path: root/man/veritytab.xml

<?xml version="1.0"?>
<!--*-nxml-*-->
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
  "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<!--
  SPDX-License-Identifier: LGPL-2.1-or-later

This is based on crypttab(5).

-->
<refentry id="veritytab" conditional='HAVE_LIBCRYPTSETUP' xmlns:xi="http://www.w3.org/2001/XInclude">

  <refentryinfo>
    <title>veritytab</title>
    <productname>systemd</productname>
  </refentryinfo>

  <refmeta>
    <refentrytitle>veritytab</refentrytitle>
    <manvolnum>5</manvolnum>
  </refmeta>

  <refnamediv>
    <refname>veritytab</refname>
    <refpurpose>Configuration for verity block devices</refpurpose>
  </refnamediv>

  <refsynopsisdiv>
    <para><filename>/etc/veritytab</filename></para>
  </refsynopsisdiv>

  <refsect1>
    <title>Description</title>

    <para>The <filename>/etc/veritytab</filename> file describes
    verity protected block devices that are set up during
    system boot.</para>

    <para>Empty lines and lines starting with the <literal>#</literal>
    character are ignored. Each of the remaining lines describes one
    verity protected block device. Fields are delimited by
    white space.</para>

    <para>Each line is in the form<programlisting><replaceable>volume-name</replaceable> <replaceable>data-device</replaceable> <replaceable>hash-device</replaceable> <replaceable>roothash</replaceable> <replaceable>options</replaceable></programlisting>
    The first four fields are mandatory, the remaining one is optional.</para>

    <para>The first field contains the name of the resulting verity volume; its block device is set up
    below <filename>/dev/mapper/</filename>.</para>

    <para>The second field contains a path to the underlying block data device, or a specification of a block device via
    <literal>UUID=</literal> followed by the UUID.</para>

    <para>The third field contains a path to the underlying block hash device, or a specification of a block device via
    <literal>UUID=</literal> followed by the UUID.</para>

    <para>The fourth field is the <literal>roothash</literal> in hexadecimal.</para>

    <para>The fifth field, if present, is a comma-delimited list of options. The following options are
    recognized:</para>

    <variablelist class='fstab-options'>

      <varlistentry>
        <term><option>superblock=<replaceable>BOOL</replaceable></option></term>

        <listitem><para>Use dm-verity with or without permanent on-disk superblock.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>format=<replaceable>NUMBER</replaceable></option></term>

        <listitem><para>Specifies the hash version type. Format type 0 is original Chrome OS version. Format type 1 is
        modern version.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>data-block-size=<replaceable>BYTES</replaceable></option></term>

        <listitem><para>Used block size for the data device. (Note kernel supports only page-size as maximum
        here; Multiples of 512 bytes.) </para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>hash-block-size=<replaceable>BYTES</replaceable></option></term>

        <listitem><para>Used block size for the hash device. (Note kernel supports only page-size as maximum
        here; Multiples of 512 bytes.)</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>data-blocks=<replaceable>BLOCKS</replaceable></option></term>

        <listitem><para>Number of blocks of data device used in verification. If not specified, the whole device is
        used.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>hash-offset=<replaceable>BYTES</replaceable></option></term>

        <listitem><para>Offset of hash area/superblock on <literal>hash-device</literal>. (Multiples of 512 bytes.)
        </para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>salt=<replaceable>HEX</replaceable></option></term>

        <listitem><para>Salt used for format or verification. Format is a hexadecimal string; 256 bytes long maximum;
        <literal>-</literal>is the special value for empty.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>uuid=<replaceable>UUID</replaceable></option></term>

        <listitem><para>Use the provided UUID for format command instead of generating new one. The UUID must be
        provided in standard UUID format, e.g. 12345678-1234-1234-1234-123456789abc.</para></listitem>
        <listitem><para></para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>ignore-corruption</option></term>
        <term><option>restart-on-corruption</option></term>
        <term><option>panic-on-corruption</option></term>

        <listitem><para>Defines what to do if a data verity problem is detected (data corruption). Without these
        options kernel fails the IO operation with I/O error. With <literal>--ignore-corruption</literal> option the
        corruption is only logged. With <literal>--restart-on-corruption</literal> or
        <literal>--panic-on-corruption</literal> the kernel is restarted (panicked) immediately.

        (You have to provide way how to avoid restart loops.)</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>ignore-zero-blocks</option></term>

        <listitem><para>Instruct kernel to not verify blocks that are expected to contain zeroes and always directly
        return zeroes instead.

        WARNING: Use this option only in very specific cases. This option is available since Linux kernel version 4.5.
        </para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>check-at-most-once</option></term>

        <listitem><para>Instruct kernel to verify blocks only the first time they are read from the data device, rather
        than every time.

        WARNING: It provides a reduced level of security because only offline tampering of the data device's content
        will be detected, not online tampering. This option is available since Linux kernel version 4.17.
        </para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>hash=<replaceable>HASH</replaceable></option></term>

        <listitem><para>Hash algorithm for dm-verity. This should be the name of the algorithm, like "sha1". For default
        see <command>veritysetup --help</command>.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>fec-device=<replaceable>PATH</replaceable></option></term>

        <listitem><para>Use forward error correction (FEC) to recover from corruption if hash verification fails. Use
        encoding data from the specified device. The fec device argument can be block device or file image. For format,
        if fec device path doesn't exist, it will be created as file. Note: block sizes for data and hash devices must
        match. Also, if the verity data_device is encrypted the fec_device should be too.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>fec-offset=<replaceable>BYTES</replaceable></option></term>

        <listitem><para>This is the offset, in bytes, from the start of the FEC device to the beginning of the encoding
        data. (Aligned on 512 bytes.)</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>fec-roots=<replaceable>NUM</replaceable></option></term>

        <listitem><para>Number of generator roots. This equals to the number of parity bytes in the encoding data. In
        RS(M, N) encoding, the number of roots is M-N. M is 255 and M-N is between 2 and 24 (including).</para>
        </listitem>
      </varlistentry>

      <varlistentry>
        <term><option>root-hash-signature=<replaceable>PATH</replaceable>|base64:<replaceable>HEX</replaceable></option></term>

        <listitem><para>A base64 string encoding the root hash signature prefixed by <literal>base64:</literal> or a
        path to roothash signature file used to verify the root hash (in kernel). This feature requires Linux kernel
        version 5.4 or more recent.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>_netdev</option></term>

        <listitem><para>Marks this veritysetup device as requiring network. It will be
        started after the network is available, similarly to
        <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry>
        units marked with <option>_netdev</option>. The service unit to set up this device
        will be ordered between <filename>remote-fs-pre.target</filename> and
        <filename>remote-veritysetup.target</filename>, instead of
        <filename>veritysetup-pre.target</filename> and
        <filename>veritysetup.target</filename>.</para>

        <para>Hint: if this device is used for a mount point that is specified in
        <citerefentry project='man-pages'><refentrytitle>fstab</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
        the <option>_netdev</option> option should also be used for the mount
        point. Otherwise, a dependency loop might be created where the mount point
        will be pulled in by <filename>local-fs.target</filename>, while the
        service to configure the network is usually only started <emphasis>after</emphasis>
        the local file system has been mounted.</para>
        </listitem>
      </varlistentry>

      <varlistentry>
        <term><option>noauto</option></term>

        <listitem><para>This device will not be added to <filename>veritysetup.target</filename>.
        This means that it will not be automatically enabled on boot, unless something else pulls
        it in. In particular, if the device is used for a mount point, it'll be enabled
        automatically during boot, unless the mount point itself is also disabled with
        <option>noauto</option>.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>nofail</option></term>

        <listitem><para>This device will not be a hard dependency of
        <filename>veritysetup.target</filename>. It'll still be pulled in and started, but the system
        will not wait for the device to show up and be enabled, and boot will not fail if this is
        unsuccessful. Note that other units that depend on the enabled device may still fail. In
        particular, if the device is used for a mount point, the mount point itself also needs to
        have the <option>nofail</option> option, or the boot will fail if the device is not enabled
        successfully.</para></listitem>
      </varlistentry>

      <varlistentry>
        <term><option>x-initrd.attach</option></term>

        <listitem><para>Setup this verity protected block device in the initrd, similarly to
        <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry>
        units marked with <option>x-initrd.mount</option>.</para>

        <para>Although it's not necessary to mark the mount entry for the root file system with
        <option>x-initrd.mount</option>, <option>x-initrd.attach</option> is still recommended with
        the verity protected block device containing the root file system as otherwise systemd
        will attempt to detach the device during the regular system shutdown while it's still in
        use. With this option the device will still be detached but later after the root file
        system is unmounted.</para>

        <para>All other verity protected block devices that contain file systems mounted in the initrd should
        use this option.</para>
        </listitem>
      </varlistentry>

    </variablelist>

    <para>At early boot and when the system manager configuration is
    reloaded, this file is translated into native systemd units by
    <citerefentry><refentrytitle>systemd-veritysetup-generator</refentrytitle><manvolnum>8</manvolnum></citerefentry>.</para>
  </refsect1>

  <refsect1>
    <title>Examples</title>
    <example>
      <title>/etc/veritytab example</title>
      <para>Set up two verity protected block devices. One using device blocks, another using files.</para>

      <programlisting>usr  PARTUUID=783e45ae-7aa3-484a-beef-a80ff9c19cbb PARTUUID=21dc1dfe-4c33-8b48-98a9-918a22eb3e37 36e3f740ad502e2c25e2a23d9c7c17bf0fdad2300b7580842d4b7ec1fb0fa263 auto
data /etc/data /etc/hash a5ee4b42f70ae1f46a08a7c92c2e0a20672ad2f514792730f5d49d7606ab8fdf auto
</programlisting>
    </example>
  </refsect1>

  <refsect1>
    <title>See Also</title>
    <para>
      <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
      <citerefentry><refentrytitle>systemd-veritysetup@.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
      <citerefentry><refentrytitle>systemd-veritysetup-generator</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
      <citerefentry project='man-pages'><refentrytitle>fstab</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
      <citerefentry project='die-net'><refentrytitle>veritysetup</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
    </para>
  </refsect1>

</refentry>