Merge pull request #303 from HOSTED-POWER/feature/monitoring

Feature/monitoring

Author: sanderdequick

.gitignore

@@ -1 +1,11 @@
 .idea
+
+# AI assistant tooling — kept local, never committed
+# (memory/guidance files, downloaded skills, agent scripts/config)
+CLAUDE.md
+AGENTS.md
+.claude/
+.cursor/
+.cursorrules
+.aider*
+.github/copilot-instructions.md

Support/monitoring.md

@@ -11,6 +11,8 @@ At **Hosted Power**, we recognize the paramount importance of application availa
 
 Should any threat arise, our skilled engineers are on standby to swiftly intervene, safeguarding your operations seamlessly.
 
+You can follow the same checks yourself in the TurboStack platform: the [Monitoring dashboard](../TurboStack%20Platform/monitoring.md) gives a fleet-wide overview of every open alert, while the per-host [Health tab](../TurboStack%20Platform/Hosts/health.md) shows the detailed status of an individual server.
+
 ## Alerting
 
 In critical situations , Hosted Power activates its On Call procedure. Our on call team is on standby 24/7, ensuring swift manual investigation and resolution of issues, regardless of office hours. If the primary On Call colleague doesn’t respond within 30 minutes, an alert automatically triggers escalation to the entire on call team, engaging our Operations engineers on call "pool" until someone acknowledges the issue en starts to research.

TurboStack Platform/Hosts/health.md

@@ -0,0 +1,78 @@
+---
+order: 50
+icon: pulse
+---
+
+# Health - server health and monitoring overview
+
+The **Health** tab is your **first stop** when opening a host. It provides a **real-time overview** of the server's condition, surfaces the most important **alerts**, and gives you direct access to detailed monitoring **reports** and **services**.
+
+![Health overview|400](img/ts_health_overview.png)
+
+Unlike the fleet-wide [Monitoring dashboard](../monitoring.md), the Health tab is fully **scoped to a single host**, giving you **deep insight** into how this specific server and its applications are behaving right now and over time.
+
+What makes the Health tab particularly powerful is that it does not just **show the status** of your server - it actively helps you **resolve issues**. For every entry in **Top Issues** and every report under **Latest Reports**, the TurboStack platform provides a **detailed analysis** of the underlying problem, together with **concrete, context-aware recommendations** on how to fix it. This turns the Health tab from a passive status page into an **actionable troubleshooting tool**, helping you move from *"something is wrong"* to *"here is exactly what to do about it"* in a single click.
+
+This page describes every section of the Health tab and how to use it to keep your infrastructure running smoothly.
+
+## Host Monitoring
+
+The **Host Monitoring** panel, located on the top right, displays the most important **resource metrics** of your server at a glance:
+
+- **CPU Usage** - the current CPU load, accompanied by a graph showing the `load1` average over time.
+- **RAM Usage** - the current memory consumption, with a detailed breakdown of used, free, cached, warning and critical thresholds.
+- **Disk Usage** - the current disk consumption on the root filesystem, plotted against warning and critical thresholds.
+
+Each metric is shown as a percentage, combined with a colour-coded **status indicator** (OK, Warning, Critical) so you can immediately spot any resource that needs attention.
+
+![Health metrics|400](img/ts_health_metrics.png)
+
+### Timeframe selector
+
+In the top right corner of the Host Monitoring panel, you can switch the timeframe of the graphs between **1H**, **8H**, **1D** and **7D**. This allows you to quickly **zoom in** on a recent spike or **zoom out** to identify longer-term trends in resource usage.
+
+!!! info
+The Host Monitoring graphs are based on the same data used by our 24/7 monitoring system. If a metric crosses a critical threshold, our team will be alerted automatically.
+!!!
+
+## Top Issues
+
+The **Top Issues** panel, on the top left, lists the **most important active alerts** on your server right now. Each issue is tagged with a **category** (e.g. `AVAILABILITY`, `STABILITY`, `PERFORMANCE`) and a **status** (e.g. `Open`).
+
+![Health top issues|400](img/ts_health_topissues.png)
+
+Clicking an issue **expands** it, revealing:
+
+1. A **detailed description** of the problem, including the affected service and the impact on your environment.
+2. A **proposed solution** suggested by the TurboStack platform, based on the type of issue and the server's configuration.
+
+This makes it easy to understand **what is wrong**, **why it matters**, and **what to do next**, without having to dig through logs or external dashboards.
+
+## Latest Reports
+
+Below the Top Issues panel, the **Latest Reports** section provides a chronological list of **detailed health reports** generated for the server and its applications.
+
+![Health reports|400](img/ts_health_reports.png)
+
+Each report is linked to a specific check (e.g. `rabbitmq-online`, `postgres-online`, `website`) and includes the **severity level** (e.g. `P1 CRITICAL`), the **affected host or URL** and the **time the report was generated**.
+
+Click **Open report** on any entry to view the full report, including timestamps, related metrics and historical context. This is particularly useful when investigating **incidents** or preparing a **post-mortem**.
+
+## Services
+
+At the bottom of the Health tab, the **Services** panel shows a complete list of every **monitored service** on the server, together with its current **status** and the **latest check output**.
+
+![Health services|400](img/ts_health_services.png)
+
+For each service you will see:
+
+- The **service name** (e.g. `postgres-online`, `ping4`, `disk /`, `smtp`, `mysql-uptime`).
+- The **status** (`OK`, `WARNING`, `CRITICAL`, `UNKNOWN`).
+- The **plugin output**, which contains the raw result of the check (e.g. response times, free disk space, error messages).
+- The **timestamp** of the most recent check.
+
+This view gives you an exhaustive overview of **everything we monitor** on your server, making it easy to confirm that all services are healthy or to drill down into a specific check that is failing.
+
+!!! info
+If a service is in an `UNKNOWN` or `CRITICAL` state and you are not sure how to resolve it, please contact our [Support](../../Support/standard_support.md) team. Our engineers have access to the same monitoring data and can intervene 24/7. See our [24/7 monitoring and alerting](../../Support/monitoring.md) overview for how this works.
+!!!

TurboStack Platform/Hosts/img/ts_health_metrics.png

@@ -0,0 +1,59 @@
+---
+order: 85
+icon: eye
+---
+# Monitoring
+
+Hosted Power **proactively monitors your infrastructure**, tracking both **server and service availability** and **resource usage** around the clock. As soon as a critical threshold is crossed - high load, low disk space, a failing service or an unreachable website - an alert is raised so the issue can be addressed before it impacts your environment.
+
+On top of automated monitoring, our engineers are available **24/7 via phone and ticket** to investigate alerts, intervene where needed and assist with any infrastructure or application issue. For more information on how to reach us, or to **open a support ticket** or **call our support line**, see the [Support](../Support/standard_support.md) page. To learn how our team acts on these alerts around the clock, see our [24/7 monitoring and alerting](../Support/monitoring.md) overview.
+
+## Dashboard
+
+The **Monitoring** page in the left-hand menu features a dedicated **Dashboard** that provides a **fleet-wide overview** of every **open alert** across all your servers and applications. Only checks in a non-OK state are shown, so the dashboard always reflects the **issues that still need attention**. For a detailed, per-host view of these alerts, resource metrics and reports, open the [Health tab](Hosts/health.md) of an individual host.
+
+![Monitoring dashboard|400](../img/turbostackapp/monitoring/ts_monitoring_dashboard.png)
+
+Alerts on the dashboard are grouped into **four categories**, each focused on a specific type of problem:
+
+### Alerts
+
+The **Alerts** panel lists the most important **server and application checks** that are currently in a non-OK state. Entries are **sorted by criticality**, first by **priority** (from **P1** - most critical - to **P4** - least critical), and within each priority by **status**: `CRITICAL` first, then `WARNING`, and finally `UNKNOWN`.
+
+This ordering ensures that the **most urgent issues** are always at the top of the list, so you know exactly **where to start** when multiple alerts are open at once.
+
+![Alerts panel|400](../img/turbostackapp/monitoring/ts_monitoring_alerts.png)
+
+### Mail Problems
+
+The **Mail Problems** panel highlights **issues related to mail applications** and their **deliverability**. This includes problems with the local mail queue (`mailq`), blocked or filtered messages, SPF or DNS-related warnings, and other indicators that mail is not flowing as expected.
+
+### Security Issues
+
+The **Security Issues** panel surfaces hosts where **security or system packages have not yet been installed or updated**. A typical example is an `apt` check reporting available **critical updates** that still need to be applied to keep the server secure.
+
+### Other
+
+The **Other** panel collects **remaining issues** that are still worth reviewing, but are **less critical and less urgent** than the items in the other categories. Use this section to keep an eye on lower-priority signals that should not be ignored, but do not require immediate intervention.
+
+!!! info
+Clicking any alert on the dashboard takes you straight to the **detailed view of that alert** on the specific server it belongs to, so you can investigate the root cause without having to navigate through the host configuration manually.
+!!!
+
+## Alert priorities
+
+Every alert is assigned a **priority** from **P1** to **P4**, indicating how urgently it needs to be addressed. This priority determines both the ordering on the dashboard and how quickly our team intervenes.
+
+- **P1 - Critical** - something is down or broken that is impacting the server and its applications. This requires **immediate action**.
+- **P2 - Very urgent** - something is down, broken or degraded that will **potentially cause a P1** if no action is taken within **2 hours**.
+- **P3 - Urgent** - something is wrong that will **potentially affect a P2 or escalate to a P1** if no action is taken within **8 hours**.
+- **P4 - Not urgent** - an issue that is best reviewed to keep the server **stable over the long term**, but does not require immediate intervention.
+
+## Website monitoring URL
+
+The [TurboStack Platform](https://my.turbostack.app/ "TurboStack Platform") is also able to **monitor a website URL** per application. In this case, the website URL will be added to our monitoring system
+and Hosted Power will **monitor the availability** of the website based on the URL, and **intervene 24/7** if an issue arises.
+
+Please follow the instructions below to activate the monitoring url on your server
+
+[!embed allowFullScreen="false"](https://player.vimeo.com/video/1056707252?title=0&byline=0&portrait=0&badge=0&autopause=0&player_id=0&app_id=58479)