The missing set of useful tools for Postgres DBAs.
git clone https://github.com/NikolayS/postgres_dba.git
cd postgres_dba
printf "%s %s %s %s\n" \\echo 🧐 🐘 'postgres_dba installed. Use ":dba" to see menu' >> ~/.psqlrc
printf "%s %s %s %s\n" \\set dba \'\\\\i $(pwd)/start.psql\' >> ~/.psqlrcThen connect to any Postgres server via psql and type :dba to open the interactive menu.
Requires psql 10+. The Postgres server itself can be older for most reports. For best results, use psql from the latest PostgreSQL release.
| ID | Report |
|---|---|
| 0 | Node information: primary/replica, lag, database size, temp files |
| 1 | Database sizes and stats |
| 2 | Table and index sizes, row counts |
| 3 | Load profile |
| ID | Report |
|---|---|
| a1 | Current activity: connections grouped by database, user, state |
| l1 | Lock trees (lightweight) |
| l2 | Lock trees, detailed (on PG14+ shows wait time from pg_locks.waitstart) |
| ID | Report |
|---|---|
| b1 | Table bloat estimation |
| b2 | B-tree index bloat estimation |
| b3 | Table bloat via pgstattuple (expensive) |
| b4 | B-tree index bloat via pgstattuple (expensive) |
| b5 | Tables and columns without stats (bloat cannot be estimated) |
| ID | Report |
|---|---|
| i1 | Unused and rarely used indexes |
| i2 | Redundant indexes |
| i3 | Foreign keys with missing indexes |
| i4 | Invalid indexes |
| i5 | Index cleanup migration DDL (DO & UNDO) |
| ID | Report |
|---|---|
| v1 | Vacuum: current activity |
| v2 | Autovacuum progress and queue |
| c1 | Index creation/reindex progress |
| ID | Report |
|---|---|
| s1 | Slowest queries by total time (requires pg_stat_statements) |
| s2 | Slowest queries report (requires pg_stat_statements) |
| ID | Report |
|---|---|
| t1 | Postgres parameters tuning |
| e1 | Installed extensions |
| p1 | Alignment padding analysis (experimental) |
| r1 | Create user with random password (interactive) |
| r2 | Alter user with random password (interactive) |
Tested with PostgreSQL 13 through 18. Older versions (9.6-12) may work but are not actively tested. Some reports require features from newer versions (noted in the report headers).
Add a .sql file to the sql/ directory. The filename format is <id>_<name>.sql (e.g., f1_my_query.sql). The first line must be an SQL comment (--) with the report description — it appears in the menu automatically.
Then regenerate the menu:
bash ./init/generate.shpspg makes tabular psql output much easier to read. After installing, add to ~/.psqlrc:
\setenv PAGER pspg
\pset border 2
\pset linestyle unicode
Based on queries by many contributors, including:
- ioguix (bloat estimation)
- Data Egret (Lesovsky, Ermakov, Boguk, Kosmodemiansky et al.)
- PostgreSQL Experts (Berkus, Weaver et al.)
Questions or ideas: nik@postgres.ai (Nikolay Samokhvalov), or open an issue.