PostgreSQL · MySQL · MariaDB · SQL Server · Oracle
ReinDB learns what normal looks like for each database, flags what drifts from it and shows the evidence. When a statement regresses, it writes the fix as a script your DBA can test.
Your first database is free, no card needed. Already have an account? Sign in to ReinDB →
DBA tuning tip
Before you tune
Ask what changed before you change anything. A deploy, a data load, fresh statistics, a new parameter: most regressions have a date.
Guide §1.4From our free 109-page guide, Rein In Your Database, for PostgreSQL, MySQL, SQL Server and Oracle.Download the guide PDF
Ask, Now, Drift, Baselines, Environment, Reports and Pre-check, each a tab in the app, with the question bar on every one. Pick a tab to see it on the built-in demo database (synthetic data).
Hourly measurements per statement, compared with a 14-day median for the same hour of day. Regressions, new heavy statements, execution spikes, plan changes (with pg_stat_monitor) and database-wide drift, each NEW, ONGOING or RESOLVED.
monitor.py · exit 4 = alert“Load over the last 3 days”, “WAL since yesterday”, “what flags fired this week?”, or a statement id. Ask from any page. Built-in rules or your approved model read the question, a badge says which, and the chart and caption always come from the data.
shareable answer linksMissing indexes found in the plan, temp spills, disk-bound reads, rows examined per row returned, N+1 query patterns, planning and JIT overhead, and unstable run times.
Analyze button · analyze_statementCREATE INDEX CONCURRENTLY, ALTER TABLE … ALGORITHM=INPLACE, LOCK=NONE, WITH (ONLINE = ON), forcing a good plan back with Query Store, per-role settings, invisible indexes before drops. Every line commented out, every step with its undo. Never runs by itself.
Cache hit ratio, dead rows and vacuum, never-analyzed tables, unused indexes, forced checkpoints, buffer pool misses, redo capacity, connection headroom, checked against this database’s own numbers.
database_fix.sqlThe connection test lists every statistic ReinDB reads as OK, missing or optional, with the exact command, my.cnf line or parameter-group setting to enable it. A grants script for each engine.
grants_postgres · mysql · sqlserver · oracle.sqlOn a managed database a fix never says ALTER SYSTEM. It names the parameter group setting to change, with the console steps, the aws rds command and whether it needs a reboot. Settings Aurora manages itself are marked as such.
One click reads the database this moment: running and idle-in-transaction sessions, who blocks whom, what sessions wait on and replica lag. Cheap read-only queries with a 5-second limit; owners only on real databases.
Live now · live_nowWho waited on whose lock, on which table, for how long, and the session holding it. Each deadlock with its tables, both statements and the one that was rolled back, from the PostgreSQL log, InnoDB status, SQL Server’s system_health session or Oracle’s trace files.
blocking · deadlocksReplica lag and replication slots holding WAL, transactions left open or idle, wait events against their usual for the hour, and when the disk fills at the current growth.
replication · long_running · storageToday against yesterday or a week ago, this week against last, in one click. Or any range from 6 hours to 30 days against the usual for those hours, or a saved baseline like “before release 4.2”, statement by statement. Copy a link to share the exact view.
Baselines pageWorkload, Top SQL, waits, I/O, capacity and stored code as standard reports, or mix them into your own (a morning check, a release check) with its range and baseline, saved for the team. Suggestions start you off, the ones that fit this week’s findings first.
Reports · My reportsMark deploys, maintenance and config changes on the page, or from your deploy script or CI job with one curl. Each shows on every chart, so a change in the numbers sits next to its likely cause.
/api/events/pushEach database’s week against the week before: load, slowest statements, waits, blocking, flags and the week’s events, with the full report attached as a page that prints to PDF.
weekly report emailEighteen read-only tools over MCP or REST with an OpenAPI description, input validation, literal redaction and an audit log of every call.
mcp_server.pyThe same pages, questions, flags and fix scripts on every engine. ReinDB reads the statistics each engine already keeps, with the smallest read-only grant it offers, on your own servers or on Amazon RDS and Aurora.
Time, buffer and disk reads, temp spill and WAL per statement. Uses pg_stat_monitor, pg_wait_sampling, pg_qualstats and track_io_timing when they are there. On Aurora, where WAL lives in AWS’s storage layer, ReinDB points to CloudWatch for it.
Statement digests with rows examined against rows returned, full scans and joins, on-disk temp tables and lock time. Explains a sample statement and reads index usage and waits when enabled.
Statement statistics from the plan cache, waits, sessions, locks and index usage, with Query Store for plan history: when a plan regresses, the fix script forces the good one back. Deadlocks with both statements from the system_health session. Nothing to install or restart.
V$SQLSTATS, waits, sessions and the DBA_ views, snapshotted by ReinDB itself, so no AWR and no Diagnostics Pack. If you license the Pack, tick it and ReinDB reads ASH and AWR too; it never assumes. Connects without an Oracle Client.
Measure, compare, flag, explain, fix.
A read-only user and a schedule are all it needs. ReinDB installs nothing in the database, and the website and agent tools read the stored history; only the hourly monitor and the Analyze button connect, read-only.
Differences between hourly reads of each engine’s statement statistics (pg_stat_statements, performance_schema, SQL Server’s plan cache, Oracle’s V$SQLSTATS), plus database-wide counters. Restarts and statistics resets detected.
Median and spread of each measure over 14 days, by hour of day, so the 9 AM batch peak is normal, not an alert.
Regressions, new heavy statements, spikes, plan changes and database-wide drift, sized by the extra time they cost.
Plain-English captions and answers computed from the measurements. A model, if you use one, only chooses what to show.
Statement and plan analysis of the worst regressions and a step-by-step test script with undo, for change control.
Read-only sessions with statement time limits, and the smallest grants each engine offers (pg_monitor; SELECT on performance_schema; VIEW SERVER PERFORMANCE STATE; SELECT on the V$ and DBA_ views). TLS with certificate verification by default. Literals in SQL are redacted before anything reaches a model. Every question and tool call is written to an audit log. Hosted sign-in uses scrypt password hashing, lockouts, single-use reset links and encrypted database credentials. If ReinDB may not connect at all, a one-file Python collector runs on your side instead: it runs only read-only statements, sends the raw numbers over HTTPS, and holds none of ReinDB’s analysis.
The question bar is on every page. Every answer shows how the question was understood, whether built-in rules or a model read it, and is computed from stored numbers, so a model cannot invent a figure. Answers are links: send one to whoever needs to see it.
Developers paste a query, a stored procedure or a schema change and get Pass, Review or Risky before it reaches production. ReinDB finds the production statement it replaces and how often that runs, plans the new code with EXPLAIN, checks it for full scans, missing indexes and code that runs row by row, flags schema changes that lock busy tables, and names the quietest hour to deploy. Pasted code is never run: only EXPLAIN, without ANALYZE, in a read-only transaction.
ReinDB installs from the browser on iPhone, Android and desktops, with no app store. Turn on alerts and a new regression arrives as a notification that names the database, never the SQL or the numbers; tap it and the chart opens.
Hourly by default, or every 30 or 15 minutes per database, and always compared with the same hour on earlier days. Monday 9 AM is judged against other 9 AMs, not against 3 AM. You control when it stops and when it stays quiet.
Time and reads for every statement (with temp spill and WAL where the engine records them), database-wide counters including log and redo volume, table activity, waits and size. Each run measures the time since the last one, flags what drifted, and analyzes the worst statements. Shorter intervals show more detail in a 6-hour view and alert sooner.
Who is blocking whom, what sessions wait on, replica lag and long or idle-in-transaction sessions. One cheap query each, so short lock waits between hourly runs are not missed. Optional.
One click reads the database at that moment, with a 5-second limit. For live connections; a collector never lets ReinDB connect.
Pause a database during maintenance or before retiring it; its history is kept. Quiet hours hold alerts overnight or at weekends and send one summary when they end. Measuring never stops for quiet hours, so the baseline stays whole.
Add a database with the read-only user and ReinDB tests the connection and every statistic it reads before saving. Within a week it knows what normal looks like for each hour of the day.
The same ReinDB three ways. Ask about onsite and in-cloud
Nothing to install. ReinDB connects to your database over TLS with a read-only user, from one fixed address you allow through the firewall or security group. Live now reads it on demand. Your first database is free; Starter is $49/month. See pricing.
For databases that never leave your network. One Linux or Windows server or VM inside your network runs the website and the hourly checks; statistics and history stay on it. Annual license.
Deployed into your own AWS or Google Cloud account from our template: scheduled containers beside Amazon RDS and Aurora, or Cloud SQL and AlloyDB. Passwords stay in your secrets manager; nothing is opened to Bandlei.
Tested on Amazon RDS for PostgreSQL, MySQL, SQL Server and Oracle, and on Aurora PostgreSQL. ReinDB uses the services’ own statistics, and its fix scripts name parameter group changes. Google Cloud SQL and Azure use the same statistics and are next to test.
The website, questions and flagging are engine-independent; each engine needs only a collector.
Start free, no card needed. Hosted plans are billed monthly. Onsite and in-your-cloud editions are licensed annually.
Your first database, for one person, with no time limit.
Every check, drift flag and fix script, plain-English questions, reports and the weekly email. Phone alerts, teammates and more databases come with Starter. New accounts start with 14 days of Starter.
Free is offered while we have room for it. If demand outgrows our capacity we may pause new Free sign-ups, or limit what Free includes after at least 30 days’ notice by email.
Start freeFor one person looking after their own databases. Up to 3 databases, then $8/month for each extra one, up to 10.
Everything in Free, plus phone and email alerts, teammates, a Fleet page with every database’s status, version and uptime (down and degraded ones flagged in red), and a Runs page with every collection and its timing. Upgrade from Plan & billing in the app at any time.
Try Starter freeFor teams. 10 databases included, then $12/month for each extra one, up to 25, for up to 10 people.
Everything in Starter, plus database groups (prod, staging, by customer) with group comparisons over time, alerts to Slack, Microsoft Teams, PagerDuty and Opsgenie, a Jira issue from any flag with its fix script attached, Pre-check as a step in your build pipeline, scheduled exports for Grafana and other tools, database import from Excel, webhooks and audit export.
Join the Pro listFor larger estates and networks that stay closed: hosted, onsite, in your own cloud or air-gapped, on an annual license with support.
More databases and people than Pro, with terms that fit how you buy.
Talk to usPartners and managed service providers: ask about partner terms. A free account nobody signs in to for 90 days is deleted, after two reminder emails; see the terms.
ReinDB is open: create a free account and your first database is free. For a guided evaluation, Pro, Enterprise, onsite or in-your-cloud editions, leave your details and we will get back to you within a business day.
ReinQ, plain-English questions over your own data, is still in an invite-only beta. It’s free during the beta, for up to 100 questions a day per organization.
Ask for a guided evaluation. All it takes is one read-only database user.
