mirror of
https://github.com/systemd/systemd.git
synced 2026-11-06 05:17:24 +00:00
Lennart Poettering
e0c648653b
This adds systemd-report-diskhealth, which reports the key health parameters of NVMe and ATA disks, plus basic identity information about all disks. For all hardware backed disks (including USB ones) size, logical and physical sector size are reported from sysfs, and model, vendor and serial from the udev database, preferring hwdb data, and falling back to the PCI or USB device if the block device itself carries no model/vendor information. For NVMe disks the SMART / Health Information log page is read via the NVMe admin command ioctl, and the critical warning bitmask, available spare, percentage used and media error counter are reported. For ATA disks with SMART enabled the overall SMART status is read via SG_IO ATA pass-through, and for rotating ones also the reallocated, pending and offline uncorrectable sector counts (SMART attributes 5, 197, 198), since various SSD vendors reused these attributes with incompatible semantics. Power-on hours and power cycle counts are reported for both under a common metric name, for ATA disks taken from the standardized Device Statistics log rather than from the vendor specific SMART attributes. ATA disks in standby are not queried, in order not to spin them up, and neither are ATA disks behind USB bridges. The temperature is deliberately not reported, since systemd-report-hwmon covers that already.
222 lines
10 KiB
XML
222 lines
10 KiB
XML
<?xml version="1.0"?> <!--*-nxml-*-->
|
|
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
|
|
"http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd">
|
|
<!-- SPDX-License-Identifier: LGPL-2.1-or-later -->
|
|
|
|
<refentry id="systemd-report-diskhealth_.service" xmlns:xi="http://www.w3.org/2001/XInclude"
|
|
conditional='ENABLE_REPORT'>
|
|
|
|
<refentryinfo>
|
|
<title>systemd-report-diskhealth@.service</title>
|
|
<productname>systemd</productname>
|
|
</refentryinfo>
|
|
|
|
<refmeta>
|
|
<refentrytitle>systemd-report-diskhealth@.service</refentrytitle>
|
|
<manvolnum>8</manvolnum>
|
|
</refmeta>
|
|
|
|
<refnamediv>
|
|
<refname>systemd-report-diskhealth@.service</refname>
|
|
<refname>systemd-report-diskhealth.socket</refname>
|
|
<refname>systemd-report-diskhealth</refname>
|
|
<refpurpose>Report disk identity and NVMe/ATA SMART health information as system report metrics</refpurpose>
|
|
</refnamediv>
|
|
|
|
<refsynopsisdiv>
|
|
<para><filename>systemd-report-diskhealth@.service</filename></para>
|
|
<para><filename>systemd-report-diskhealth.socket</filename></para>
|
|
<para><filename>/usr/lib/systemd/systemd-report-diskhealth</filename></para>
|
|
</refsynopsisdiv>
|
|
|
|
<refsect1>
|
|
<title>Description</title>
|
|
|
|
<para><command>systemd-report-diskhealth@.service</command> is a metrics source for
|
|
<citerefentry><refentrytitle>systemd-report</refentrytitle><manvolnum>1</manvolnum></citerefentry>. It
|
|
reports basic identity information about the disks of the system, as well as the key health parameters
|
|
of NVMe and ATA disks, by implementing the <constant>io.systemd.Metrics</constant> Varlink interface on a
|
|
socket linked into the <filename>/run/systemd/report/</filename> directory.</para>
|
|
|
|
<para>All whole-disk block devices backed by hardware are reported, i.e. partitions, loopback devices,
|
|
device mapper devices and other virtual block devices, as well as drives without a medium, are skipped.
|
|
Only block devices that have been fully probed by
|
|
<citerefentry><refentrytitle>systemd-udevd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>
|
|
are considered. The metric object is the device node of the block device, for example
|
|
<filename>/dev/nvme0n1</filename> or <filename>/dev/sda</filename>. No metric fields are set.</para>
|
|
|
|
<para>The following metrics are reported for all disks. The model, vendor and serial strings are taken
|
|
from the udev database, preferring information from the hardware database over the information the
|
|
device reports about itself. If the block device carries no model or vendor information, the one of the
|
|
PCI device of an NVMe disk or of the USB device of a USB disk is used instead. Metrics for which no
|
|
information is available are omitted.</para>
|
|
|
|
<variablelist>
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskLogicalSectorSizeBytes</literal></term>
|
|
<listitem><para>The logical sector size of the disk in bytes, i.e. the smallest unit the disk can
|
|
be addressed in.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskModel</literal></term>
|
|
<listitem><para>The model name of the disk.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskPhysicalSectorSizeBytes</literal></term>
|
|
<listitem><para>The physical sector size of the disk in bytes, i.e. the smallest unit the disk can
|
|
write without a read-modify-write cycle, as reported by the kernel.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskSerial</literal></term>
|
|
<listitem><para>The serial number of the disk.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskSizeBytes</literal></term>
|
|
<listitem><para>The size of the disk in bytes.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskVendor</literal></term>
|
|
<listitem><para>The vendor name of the disk. Note that ATA disks do not report a vendor, hence this
|
|
metric is typically not available for them.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
</variablelist>
|
|
|
|
<para>The following metrics are reported for NVMe disks and for ATA disks. For NVMe disks they are taken
|
|
from the SMART / Health Information log page, for ATA disks from the General Statistics page of the
|
|
Device Statistics log. ATA disks which do not implement the Device Statistics log do not report these
|
|
metrics.</para>
|
|
|
|
<variablelist>
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskPowerCycles</literal></term>
|
|
<listitem><para>The number of power cycles over the lifetime of the disk.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.DiskPowerOnHours</literal></term>
|
|
<listitem><para>The number of hours the disk has been powered on over its lifetime.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
</variablelist>
|
|
|
|
<para>The following metrics are reported for NVMe disks only, and are taken from the SMART / Health
|
|
Information log page of the controller. If multiple namespaces are attached to the same controller, the
|
|
same values are reported for all of them.</para>
|
|
|
|
<variablelist>
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.NVMeAvailableSparePercent</literal></term>
|
|
<listitem><para>The remaining spare capacity in percent.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.NVMeCriticalWarning</literal></term>
|
|
<listitem><para>The critical warning bitmask. This is zero if all is well. Otherwise, bit 0 indicates
|
|
that the spare capacity dropped below the threshold, bit 1 that a temperature threshold was
|
|
exceeded, bit 2 that the reliability is degraded due to media errors, bit 3 that the media has been
|
|
placed in read-only mode, bit 4 that a volatile memory backup device failed, and bit 5 that the
|
|
persistent memory region has become read-only.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.NVMeMediaErrors</literal></term>
|
|
<listitem><para>The number of unrecovered data integrity errors over the lifetime of the
|
|
disk.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.NVMePercentageUsed</literal></term>
|
|
<listitem><para>The vendor specific estimate of the used up life time of the disk in percent. Note
|
|
that this value may exceed 100.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
</variablelist>
|
|
|
|
<para>The following metrics are reported for ATA disks with SMART enabled only. Disks attached via USB
|
|
are not queried. In order not to spin up disks merely for collecting metrics, disks in standby mode are
|
|
not queried either. The metrics derived from SMART attributes 5, 197 and 198 are only reported for
|
|
rotating disks (as indicated by the <filename>queue/rotational</filename> sysfs attribute), since various
|
|
SSD vendors have reused these attributes for flash specific concepts with incompatible
|
|
semantics.</para>
|
|
|
|
<variablelist>
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.ATAOfflineUncorrectable</literal></term>
|
|
<listitem><para>The number of uncorrectable sectors found during offline scans (SMART attribute
|
|
198).</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.ATAPendingSectors</literal></term>
|
|
<listitem><para>The number of unstable sectors waiting to be remapped (SMART attribute
|
|
197).</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.ATAReallocatedSectors</literal></term>
|
|
<listitem><para>The number of sectors remapped to the spare area (SMART attribute
|
|
5).</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
|
|
<varlistentry>
|
|
<term><literal>io.systemd.DiskHealth.ATASmartStatusFailing</literal></term>
|
|
<listitem><para>1 if the overall SMART status of the disk reports that a threshold has been exceeded,
|
|
i.e. that failure of the disk is predicted, 0 otherwise.</para>
|
|
|
|
<xi:include href="version-info.xml" xpointer="v263"/></listitem>
|
|
</varlistentry>
|
|
</variablelist>
|
|
|
|
<para>The disk temperature is not reported by this service, see
|
|
<citerefentry><refentrytitle>systemd-report-hwmon@.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>
|
|
for that.</para>
|
|
</refsect1>
|
|
|
|
<!-- Note: we do not document the command line switches here. The systemd-report-diskhealth binary itself
|
|
should not be invoked by users, but only as a service. Use systemd-report(1) to acquire metrics. -->
|
|
|
|
<refsect1>
|
|
<title>See Also</title>
|
|
<para><simplelist type="inline">
|
|
<member><citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry></member>
|
|
<member><citerefentry><refentrytitle>systemd-report</refentrytitle><manvolnum>1</manvolnum></citerefentry></member>
|
|
<member><citerefentry><refentrytitle>systemd-report-hwmon@.service</refentrytitle><manvolnum>8</manvolnum></citerefentry></member>
|
|
<member><citerefentry><refentrytitle>systemd-udevd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry></member>
|
|
</simplelist></para>
|
|
</refsect1>
|
|
|
|
</refentry>
|