Health

See a host's live health - the management overview, top issues, resource usage, and service checks.

The Health tab is a live overview of a host. It surfaces the most important problems first and shows real-time resource usage and the status of every monitored service. It is the tab a host opens on.

The Health tab showing the Management overview, the top issues and the host monitoring graphs

This tab covers:

Management overview - the decisions to take, in business terms
Top issues - ranked list of detected problems
Latest reports - investigation reports for this host
Host monitoring - CPU, memory, swap, and disk graphs
Services - status of every monitored service check

Management overview

The Management overview card at the top summarizes the host in business terms. It shows a status (Needs attention, Stable or Awaiting analysis), a short summary and decision cards. Each decision has an urgency: Now, 7 days, 30 days or Monitor. Where it applies, a card also carries an opportunity label, such as Capacity headroom, Continuity and SLA (Service Level Agreement), Performance review, Storage planning, Resilience or Efficiency saving. The action on a card jumps to the matching entry under Top issues and highlights it, so the technical evidence is one click away.

Before the first analysis, the card reads "Management summary pending the next server analysis." and points you to the technical findings below.

Refresh in the card header asks the platform for a new analysis of this host. The button reads Analysing... while it runs. The footer shows Last generated with the date and time of the analysis you are looking at, and how long ago that was. The same Refresh and Last generated appear on the Top issues card.

Top issues

A ranked list of the most important problems detected on the host. Each issue shows a severity (for example Critical, High, Moderate or Low, as the analysis reports it) and a category (PERFORMANCE / AVAILABILITY / SECURITY / STABILITY). Expand an issue to see what is happening, a recommended fix, and links to open the related investigation reports.

Tip

The list is ordered by importance, and not every issue is equally urgent. Start with CRITICAL and HIGH - these affect availability, performance or security right now. MEDIUM, LOW and INFO are worth knowing about but rarely need immediate action; handle them during normal maintenance.

When nothing has been detected, the panel is empty and reports that there are no top issues for the host.

The Top issues card of a host with detected problems

Latest reports

The most recent investigation reports for the host, newest first. Open a report to read its full details, or download the Portable Document Format (PDF) when available.

When you open a report, its address appears in the browser URL, so the report is bookmarkable and shareable. Use the Link button in the report header to copy that address to your clipboard (the button confirms with Copied). Anyone who opens the copied link, and who has access to the host, lands on the Health tab with the same report already open.

An open investigation report with the Link button that copies a shareable address

Tip

Paste the link into a support ticket or a message to a colleague so they see the exact report you are looking at, instead of describing it.

Host monitoring

The Host Monitoring panel has resource-usage cards for the four things that most often cause a slow or failing site. They appear in this order: CPU, RAM, Disk Swap Usage and Disk. Each card shows a current percentage, an OK / Warning / Critical status, and a graph. Use the 1H / 8H / 1D / 7D buttons to change the time range (1 hour to 7 days). The view refreshes automatically, and selecting a graph opens the alert detail page of that check.

Card	What it measures	What a high value means and what to do
CPU (Central Processing Unit)	How busy the server's processor is.	Sustained high CPU means the server is doing more work than it can keep up with, so pages slow down. Look for a traffic spike, a heavy task, or inefficient code. See High CPU and load.
RAM (Random Access Memory)	How much of the server's working memory is in use. The card also shows the extra capacity TurboRAM provides ("+X TurboRAM", the effective and physical memory and the compression ratio).	High memory use is normal up to a point. The risk is running out - see swap below.
Disk Swap Usage	Memory moved out of active use to the swap file on disk, the slower overflow behind TurboRAM.	Sustained use of the disk swap file means the server has run out of fast memory and is using storage instead. See Out of memory.
Disk	How full the server's storage is.	A full disk causes uploads, logs, databases and backups to fail. Free up space before it reaches 100%. See Disk space.

Tip

Use the 7-day (7D) view to tell a one-off spike from a steady trend. A single short spike is usually harmless; a line that climbs day after day needs attention.

Note

The RAM card can report an effective capacity larger than the server's physical RAM. That is TurboRAM compressing cold pages in memory, not a reporting error. When no data arrives, the panel reads "No monitoring data".

Services

A list of all monitored service checks for the host - for example the database, web server, cache and connectivity checks. Each check shows its status (OK, Warning, Critical or Unknown), the latest output from the check, and when it last ran. An Unknown status means the check could not determine the service state, for example because it did not return a result.

A check in Warning, Critical or Unknown state is the place to start when something is wrong: the output usually names the exact problem (for example a service that is not running, or a certificate about to expire). Many issues can be resolved from the command-line tool or by publishing a corrected configuration.

Tip

The Health tab covers one host. For an overview across all your hosts, use the Health page in the sidebar. For how monitoring works, see Monitoring (concepts).