mirror of
https://github.com/supabase/supabase.git
synced 2026-10-05 09:25:06 +03:00
43 lines
2.7 KiB
Plaintext
43 lines
2.7 KiB
Plaintext
---
|
|
id: 'manual-replication-faq'
|
|
title: 'Manual replication FAQ'
|
|
description: 'Frequently asked questions about manual database replication.'
|
|
subtitle: 'Common questions and considerations when setting up manual replication.'
|
|
sidebar_label: 'FAQ'
|
|
---
|
|
|
|
## Which connection string should be used?
|
|
|
|
Use the direct connection string for logical replication. Connections through a pooler, such as Supavisor, will not work.
|
|
|
|
## The tool in use does not support IPv6
|
|
|
|
You can enable the [IPv4 add-on](/docs/guides/platform/ipv4-address) for your project.
|
|
|
|
## Can replication be configured in the Dashboard?
|
|
|
|
You can inspect [publications](/dashboard/project/_/database/publications) and change their published operations and table selection in the Dashboard. Use the [SQL Editor](/dashboard/project/_/sql/new), a CLI, or your consumer's setup flow for the remaining setup. Connect and operate the external consumer separately; the Dashboard doesn't manage a manual replication pipeline.
|
|
|
|
## How to configure database settings for replication?
|
|
|
|
Using the Supabase CLI, you can [configure database settings](/docs/guides/database/custom-postgres-config#cli-configurable-settings) to optimize them for your replication needs. These values can vary depending on your database size and activity.
|
|
|
|
## What are some important configuration options?
|
|
|
|
Some of the more important options to be aware of are:
|
|
|
|
- `max_wal_size`: Soft limit for WAL written between automatic checkpoints. Postgres can exceed it under some conditions.
|
|
- `max_slot_wal_keep_size`: Per-slot limit on retained WAL at checkpoint time. `-1` lets replication slots retain unlimited WAL. A finite limit can cause a lagging slot to lose required WAL.
|
|
- `wal_keep_size`: Minimum amount of old WAL retained for standby servers. It does not replace the per-slot limit.
|
|
- `max_wal_senders`: Maximum number of concurrent WAL sender processes for standby servers, logical replication consumers, and streaming backups.
|
|
|
|
Size WAL retention for the consumer's maximum expected lag or downtime and the database's WAL generation rate. Higher values reduce the risk that a consumer needs a new initial sync, but can use more database storage. Monitor each slot's retained WAL and `wal_status` instead of treating a configured limit as a guarantee.
|
|
|
|
## What is XMIN and should it be used?
|
|
|
|
Xmin is a different form of replication from logical replication and should only be used if logical replication is not available for your database, such as with older versions of Postgres.
|
|
|
|
Xmin performs replication by checking the [xmin system column](https://www.postgresql.org/docs/current/ddl-system-columns.html) and determining if that row has already been synchronized.
|
|
|
|
It does not capture deletion of data and is **not recommended**, particularly for larger databases.
|