System and storage inspection
Applies to: Rubian 4.0.2; available observations depend on connection and host platform
Updated
Rubian offers a session summary, connected-system observations and host inspection tools. Begin with the summary so the scope of later measurements is clear.
system_status
status is an alias. The summary prints the product version, current location, connection mode and discovered inventory counts. version also displays version information. These observations provide context for inspection; they do not establish that a particular workload has finished successfully.
Memory and process observations#
| Command | What it reports |
|---|---|
free |
Connected-system memory information when available; host memory-tool output in host mode. |
sysmon |
The supplied system monitor's summary. Requires the optional monitor. |
sysmon_top(n = 10) |
The monitor's selected process observations. |
sysmon_snap(n = 12) |
A structured monitor snapshot with load, memory and process information. |
uptime(quiet: false) |
Connected-environment uptime information, or host-tool information in host mode. |
kernel_health |
A diagnostic display of available connected-system observations. |
proc_tree |
A task hierarchy when the connected environment supplies it; otherwise a limited process-ID display or a missing-data message. |
kernel_health is intended for reading. It does not return a structured health verdict, and its concluding display does not prove that every individual observation succeeded. Retain missing or failed observations when investigating a problem.
On supported hosts, vmstat, iostat, top and htop provide additional host-tool views. The optional monitor also offers sysmon_local for its local mode and administrator setup through sysmon_build where supplied. Use the environment's setup procedure before changing the selected monitor mode. Taskman covers process selection and termination.
Storage and mounted locations#
In a connected environment, mount and df can display available managed mounts and modules. This output is an inventory; do not read it as a conventional disk-capacity report. In host mode, the corresponding host tools provide their host-specific information.
du(path = nil, depth: 1) provides a bounded view of host directory sizes. Its managed-file observations should not be used as a capacity guarantee. fs_storage_stats(path) reports the available storage observations for a managed location in a connected environment; fields depend on the supplied storage configuration.
sync(path = nil) requests that connected storage writes be flushed, optionally for a selected path. A flush is not a backup or recovery test.
Read logs#
| Command | Use |
|---|---|
log_tail(source = 'kernel', lines: 30) |
Read the selected known environment log. |
tail_log(path = nil, n: 50, follow: false) |
Read a host log; follow: true continues watching through the host utility. |
journalctl(*arguments) |
Read the host system journal where available. With no arguments it requests recent diagnostic context without a pager. |
dmesg(lines: 50) |
Display recent connected-system messages or host kernel messages. The connected form can return more text than the displayed line limit. |
Select a small relevant time window or line count. Keep the timestamp, failing operation and corresponding error together. Logs may contain application data and credentials; redact a support extract rather than publishing an entire log.
Built-in environment checks#
rtest(filter = nil) runs the supplied built-in environment checks and returns a summary with total, passed, failed and individual results. rtest_report returns the last summary for the current session. The filter selects check names using a regular expression.
These checks can create project files, append notes, send test traffic and start temporary work. Run them in the designated evaluation or support environment. They test the supplied built-in cases; they are not an automatic test suite for every project you create. A report-persistence error also needs attention even if individual checks passed.
Administrative storage operations#
unmount(path) can attempt a managed unmount and fall back to a privileged host unmount. Confirm the resource and its users before invoking it. mkfs(type, device) invokes host filesystem creation, asks for an explicit confirmation and can erase existing data on the selected device. These commands belong in a provisioned storage-administration procedure, not a routine diagnostic sequence.
Use the relevant host or release-specific instructions for capacity planning, device formatting and recovery. This reference does not qualify a storage device, filesystem or recovery procedure simply because its command is available.