What our engineers handle on PostgreSQL
This is the hands-on work, done by a person: the jobs the automatic tools cannot do for you.
- 1Tune memory, connections and autovacuum for the workload
- 2Set up backups with pg_dump, base backups and WAL archiving for point-in-time recovery
- 3Run streaming replication and test a failover
- 4Upgrade between major versions with pg_upgrade or logical replication
Problems we fix
Problem
Queries slow down over time and the tables grow
What we do
We check bloat and autovacuum, repack or vacuum where needed and set the autovacuum parameters so it does not return.
Problem
The disk fills with WAL files
What we do
We find the stuck replica or archive command, fix it, clean up safely and add monitoring for WAL growth.
Problem
Application connections run out
What we do
We add a connection pooler such as PgBouncer and set sensible per-application limits.
Automatic tools, and our team for the rest
RootManage is automatic server management with every tool in one place. The tools run by themselves on a schedule and report to one dashboard. Everything they cannot do, and every emergency, is handled by our engineers by hand.
Automatic
RootManage tools, running by themselves
- Server Health agent reports load, disk use and failed login attempts to your dashboard.
- Metrics agent draws CPU, memory, disk and network charts, sampled every five minutes.
- Uptime checks check the server every minute and email you when it stops answering, and again when it recovers.
- Backup agent runs scheduled backups at 02:00 IST to storage we provide or to your own SFTP or S3 box, with help restoring.
Manual, by our engineers
Everything else, and every emergency
- Hands-on work. Fixes, tuning, migrations, upgrades and the other jobs listed above on this page.
- Emergency work. A hacked site, an outage or a failed update is handled by a person who investigates and fixes it, not by a script.
- Things we do not automate. Anything the tools do not cover is done by hand, over SSH, WHM or RDP, on a login you can remove at any time.
The tools are optional, run on Linux servers and are installed by you with one command each. Every job our team does is recorded as a task in your dashboard with the details, and you get an email when it is updated.
How it works
- 1
Sign up
Create an account, choose a plan and add the server you want managed.
- 2
Install the tools
Run one command per tool. They then work by themselves and report to your dashboard. For hands-on work you give us a login you can remove at any time.
- 3
Automatic, with people behind it
The tools watch, scan and back up automatically and alert you. Our engineers do the rest by hand and handle every emergency, and you follow each task in your dashboard.
Questions about PostgreSQL
Can you restore the database to an earlier point in time?
Yes, if WAL archiving and base backups were running. We set that up so you can recover to just before a bad change.
Can you upgrade PostgreSQL without long downtime?
Yes. For big databases we use logical replication or pg_upgrade with links, rehearse on a copy and keep the old cluster until you are sure.
What does it cost to have PostgreSQL managed?
Managed plans start at ₹1,999 a month for one server, and PostgreSQL is managed as part of the server. The pricing page lists what each plan includes. If a server is already infected, the emergency cleanup is a one-time ₹9,000 job.