Add docs for exported open telemetry trace files

public/docs/img/support/images/download-trace-files.png.json

@@ -0,0 +1 @@
+{"width":1730,"height":1383,"updated":"2026-02-23T11:13:48.981Z"}

public/docs/img/support/images/trace-file-export-configuration.png.json

@@ -0,0 +1 @@
+{"width":1709,"height":1123,"updated":"2026-02-23T11:13:49.011Z"}

src/pages/docs/support/opentelemetry-trace-files.md

@@ -0,0 +1,90 @@
+---
+layout: src/layouts/Default.astro
+pubDate: 2026-02-23
+modDate: 2026-02-23
+title: OpenTelemetry trace files
+description: OpenTelemetry trace files saved to disk by Octopus Server for diagnostic purposes.
+navOrder: 15
+---
+
+Octopus Server records [OpenTelemetry](https://opentelemetry.io/) (OTEL) traces that capture internal operations like HTTP requests, task execution, and more. These traces are saved as `.tar` files to disk and can be sent to Octopus support to help diagnose issues.
+
+:::div{.hint}
+**OpenTelemetry trace files are only available for self-hosted instances of Octopus Server.**
+
+This feature is available from 2026.1. It is disabled by default and can be enabled at **Configuration ➜ Diagnostics**.
+:::
+
+## Permissions
+
+Viewing and changing the OpenTelemetry trace files configuration, and downloading or deleting trace files, requires the [`AdministerSystem`](/docs/security/users-and-teams/default-permissions) permission.
+
+## Enabling and configuring
+
+1. Navigate to **Configuration ➜ Diagnostics**.
+2. Under **Server Traces**, click `Configure`.
+3. Toggle **Enabled**.
+4. Optionally configure **Max storage size** and **Retention days**.
+5. Click `Save`.
+
+:::figure
+![OpenTelemetry trace files configuration page](/docs/img/support/images/trace-file-export-configuration.png)
+:::
+
+:::div{.hint}
+Configuration changes take effect within about 1 minute as the server syncs settings in the background.
+:::
+
+## What traces contain
+
+Traces are made up of *spans*, each representing a unit of work performed by the server. The spans captured include:
+
+- **HTTP requests** - inbound and outbound HTTP requests with timing and status information.
+- **Task execution** - deployments, runbook runs, and other server tasks.
+- **Internal operations** - other server-side work that provides context when diagnosing problems.
+
+Each span includes attributes such as timing, status codes, and contextual metadata that help Octopus support engineers understand what the server was doing and where time was spent.
+
+Trace data is stored in OTLP JSON format, which is compatible with the open source [OpenTelemetry Collector](https://opentelemetry.io/docs/collector/) after decompression.
+
+## File location
+
+Trace files are stored under the `Telemetry/OpenTelemetry/traces/` subdirectory of the server's **ClusterSharedDirectory** as `.tar` files. ClusterSharedDirectory is the shared storage location Octopus uses for logs and diagnostics data - its path can be found at [**Configuration ➜ Settings ➜ Server Folders**](/docs/administration/managing-infrastructure/server-configuration-and-file-storage/#server-folders) (or configured via the [`path` command](/docs/octopus-rest-api/octopus.server.exe-command-line/path)), and defaults to the server's Home Directory if not explicitly set.
+
+In a [High Availability (HA)](/docs/administration/high-availability) cluster, all nodes write to this shared path, so trace files from every node are available in one location.
+
+## Retention and disk usage
+
+Traces are written to disk in 50 MB chunks. The configurable size limit controls how many of these chunk files will be written. When the total size approaches the configured size limit, the oldest trace files are deleted first to make room. The minimum configurable storage size is **250 MB**.
+
+Retention by age is **unlimited by default**, but you can configure a maximum number of days to keep trace files. Size-based and time-based retention work together - files are removed when they exceed either limit.
+
+If a node crashes or restarts mid-write, any `.tar.inprogress` files left behind are automatically recovered after 10 minutes of inactivity and treated as completed trace files.
+
+## Downloading trace files
+
+Trace files can be downloaded directly from the Diagnostics page:
+
+1. Navigate to **Configuration ➜ Diagnostics**.
+2. Under **Server Traces**, click `Download` to save all current trace files as a single `.tar` archive.
+
+:::figure
+![Download button for OpenTelemetry trace files on the Diagnostics page](/docs/img/support/images/download-trace-files.png)
+:::
+
+:::div{.hint}
+Downloading is available even while trace export is actively running - the download reads safely alongside the active writer.
+:::
+
+## Deleting trace files
+
+To free disk space without waiting for retention limits to apply:
+
+1. Navigate to **Configuration ➜ Diagnostics**.
+2. Under **Server Traces**, click `Configure`.
+3. Disable the feature and click `Save`.
+4. Click the `Clear trace files` button.
+
+## Sending trace files to support
+
+If Octopus support needs your trace files to help diagnose an issue, they will guide you through how to provide them.