diff options
author | Lennart Poettering <lennart@poettering.net> | 2021-03-02 18:57:04 +0100 |
---|---|---|
committer | Yu Watanabe <watanabe.yu+github@gmail.com> | 2021-03-03 04:16:32 +0100 |
commit | 725ad3b06288b2beeaaf178120010612a30646e4 (patch) | |
tree | ed62db87f169bc887b495d327696270862554b44 | |
parent | time-util: simplify overflow check (diff) | |
download | systemd-725ad3b06288b2beeaaf178120010612a30646e4.tar.xz systemd-725ad3b06288b2beeaaf178120010612a30646e4.zip |
fstab-generator: add new root=tmpfs option
It's useful to be able to combine a regular /usr/ file system with a
tmpfs as root, for an OS that boots up in volatile mode on every single
boot. Let's add explicit support for this via root=tmpfs.
Note the relationship to the existing systemd.volatile= option:
1. The kernel command line "root=/dev/… systemd.volatile=yes" will mount
the specified root fs, and then hide everything at the top by
overmounting it with a tmpfs, except for the /usr subtree.
2. The kernel command line "root=tmpfs mount.usr=/dev/…" otoh will mount
a toot fs at the top (just like the case above), but will then mount
the top-level dir of the fs specified in mount.usr= directly below
it.
Or to say this differently: in the first case /usr/ from the physical
storage fs is going to become /usr/ of the hierarchy ultimately booted,
while in the second case / from the physical storage fs is going to
become /usr of the hierarchy booted.
Philosophically I figure systemd.volatile= is more an option for
"one-off" boots, while root=tmpfs is something to have as default mode
of operation for suitable images.
This is currently hard to test reasonably, since Dracut refuses to
accept root=tmpfs. This needs to be addressed separately though.
-rw-r--r-- | man/systemd-fstab-generator.xml | 35 | ||||
-rw-r--r-- | src/fstab-generator/fstab-generator.c | 30 |
2 files changed, 52 insertions, 13 deletions
diff --git a/man/systemd-fstab-generator.xml b/man/systemd-fstab-generator.xml index ec8f5c988e..3c5a5cc50a 100644 --- a/man/systemd-fstab-generator.xml +++ b/man/systemd-fstab-generator.xml @@ -82,9 +82,20 @@ <varlistentry> <term><varname>root=</varname></term> - <listitem><para>Takes the root filesystem to mount in the - initrd. <varname>root=</varname> is honored by the - initrd.</para></listitem> + <listitem><para>Configures the operating system's root filesystem to mount when running in the + initrd. This accepts a device node path (usually <filename>/dev/disk/by-uuid/…</filename> or + <filename>/dev/disk/by-label/…</filename> or similar), or the special values <literal>gpt-auto</literal> + and <literal>tmpfs</literal>.</para> + + <para>Use <literal>gpt-auto</literal> to explicitly request automatic root file system discovery via + <citerefentry><refentrytitle>systemd-gpt-auto-generator</refentrytitle><manvolnum>8</manvolnum></citerefentry>.</para> + + <para>Use <literal>tmpfs</literal> in order to mount a <citerefentry + project='man-pages'><refentrytitle>tmpfs</refentrytitle><manvolnum>5</manvolnum></citerefentry> file + system as root file system of the OS. This is useful in combination with + <varname>mount.usr=</varname> (see below) in order to combine a volatile root file system with a + separate, immutable <filename>/usr/</filename> file system. Also see + <varname>systemd.volatile=</varname> below.</para></listitem> </varlistentry> <varlistentry> @@ -193,10 +204,19 @@ or any other resources stored in the root file system are physically removed. It's thus safe to boot a system that is normally operated in non-volatile mode temporarily into volatile mode, without losing data.</para> - <para>Note that with the exception of <literal>overlay</literal> mode, enabling this setting will only work - correctly on operating systems that can boot up with only <filename>/usr/</filename> mounted, and are able to - automatically populate <filename>/etc/</filename>, and also <filename>/var/</filename> in case of - <literal>systemd.volatile=yes</literal>.</para></listitem> + <para>Note that with the exception of <literal>overlay</literal> mode, enabling this setting will + only work correctly on operating systems that can boot up with only <filename>/usr/</filename> + mounted, and are able to automatically populate <filename>/etc/</filename>, and also + <filename>/var/</filename> in case of <literal>systemd.volatile=yes</literal>.</para> + + <para>Also see <varname>root=tmpfs</varname> above, for a method to combine a + <literal>tmpfs</literal> file system with a regular <filename>/usr/</filename> file system (as + configured via <varname>mount.usr=</varname>). The main distinction between + <varname>systemd.volatile=yes</varname>, and <varname>root=tmpfs</varname> in combination + <varname>mount.usr=</varname> is that the former operates on top of a regular root file system and + temporarily obstructs the files and directories above its <filename>/usr/</filename> subdirectory, + while the latter does not hide any files, but simply mounts a unpopulated tmpfs as root file system + and combines it with a user picked <filename>/usr/</filename> file system.</para></listitem> </varlistentry> <varlistentry> @@ -218,6 +238,7 @@ <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry>, <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>, <citerefentry><refentrytitle>systemd-cryptsetup-generator</refentrytitle><manvolnum>8</manvolnum></citerefentry>, + <citerefentry><refentrytitle>systemd-gpt-auto-generator</refentrytitle><manvolnum>8</manvolnum></citerefentry>, <citerefentry><refentrytitle>kernel-command-line</refentrytitle><manvolnum>7</manvolnum></citerefentry> </para> </refsect1> diff --git a/src/fstab-generator/fstab-generator.c b/src/fstab-generator/fstab-generator.c index a526d6e8fa..6df7fa5328 100644 --- a/src/fstab-generator/fstab-generator.c +++ b/src/fstab-generator/fstab-generator.c @@ -671,7 +671,8 @@ static int parse_fstab(bool initrd) { static int add_sysroot_mount(void) { _cleanup_free_ char *what = NULL; - const char *opts; + const char *opts, *fstype; + bool default_rw; int r; if (isempty(arg_root_what)) { @@ -691,12 +692,29 @@ static int add_sysroot_mount(void) { return 0; } - what = fstab_node_to_udev_node(arg_root_what); - if (!what) - return log_oom(); + if (streq(arg_root_what, "tmpfs")) { + /* If root=tmpfs is specified, then take this as shortcut for a writable tmpfs mount as root */ + + what = strdup("rootfs"); /* just a pretty name, to show up in /proc/self/mountinfo */ + if (!what) + return log_oom(); + + fstype = arg_root_fstype ?: "tmpfs"; /* tmpfs, unless overriden */ + + default_rw = true; /* writable, unless overriden */; + } else { + + what = fstab_node_to_udev_node(arg_root_what); + if (!what) + return log_oom(); + + fstype = arg_root_fstype; /* if not specified explicitly, don't default to anything here */ + + default_rw = false; /* read-only, unless overriden */ + } if (!arg_root_options) - opts = arg_root_rw > 0 ? "rw" : "ro"; + opts = arg_root_rw > 0 || (arg_root_rw < 0 && default_rw) ? "rw" : "ro"; else if (arg_root_rw >= 0 || !fstab_test_option(arg_root_options, "ro\0" "rw\0")) opts = strjoina(arg_root_options, ",", arg_root_rw > 0 ? "rw" : "ro"); @@ -715,7 +733,7 @@ static int add_sysroot_mount(void) { what, "/sysroot", NULL, - arg_root_fstype, + fstype, opts, is_device_path(what) ? 1 : 0, /* passno */ 0, /* makefs off, growfs off, noauto off, nofail off, automount off */ |