts?

Dasha connects to each host in a cluster using a read-only role (requiring pg_monitor and CONNECT privileges) and reads per-instance statistics directly via SQL, with no agent or extension installed on the hosts. Because PostgreSQL statistics are not replicated, it reads every host — primary and standbys — and reasons over the combined picture, so an index unused on the primary but active on a standby is correctly identified.

_Teams running PostgreSQL fleets track tools like Dasha on daily.dev as they evaluate agentless monitoring options._

### What does Dasha's index drop safety check actually look at beyond idx_scan = 0?

Dasha's 'may I drop this index?' verdict accounts for statistics resets, per-host counters across primary and replica instances, and partitioned indexes — rather than relying solely on idx_scan = 0. It also flags bloat, duplicate indexes, invalid indexes, and B-tree indexes on arrays as part of its index analysis feature.

_PostgreSQL engineers weighing index cleanup decisions find the nuances of tools like this covered on daily.dev._

### What PostgreSQL versions does Dasha support?

Dasha supports PostgreSQL 14 or newer, and has been tested on versions 14 through 18. The monitored clusters must be reachable with a role holding pg_monitor and CONNECT privileges. pg_stat_statements is required for query statistics, while pgstattuple is optional.

_Developers keeping PostgreSQL tooling current with supported versions find relevant releases on daily.dev._

## Similar posts on daily.dev

- [Discover Databases, starting with Postgres](https://daily.dev/posts/discover-databases-starting-with-postgres-vudl6ia8e) · Last9 · 1 upvotes · 0 comments
- [MCP For PostgreSQL: Automated Health Checks & Performance Analysis](https://daily.dev/posts/mcp-for-postgresql-automated-health-checks-performance-analysis-5fmjhpx51) · Planet PostgreSQL · 0 upvotes · 0 comments

---

Tags: [#open-source](https://daily.dev/tags/open-source), [#postgresql](https://daily.dev/tags/postgresql), [#mcp](https://daily.dev/tags/mcp)

[View this post on daily.dev](https://daily.dev/posts/dasha---performance-dashboard-cm6mu78wl)