[-] rewrite "Custom installation" tutorial, fixes #602

Author: pashagolub

docs/concept/components.md

@@ -2,6 +2,9 @@
 title: Components
 ---
 
+
+## Component diagram
+
 The main development idea around pgwatch was to do the minimal work
 needed and not to reinvent the wheel - meaning that pgwatch is mostly
 just about gluing together already some proven pieces of software for
@@ -11,6 +14,14 @@ pgwatch metrics collector. Note that most components are not mandatory
 and for tasks like metrics storage there are many components to choose
 from.
 
+[![pgwatch typical deployment architecture diagram](../gallery/pgwatch_architecture.jpg)](../gallery/pgwatch_architecture.jpg)
+
+All components are *loosely coupled*, thus for non-pgwatch components
+(pgwatch components are only the metrics collector)
+you can decide to make use of an already existing installation of
+Postgres, Grafana or Prometheus and run additionally just the pgwatch
+collector.
+
 ## The metrics gathering daemon
 
 The metrics collector, written in Go, is the only mandatory and most
@@ -89,12 +100,3 @@ always want to customize some aspects though, so it's not meant as a
 one-size-fits-all solution. Also as metrics are stored in a DB, they can
 be visualized or processed in any other way.
 
-## Component diagram
-
-[![pgwatch typical deployment architecture diagram](../gallery/pgwatch_architecture.jpg)](../gallery/pgwatch_architecture.jpg)
-
-All components are *loosely coupled*, thus for non-pgwatch components
-(pgwatch components are only the metrics collector)
-you can decide to make use of an already existing installation of
-Postgres, Grafana or Prometheus and run additionally just the pgwatch
-collector.

docs/concept/installation_options.md

@@ -9,7 +9,7 @@ also choose how is the monitoring configuration
 ## Configuration database based operation
 
 This is the original central pull mode depicted on the
-[architecture diagram](../gallery/pgwatch_architecture.png). 
+[architecture diagram](components.md#component-diagram).
 It requires a small schema to be rolled out on any Postgres
 database accessible to the metrics gathering daemon, which will hold the
 connect strings, metric definition SQLs and preset configurations and
@@ -24,10 +24,10 @@ One can deploy the gatherer daemon(s) decentralized with
 *sources to be monitored* defined in simple YAML files. In that case there
 is no need for the central Postgres configuration database. See the
 [sample.sources.yaml](https://github.com/cybertec-postgresql/pgwatch/blob/master/internal/sources/sample.sources.yaml)
-config file for an example. 
+config file for an example.
 
 !!! Note
-    In this mode you also may want, but not forced, to point out the path to 
+    In this mode you also may want, but not forced, to point out the path to
     metric definition YAML file when starting the
     gatherer. Also note that the configuration system supports multiple
     YAML files in a folder so that you could easily programmatically manage

docs/index.md

@@ -14,10 +14,10 @@ In the world of database management, **monitoring** plays a crucial role in ensu
 
 pgwatch is designed specifically for monitoring **PostgreSQL databases** and related infrastructure. It covers a wide range of components crucial for PostgreSQL ecosystems, including:
 
-  - **PostgreSQL Databases**: pgwatch monitors the core performance and health metrics of your PostgreSQL instances.
-  - **Patroni Clusters**: Monitor the health and performance of high-availability clusters members managed by **Patroni**.
-  - **Connection Poolers (PgPool, PgBouncer)**: pgwatch provides insights into connection pooling with both **PgPool** and **PgBouncer**.
-  - **Backup solutions**: Track the performance and status of **PgBackRest** and **WAL-G** backups, ensuring that your backups are executed correctly.
+- **PostgreSQL Databases**: pgwatch monitors the core performance and health metrics of your PostgreSQL instances.
+- **Patroni Clusters**: Monitor the health and performance of high-availability clusters members managed by **Patroni**.
+- **Connection Poolers (PgPool, PgBouncer)**: pgwatch provides insights into connection pooling with both **PgPool** and **PgBouncer**.
+- **Backup solutions**: Track the performance and status of **PgBackRest** and **WAL-G** backups, ensuring that your backups are executed correctly.
 
 This extended monitoring capability allows you to gain a comprehensive view of not only your PostgreSQL databases but also the surrounding infrastructure that supports and enhances your database operations.
 
@@ -27,12 +27,12 @@ This extended monitoring capability allows you to gain a comprehensive view of n
 
 pgwatch provides out-of-the-box support for almost all essential **PostgreSQL metrics**, including:
 
-  - Database health checks
-  - Query performance
-  - Index usage
-  - Disk I/O
-  - CPU and memory consumption
-  - Locks, waits, and more
+- Database health checks
+- Query performance
+- Index usage
+- Disk I/O
+- CPU and memory consumption
+- Locks, waits, and more
 
 In addition to the standard metrics, pgwatch can be easily extended to monitor **custom metrics** based on your specific needs. The solution offers flexibility to fine-tune monitoring details and the aggressiveness of data collection.