Skip to contentSkip to conversation

package/systemd: set machine id file instalation as optional choice

PATCH In review Bartosz Bilas 4 participants 22 messages Nov 17, 2019 – Aug 5, 2021

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

In case of e.g writable systems there is no neccesity to have pre-installed empty machine-id file on target because systemd generates that automatically during system boot.
Also in case of having empty machine-id file we are not able
to use a service with ConditionFirstBoot because systemd recognizes machine-id file in system therefore we can't detect new system instance boots up. Set this option as enable by default to keep compatibility with old builds.

---
 package/systemd/Config.in  | 13 +++++++++++++
 package/systemd/systemd.mk |  2 ++
 2 files changed, 15 insertions(+)
package/systemd/Config.in +13 -0
index aef39abe27..fadc1a32c8 100644
@@ -112,6 +112,19 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
 	default "x64"   if BR2_x86_64
 	depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
+	bool "Install empty machine id file"
+	default y
+	help
+	  The /etc/machine-id file contains the unique machine ID
+	  of the local system that is set during installation or
+	  boot. The machine ID is a single newline-terminated,
+	  hexadecimal, 32-character, lowercase ID. When decoded from
+	  hexadecimal, this corresponds to a 16-byte/128-bit value.
+	  This ID may not be all zeros.
+
+	  https://www.freedesktop.org/software/systemd/man/machine-id.html
+
 config BR2_PACKAGE_SYSTEMD_JOURNAL_GATEWAY
 	bool "HTTP server for journal events"
 	select BR2_PACKAGE_LIBMICROHTTPD
package/systemd/systemd.mk +2 -0
index 92490eb86b..fc348fe120 100644
@@ -461,9 +461,11 @@ define SYSTEMD_INSTALL_INIT_HOOK
 		$(TARGET_DIR)/etc/systemd/system/multi-user.target.wants/remote-fs.target
 endef
 
+ifeq ($(BR2_PACKAGE_SYSTEMD_MACHINEID_FILE),y)
 define SYSTEMD_INSTALL_MACHINEID_HOOK
 	touch $(TARGET_DIR)/etc/machine-id
 endef
+endif
 
 SYSTEMD_POST_INSTALL_TARGET_HOOKS += \
 	SYSTEMD_INSTALL_TARGET_CRYPTSETUP \
Signature · 1 line
2.24.0
#1 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

Allow the user to choose which systemd generator should be installed on the target instead of installing all of them by default. That's usefull in case of when someone is using firstboot condition and doesn't want to have e.g getty services created automatically during system boot by systemd-getty-generator. Set them enabled by default to keep compatibility with old builds.

---
 package/systemd/Config.in  | 94 ++++++++++++++++++++++++++++++++++++++
 package/systemd/systemd.mk | 56 +++++++++++++++++++++++
 2 files changed, 150 insertions(+)
package/systemd/Config.in +94 -0
index fadc1a32c8..052b69f4de 100644
@@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
 	default "x64"   if BR2_x86_64
 	depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+menu systemd-generators
+
+config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR
+	bool "systemd debug generator"
+	default y
+	help
+	  systemd-debug-generator is for enabling a runtime debug
+	  shell and masking specific units at boot.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html
+
+config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
+	bool "systemd fstab generator"
+	default y
+	help
+	  systemd-fstab-generator is a generator that translates
+	  /etc/fstab into native systemd units early at boot
+	  and when configuration of the system manager is reloaded.
+	  This will instantiate mount and swap units as necessary.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html
+
+config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
+	bool "systemd getty generator"
+	default y
+	help
+	  systemd-getty-generator is a generator that automatically
+	  instantiates serial-getty at .service on the kernel
+	  console(s), if they can function as ttys and are not
+	  provided by the virtual console subsystem. It will also
+	  instantiate serial-getty at .service instances for
+	  virtualizer consoles, if execution in a virtualized
+	  environment is detected.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html
+
+config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR
+	bool "systemd gpt auto generator"
+	default y
+	help
+	  systemd-gpt-auto-generator is a unit generator that
+	  automatically discovers root, /home/, /srv/, the EFI
+	  System Partition, the Extended Boot Loader Partition and
+	  swap partitions and creates mount and swap units for them,
+	  based on the partition type GUIDs of GUID partition tables
+	  (GPT). It implements the Discoverable Partitions
+	  Specification.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html
+
+config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR
+	bool "systemd rc local generator"
+	default y
+	help
+	  systemd-rc-local-generator is a generator that checks
+	  whether /etc/rc.local exists and is executable,
+	  and if it is pulls the rc-local.service unit into the
+	  boot process.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html
+
+config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
+	bool "systemd run generator"
+	default y
+	help
+	  systemd-run-generator is for invoking commands specified
+	  on the kernel command line as system service.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html
+
+config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+	bool "systemd system update generator"
+	default y
+	help
+	  systemd-system-update-generator is a generator that
+	  automatically redirects the boot process to
+	  system-update.target, if /system-update exists.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
+
+config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
+	bool "systemd sysv generator"
+	default y
+	help
+	  systemd-sysv-generator is a generator that creates wrapper
+	  .service units for SysV init scripts in /etc/init.d/* at
+	  boot and when configuration of the system manager is
+	  reloaded. This will allow systemd(1) to support them
+	  similarly to native units/
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
+
+endmenu
+
 config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
 	bool "Install empty machine id file"
 	default y
package/systemd/systemd.mk +56 -0

62 lines not on this page · raw message

Signature · 1 line
2.24.0
#2 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

Quoted 3 lines · Bartosz Bilas

Allow the user to choose which systemd generator should be installed on the target instead of installing all of them by default. That's usefull in case of when someone is using firstboot condition

useful

Quoted 15 lines

and doesn't want to have e.g getty services created automatically during system boot by systemd-getty-generator. Set them enabled by default to keep compatibility with old builds.

---
 package/systemd/Config.in  | 94 ++++++++++++++++++++++++++++++++++++++
 package/systemd/systemd.mk | 56 +++++++++++++++++++++++
 2 files changed, 150 insertions(+)
package/systemd/Config.in +0 -0
index fadc1a32c8..052b69f4de 100644

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 115

Quoted 4 lines
package/systemd/Config.in +1 -0
@@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
 	default "x64"   if BR2_x86_64
 	depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+menu systemd-generators

I don't like adding submenus. Instead, I'd just do it with a comment.

However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In that
case, it could be done with a condition just under that option and the generators would be indented. Although, maybe they don't only get executed on firstboot...

So, then I'd move this section to the end of Config.in, and use a comment
instead of a menu.

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 119

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR
+	bool "systemd debug generator"
+	default y

Although this one should default y for backward compatibility, I'm of a mind to
change the default here. It doesn't sound like something you want to have on by default.

Quoted 4 lines
package/systemd/Config.in +4 -0
+	help
+	  systemd-debug-generator is for enabling a runtime debug
+	  shell and masking specific units at boot.
+

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 133

Quoted 10 lines
package/systemd/Config.in +10 -0
+	  https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html
+
+config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
+	bool "systemd fstab generator"
+	default y
+	help
+	  systemd-fstab-generator is a generator that translates
+	  /etc/fstab into native systemd units early at boot
+	  and when configuration of the system manager is reloaded.
+	  This will instantiate mount and swap units as necessary.

Makes me realize that we probably shouldn't have an fstab in
skeleton-init-systemd...

Quoted 10 lines
package/systemd/Config.in +10 -0
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html
+
+config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
+	bool "systemd getty generator"
+	default y
+	help
+	  systemd-getty-generator is a generator that automatically
+	  instantiates serial-getty at .service on the kernel
+	  console(s), if they can function as ttys and are not

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 153

Quoted 10 lines
package/systemd/Config.in +10 -0
+	  provided by the virtual console subsystem. It will also
+	  instantiate serial-getty at .service instances for
+	  virtualizer consoles, if execution in a virtualized
+	  environment is detected.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html
+
+config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR
+	bool "systemd gpt auto generator"
+	default y

For this one, I also doubt that we want to have this default on.

Quoted 4 lines
package/systemd/Config.in +4 -0
+	help
+	  systemd-gpt-auto-generator is a unit generator that
+	  automatically discovers root, /home/, /srv/, the EFI
+	  System Partition, the Extended Boot Loader Partition and

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 167

Quoted 10 lines
package/systemd/Config.in +10 -0
+	  swap partitions and creates mount and swap units for them,
+	  based on the partition type GUIDs of GUID partition tables
+	  (GPT). It implements the Discoverable Partitions
+	  Specification.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html
+
+config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR
+	bool "systemd rc local generator"
+	default y

Same here. It's pretty useless in Buildroot context.

Quoted 49 lines
package/systemd/Config.in +41 -0
+	help
+	  systemd-rc-local-generator is a generator that checks
+	  whether /etc/rc.local exists and is executable,
+	  and if it is pulls the rc-local.service unit into the
+	  boot process.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html
+
+config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
+	bool "systemd run generator"
+	default y
+	help
+	  systemd-run-generator is for invoking commands specified
+	  on the kernel command line as system service.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html
+
+config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+	bool "systemd system update generator"
+	default y
+	help
+	  systemd-system-update-generator is a generator that
+	  automatically redirects the boot process to
+	  system-update.target, if /system-update exists.
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
+
+config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
+	bool "systemd sysv generator"
+	default y
+	help
+	  systemd-sysv-generator is a generator that creates wrapper
+	  .service units for SysV init scripts in /etc/init.d/* at
+	  boot and when configuration of the system manager is
+	  reloaded. This will allow systemd(1) to support them
+	  similarly to native units/
+
+	  https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
+
+endmenu
+
 config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
 	bool "Install empty machine id file"
 	default y
package/systemd/systemd.mk +0 -0
index fc348fe120..32349980e3 100644

package/systemd/systemd.mk define SYSTEMD_INSTALL_MACHINEID_HOOK · line 470

Quoted 4 lines
package/systemd/systemd.mk +1 -0
@@ -467,6 +467,62 @@ define SYSTEMD_INSTALL_MACHINEID_HOOK
 endef
 endif
 
+ifneq ($(BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR),y)

We prefer ifeq ($(...),) to ifneq ($(...),y). Yes, there are still 145
occurrences of the latter, some of them committed just a couple of days ago, but that's because of incomplete review :-)

Quoted 43 lines
package/systemd/systemd.mk +43 -0
+define SYSTEMD_DISABLE_DEBUG_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-debug-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_DEBUG_GENERATOR
+endif> +
+ifneq ($(BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR),y)
+define SYSTEMD_DISABLE_FSTAB_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-fstab-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_FSTAB_GENERATOR
+endif
+
+ifneq ($(BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR),y)
+define SYSTEMD_DISABLE_GETTY_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-getty-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_GETTY_GENERATOR
+endif
+
+ifneq ($(BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR),y)
+define SYSTEMD_DISABLE_GPT_AUTO_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-gpt-auto-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_GPT_AUTO_GENERATOR
+endif
+
+ifneq ($(BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR),y)
+define SYSTEMD_DISABLE_RC_LOCAL_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-rc-local-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_RC_LOCAL_GENERATOR
+endif
+
+ifneq ($(BR2_PACKAGE_SYSTEMD_RUN_GENERATOR),y)
+define SYSTEMD_DISABLE_RUN_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-run-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_RUN_GENERATOR
+endif
+
+ifneq ($(BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR),y)
+define SYSTEMD_DISABLE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-system-update-generator

package/systemd/systemd.mk define SYSTEMD_INSTALL_MACHINEID_HOOK · line 524

Quoted 10 lines
package/systemd/systemd.mk +10 -0
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+endif
+
+ifneq ($(BR2_PACKAGE_SYSTEMD_SYSTEMD_SYSV_GENERATOR),y)
+define SYSTEMD_DISABLE_SYSTEMD_SYSV_GENERATOR
+	rm -f $(TARGET_DIR)/usr/lib/systemd/system-generators/systemd-sysv-generator
+endef
+SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_DISABLE_SYSTEMD_SYSV_GENERATOR
+endif

That's quite a lot of code... How about this refactoring:

SYSTEMD_GENERATORS_TO_DISABLE = \
	$(if $(BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR),,systemd-debug-generator) \
	...

ifneq ($(strip $(SYSTEMD_GENERATORS_TO_DISABLE)),) define SYSTEMD_INSTALL_TARGET_GENERATORS

	rm -f $(addprefix

$(TARGET_DIR)/usr/lib/systemd/system-generators/,$(SYSTEMD_GENERATORS_TO_DISABLE)) endef
SYSTEMD_POST_INSTALL_TARGET_HOOKS += SYSTEMD_INSTALL_TARGET_GENERATORS endif

Also, please put this below...

package/systemd/systemd.mk define SYSTEMD_INSTALL_MACHINEID_HOOK · line 528

Quoted 4 lines
package/systemd/systemd.mk +1 -0
+
 SYSTEMD_POST_INSTALL_TARGET_HOOKS += \
 	SYSTEMD_INSTALL_TARGET_CRYPTSETUP \
 	SYSTEMD_INSTALL_TARGET_MACHINED \

... here, so that the above statement stays closer to the definitions of the variables that are used in it.

Regards,
Arnout

#3 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

overall I'm not very keen on adding options for the various generators...

there are special case where you want to get rid of a specific generator, but it's really special and probably better done in a post-image script

detailed answers below

Le dim. 17 nov. 2019 ? 15:12, Arnout Vandecappelle <[email protected]> a ?crit :

Quoted 30 lines · Bartosz Bilas
Quoted 3 lines

Allow the user to choose which systemd generator should be installed on the target instead of installing all of them by default. That's usefull in case of when someone is using firstboot condition

useful

Quoted 15 lines

and doesn't want to have e.g getty services created automatically during system boot by systemd-getty-generator. Set them enabled by default to keep compatibility with old builds.

---
 package/systemd/Config.in  | 94 ++++++++++++++++++++++++++++++++++++++
 package/systemd/systemd.mk | 56 +++++++++++++++++++++++
 2 files changed, 150 insertions(+)
package/systemd/Config.in +0 -0
index fadc1a32c8..052b69f4de 100644

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 115

Quoted 4 lines
package/systemd/Config.in +1 -0
@@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
      default "x64"   if BR2_x86_64
      depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+menu systemd-generators

I don't like adding submenus. Instead, I'd just do it with a comment.

Quoted 7 lines

However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In that
case, it could be done with a condition just under that option and the generators would be indented. Although, maybe they don't only get executed on
firstboot...

No... generators are a core feature of systemd, it's not related to first-boot in any way that I know

Quoted 7 lines

So, then I'd move this section to the end of Config.in, and use a comment
instead of a menu.

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 119

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR
+     bool "systemd debug generator"
+     default y
Quoted 7 lines

Although this one should default y for backward compatibility, I'm of a
mind to
change the default here. It doesn't sound like something you want to have on by
default.

Despite its name, this is not a debug tool, but more of a rescue tool. It allows you to get a shell back on a
very broken system, or to prevent a unit from starting from the kernel-command-line

This should really be here all the time except on very specific use-cases, I don't think making it optional
is a good idea

Quoted 19 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 125

Quoted 5 lines
package/systemd/Config.in +5 -0
+     help
+       systemd-debug-generator is for enabling a runtime debug
+       shell and masking specific units at boot.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 133

Quoted 9 lines
package/systemd/Config.in +9 -0
+
+config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
+     bool "systemd fstab generator"
+     default y
+     help
+       systemd-fstab-generator is a generator that translates
+       /etc/fstab into native systemd units early at boot
+       and when configuration of the system manager is reloaded.
+       This will instantiate mount and swap units as necessary.

Makes me realize that we probably shouldn't have an fstab in
skeleton-init-systemd...

Quoted 1 line

fstab is supported in systemd and it is not a deprecated feature.

It is the recommended way to automatically mount filesystems at boot.

this one is really fundamental, and should really not be removed. lots of things might break.

there are a couple of reasons to use .mount units instead (and the fstab-generator simply
creates .mount units from /etc/fstab) but in the common case you are supposed to use /etc/fstab

so again, I don't think this is a good idea.

Quoted 20 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 134

Quoted 2 lines
package/systemd/Config.in +2 -0
+
+

https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html

Quoted 4 lines
package/systemd/Config.in +4 -0
+
+config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
+     bool "systemd getty generator"
+     default y

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 150

Quoted 10 lines
package/systemd/Config.in +10 -0
+     help
+       systemd-getty-generator is a generator that automatically
+       instantiates serial-getty at .service on the kernel
+       console(s), if they can function as ttys and are not
+       provided by the virtual console subsystem. It will also
+       instantiate serial-getty at .service instances for
+       virtualizer consoles, if execution in a virtualized
+       environment is detected.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

That one could be made optional, since it's about runtime-detection of the right way to start a getty
(i.e figure out if the console is a serial tty, a normal tty, or a vconsole) and in the embedded case
we usually know what we have

note that it means manually enabling a service instead... which is probably harder

but otoh, it's typically very small and it does "the right thing" most of the time. So again, probably a
better fit for a post-image script

Quoted 7 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 153

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR
+     bool "systemd gpt auto generator"
+     default y

For this one, I also doubt that we want to have this default on.

That one can probably be made optional, yes... it saves 34k approximately

Quoted 11 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 164

Quoted 10 lines
package/systemd/Config.in +10 -0
+     help
+       systemd-gpt-auto-generator is a unit generator that
+       automatically discovers root, /home/, /srv/, the EFI
+       System Partition, the Extended Boot Loader Partition and
+       swap partitions and creates mount and swap units for them,
+       based on the partition type GUIDs of GUID partition tables
+       (GPT). It implements the Discoverable Partitions
+       Specification.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html

Quoted 8 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 167

Quoted 4 lines
package/systemd/Config.in +4 -0
+
+config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR
+     bool "systemd rc local generator"
+     default y

Same here. It's pretty useless in Buildroot context.

this one is already conditional to compiling sysV backward compatibility support...
which i think is always on in buildroot (neet to double-check that)

I think having an option to enable/disable sysV support would be more generic and more useful

Quoted 8 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 175

Quoted 7 lines
package/systemd/Config.in +7 -0
+     help
+       systemd-rc-local-generator is a generator that checks
+       whether /etc/rc.local exists and is executable,
+       and if it is pulls the rc-local.service unit into the
+       boot process.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 184

Quoted 9 lines
package/systemd/Config.in +9 -0
+
+config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
+     bool "systemd run generator"
+     default y
+     help
+       systemd-run-generator is for invoking commands specified
+       on the kernel command line as system service.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

This one could be optional... it's very usefull for containers, less for full-blown systems

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 194

Quoted 9 lines
package/systemd/Config.in +9 -0
+config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+     bool "systemd system update generator"
+     default y
+     help
+       systemd-system-update-generator is a generator that
+       automatically redirects the boot process to
+       system-update.target, if /system-update exists.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

this one could be optional

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 206

Quoted 11 lines
package/systemd/Config.in +11 -0
+config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
+     bool "systemd sysv generator"
+     default y
+     help
+       systemd-sysv-generator is a generator that creates wrapper
+       .service units for SysV init scripts in /etc/init.d/* at
+       boot and when configuration of the system manager is
+       reloaded. This will allow systemd(1) to support them
+       similarly to native units/
+
+

https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html

Quoted 4 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 208

Quoted 3 lines
package/systemd/Config.in +3 -0
+
+endmenu
+

Again, i'd rather have a generic way to disable the sysV support at compile time...

Ok so overall, I think most generators should stay in and only one or two should be made optional...
Sorry for the negative feedback, but I don't like adding complex compilation options that are not supported
upstream, especially when all generators together wait 272k on my debian

The few that could legitimately be made optional are easy enough to remove in a post-image that I'm not sure
it's worth the trouble

i'd be ok for a patch doing that, though, but only for the few ones that are worth it...

Cheers
Jeremy

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191117/2c924df3/attachment.html>
#4 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

Hello guys,

Quoted 244 lines · J?r?my ROSEN

overall I'm not very keen on adding options for the various generators...

there are special case where you want to get rid of a specific generator, but it's really special and probably better done in a post-image script

detailed answers below

Le?dim. 17 nov. 2019 ??15:12, Arnout Vandecappelle <[email protected] <mailto:[email protected]>> a ?crit?:

On 17/11/2019 12:33, Bartosz Bilas wrote:
> Allow the user to choose which systemd generator should be installed > on the target instead of installing all of them by default. > That's usefull in case of when someone is using firstboot condition ? ? ? ? ?useful

> and doesn't want to have e.g getty services created automatically during
> system boot by systemd-getty-generator. Set them enabled by default to
> keep compatibility with old builds.
>
> Signed-off-by: Bartosz Bilas <[email protected] <mailto:[email protected]>>
> ---
>? package/systemd/Config.in? | 94
++++++++++++++++++++++++++++++++++++++
>? package/systemd/systemd.mk <http://systemd.mk> | 56
+++++++++++++++++++++++
>? 2 files changed, 150 insertions(+)
>
> diff --git a/package/systemd/Config.in b/package/systemd/Config.in > index fadc1a32c8..052b69f4de 100644
> --- a/package/systemd/Config.in
> +++ b/package/systemd/Config.in
> @@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH >? ? ? ?default "x64"? ?if BR2_x86_64
>? ? ? ?depends on BR2_PACKAGE_SYSTEMD_BOOT
>
> +menu systemd-generators

?I don't like adding submenus. Instead, I'd just do it with a comment.

?However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In that
case, it could be done with a condition just under that option and the generators would be indented. Although, maybe they don't only get executed on
firstboot...

No... generators are a core feature of systemd, it's not related to first-boot in any way that I know

?So, then I'd move this section to the end of Config.in, and use a comment
instead of a menu.

> +config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR
> +? ? ?bool "systemd debug generator"
> +? ? ?default y

?Although this one should default y for backward compatibility, I'm of a mind to
change the default here. It doesn't sound like something you want to have on by
default.

Despite its name, this is not a debug tool, but more of a rescue tool. It allows you to get a shell back on a
very broken system, or to prevent a unit from starting from the kernel-command-line

This should really be here all the time except on very specific use-cases, I don't think making it optional
is a good idea

Review

Quoted 5 lines

+? ? ?help
+? ? ? ?systemd-debug-generator is for enabling a runtime debug +? ? ? ?shell and masking specific units at boot.

(quoted diff) +2 -0
+
+

https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
> +? ? ?bool "systemd fstab generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-fstab-generator is a generator that translates > +? ? ? ?/etc/fstab into native systemd units early at boot > +? ? ? ?and when configuration of the system manager is reloaded. > +? ? ? ?This will instantiate mount and swap units as necessary.

?Makes me realize that we probably shouldn't have an fstab in skeleton-init-systemd...

fstab is supported in systemd and it is not a deprecated feature. It is the recommended?way to automatically mount filesystems at boot.

this one is really fundamental, and should really not be removed. lots of things might break.

there are a couple of reasons to use .mount units instead (and the fstab-generator simply
creates .mount units from /etc/fstab) but in the common case you are supposed to use /etc/fstab

so again, I don't think this is a good idea.

Review

Quoted 2 lines
(quoted diff) +2 -0
+
+

https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
> +? ? ?bool "systemd getty generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-getty-generator is a generator that automatically > +? ? ? ?instantiates serial-getty at .service on the kernel > +? ? ? ?console(s), if they can function as ttys and are not > +? ? ? ?provided by the virtual console subsystem. It will also > +? ? ? ?instantiate serial-getty at .service instances for > +? ? ? ?virtualizer consoles, if execution in a virtualized > +? ? ? ?environment is detected.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html
> +

That one could be made optional, since it's about runtime-detection of the right way to start a getty
(i.e figure out if the console is a serial tty, a normal tty, or a vconsole) and in the embedded case
we usually know what we have

note that it means manually enabling a service instead... which is probably harder

but otoh, it's typically very small and it does "the right thing" most of the time. So again, probably a
better fit for a post-image script

Quoted 3 lines

+config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR
+? ? ?bool "systemd gpt auto generator"
+? ? ?default y

?For this one, I also doubt that we want to have this default on.

That one can probably be made optional, yes... it saves 34k approximately

Review

Quoted 10 lines

+? ? ?help
+? ? ? ?systemd-gpt-auto-generator is a unit generator that +? ? ? ?automatically discovers root, /home/, /srv/, the EFI +? ? ? ?System Partition, the Extended Boot Loader Partition and +? ? ? ?swap partitions and creates mount and swap units for them, +? ? ? ?based on the partition type GUIDs of GUID partition tables +? ? ? ?(GPT). It implements the Discoverable Partitions +? ? ? ?Specification.

(quoted diff) +2 -0
+
+

https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR > +? ? ?bool "systemd rc local generator" > +? ? ?default y

?Same here. It's pretty useless in Buildroot context.

this one is already conditional to compiling sysV backward compatibility support...
which i think is always on in buildroot (neet to double-check that)

I think having an option to enable/disable sysV support would be more generic and more useful

Review

Quoted 7 lines

+? ? ?help
+? ? ? ?systemd-rc-local-generator is a generator that checks +? ? ? ?whether /etc/rc.local exists and is executable, +? ? ? ?and if it is pulls the rc-local.service unit into the +? ? ? ?boot process.

(quoted diff) +2 -0
+
+
    https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html
    > +
    > +config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
    > +? ? ?bool "systemd run generator"
    > +? ? ?default y
    > +? ? ?help
    > +? ? ? ?systemd-run-generator is for invoking commands specified
    > +? ? ? ?on the kernel command line as system service.
    > +
    > +
    https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html
    > +

This one could be optional... it's very usefull?for containers, less for full-blown systems

Review

Quoted 9 lines
(quoted diff) +9 -0
+config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+? ? ?bool "systemd system update generator"
+? ? ?default y
+? ? ?help
+? ? ? ?systemd-system-update-generator is a generator that
+? ? ? ?automatically redirects the boot process to
+? ? ? ?system-update.target, if /system-update exists.
+
+
    https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
    > +

this one could be optional

Review

Quoted 11 lines
(quoted diff) +11 -0
+config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
+? ? ?bool "systemd sysv generator"
+? ? ?default y
+? ? ?help
+? ? ? ?systemd-sysv-generator is a generator that creates wrapper
+? ? ? ?.service units for SysV init scripts in /etc/init.d/* at
+? ? ? ?boot and when configuration of the system manager is
+? ? ? ?reloaded. This will allow systemd(1) to support them
+? ? ? ?similarly to native units/
+
+
    https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
    > +
    > +endmenu
    > +

Again, i'd rather have a generic way to disable the sysV support at compile time...

Quoted 5 lines

Ok so overall, I think most generators should stay in and only one or two should be made optional...
Sorry for the negative feedback, but I don't like adding complex compilation options that are not supported
upstream, especially when all generators together wait 272k on my debian

Until we don't decide about the installation machine-id file that patch doesn't make any sense due to the impossibility to run those generators at all.

Best
Bartek

Quoted 5 lines

The few that could legitimately be made optional are easy enough to remove in a post-image that I'm not sure
it's worth the trouble

Quoted 28 lines

i'd?be ok for a patch doing that, though, but only for the few ones that are worth it...

Cheers
Jeremy

Signature · 21 lines
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]>
phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook 
<https://www.facebook.com/smileopensource> LinkedIn 
<https://www.linkedin.com/company/smile> Github 
<https://github.com/Smile-SA>


D?couvrez l?univers Smile, rendez-vous sur smile.eu 
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191125/714955a0/attachment.html>

#5 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

I don't understand that remark...

generators are run at every boot and also when "systemctl daemon-reload" is called
They have nothing to do with the machine-id or the firstboot logic...

Le lun. 25 nov. 2019 ? 17:44, Bartosz Bilas <[email protected]> a ?crit :

Quoted 297 lines

Hello guys,
On 17.11.2019 16:06, J?r?my ROSEN wrote:

overall I'm not very keen on adding options for the various generators...

there are special case where you want to get rid of a specific generator, but it's really special and probably better done in a post-image script

detailed answers below

Le dim. 17 nov. 2019 ? 15:12, Arnout Vandecappelle <[email protected]> a ?crit :

Quoted 30 lines · Bartosz Bilas
Quoted 3 lines

Allow the user to choose which systemd generator should be installed on the target instead of installing all of them by default. That's usefull in case of when someone is using firstboot condition

useful

Quoted 15 lines

and doesn't want to have e.g getty services created automatically during system boot by systemd-getty-generator. Set them enabled by default to keep compatibility with old builds.

---
 package/systemd/Config.in  | 94 ++++++++++++++++++++++++++++++++++++++
 package/systemd/systemd.mk | 56 +++++++++++++++++++++++
 2 files changed, 150 insertions(+)
package/systemd/Config.in +0 -0
index fadc1a32c8..052b69f4de 100644

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 115

Quoted 4 lines
package/systemd/Config.in +1 -0
@@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
      default "x64"   if BR2_x86_64
      depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+menu systemd-generators

I don't like adding submenus. Instead, I'd just do it with a comment.

Quoted 8 lines

However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In
that
case, it could be done with a condition just under that option and the generators would be indented. Although, maybe they don't only get executed on
firstboot...

No... generators are a core feature of systemd, it's not related to first-boot in any way that I know

Quoted 7 lines

So, then I'd move this section to the end of Config.in, and use a comment
instead of a menu.

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 119

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR
+     bool "systemd debug generator"
+     default y
Quoted 7 lines

Although this one should default y for backward compatibility, I'm of a
mind to
change the default here. It doesn't sound like something you want to have on by
default.

Despite its name, this is not a debug tool, but more of a rescue tool. It allows you to get a shell back on a
very broken system, or to prevent a unit from starting from the kernel-command-line

This should really be here all the time except on very specific use-cases, I don't think making it optional
is a good idea

Quoted 19 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 125

Quoted 5 lines
package/systemd/Config.in +5 -0
+     help
+       systemd-debug-generator is for enabling a runtime debug
+       shell and masking specific units at boot.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 133

Quoted 9 lines
package/systemd/Config.in +9 -0
+
+config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
+     bool "systemd fstab generator"
+     default y
+     help
+       systemd-fstab-generator is a generator that translates
+       /etc/fstab into native systemd units early at boot
+       and when configuration of the system manager is reloaded.
+       This will instantiate mount and swap units as necessary.

Makes me realize that we probably shouldn't have an fstab in
skeleton-init-systemd...

Quoted 1 line

fstab is supported in systemd and it is not a deprecated feature.

It is the recommended way to automatically mount filesystems at boot.

this one is really fundamental, and should really not be removed. lots of things might break.

there are a couple of reasons to use .mount units instead (and the fstab-generator simply
creates .mount units from /etc/fstab) but in the common case you are supposed to use /etc/fstab

so again, I don't think this is a good idea.

Quoted 20 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 134

Quoted 2 lines
package/systemd/Config.in +2 -0
+
+

https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html

Quoted 4 lines
package/systemd/Config.in +4 -0
+
+config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
+     bool "systemd getty generator"
+     default y

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 150

Quoted 10 lines
package/systemd/Config.in +10 -0
+     help
+       systemd-getty-generator is a generator that automatically
+       instantiates serial-getty at .service on the kernel
+       console(s), if they can function as ttys and are not
+       provided by the virtual console subsystem. It will also
+       instantiate serial-getty at .service instances for
+       virtualizer consoles, if execution in a virtualized
+       environment is detected.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

That one could be made optional, since it's about runtime-detection of the right way to start a getty
(i.e figure out if the console is a serial tty, a normal tty, or a vconsole) and in the embedded case
we usually know what we have

note that it means manually enabling a service instead... which is probably harder

but otoh, it's typically very small and it does "the right thing" most of the time. So again, probably a
better fit for a post-image script

Quoted 7 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 153

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR
+     bool "systemd gpt auto generator"
+     default y

For this one, I also doubt that we want to have this default on.

That one can probably be made optional, yes... it saves 34k approximately

Quoted 11 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 164

Quoted 10 lines
package/systemd/Config.in +10 -0
+     help
+       systemd-gpt-auto-generator is a unit generator that
+       automatically discovers root, /home/, /srv/, the EFI
+       System Partition, the Extended Boot Loader Partition and
+       swap partitions and creates mount and swap units for them,
+       based on the partition type GUIDs of GUID partition tables
+       (GPT). It implements the Discoverable Partitions
+       Specification.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html

Quoted 8 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 167

Quoted 4 lines
package/systemd/Config.in +4 -0
+
+config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR
+     bool "systemd rc local generator"
+     default y

Same here. It's pretty useless in Buildroot context.

this one is already conditional to compiling sysV backward compatibility support...
which i think is always on in buildroot (neet to double-check that)

I think having an option to enable/disable sysV support would be more generic and more useful

Quoted 8 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 175

Quoted 7 lines
package/systemd/Config.in +7 -0
+     help
+       systemd-rc-local-generator is a generator that checks
+       whether /etc/rc.local exists and is executable,
+       and if it is pulls the rc-local.service unit into the
+       boot process.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 184

Quoted 9 lines
package/systemd/Config.in +9 -0
+
+config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
+     bool "systemd run generator"
+     default y
+     help
+       systemd-run-generator is for invoking commands specified
+       on the kernel command line as system service.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

This one could be optional... it's very usefull for containers, less for full-blown systems

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 194

Quoted 9 lines
package/systemd/Config.in +9 -0
+config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+     bool "systemd system update generator"
+     default y
+     help
+       systemd-system-update-generator is a generator that
+       automatically redirects the boot process to
+       system-update.target, if /system-update exists.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

this one could be optional

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 206

Quoted 11 lines
package/systemd/Config.in +11 -0
+config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
+     bool "systemd sysv generator"
+     default y
+     help
+       systemd-sysv-generator is a generator that creates wrapper
+       .service units for SysV init scripts in /etc/init.d/* at
+       boot and when configuration of the system manager is
+       reloaded. This will allow systemd(1) to support them
+       similarly to native units/
+
+

https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html

Quoted 4 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 208

Quoted 3 lines
package/systemd/Config.in +3 -0
+
+endmenu
+

Again, i'd rather have a generic way to disable the sysV support at compile time...

Ok so overall, I think most generators should stay in and only one or two should be made optional...
Sorry for the negative feedback, but I don't like adding complex compilation options that are not supported
upstream, especially when all generators together wait 272k on my debian

Until we don't decide about the installation machine-id file that patch doesn't make any sense due to the impossibility to run those generators at all.

Best
Bartek

The few that could legitimately be made optional are easy enough to remove in a post-image that I'm not sure
it's worth the trouble

i'd be ok for a patch doing that, though, but only for the few ones that are worth it...

Cheers
Jeremy

--
[image: SMILE] <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone] +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191125/91bcc1e9/attachment.html>
#6 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

Hello,

Quoted 5 lines · J?r?my ROSEN

I don't understand that remark...

generators are run at every boot and also when "systemctl daemon-reload" is called
They have nothing to do with the machine-id or the firstboot logic...

that's weird because when I was testing I noticed the generators were executing only once during the first new system instance boots up. I'll double-check that then.

Best
Bartek

Quoted 306 lines

Le?lun. 25 nov. 2019 ??17:44, Bartosz Bilas <[email protected] <mailto:[email protected]>> a ?crit?:

    Hello guys,

    On 17.11.2019 16:06, J?r?my ROSEN wrote:
Quoted 262 lines

overall I'm not very keen on adding options for the various generators...

there are special case where you want to get rid of a specific generator, but it's really special and probably better done in a post-image script

detailed answers below

Le?dim. 17 nov. 2019 ??15:12, Arnout Vandecappelle <arnout at mind.be <mailto:[email protected]>> a ?crit?:

On 17/11/2019 12:33, Bartosz Bilas wrote:
> Allow the user to choose which systemd generator should be installed
> on the target instead of installing all of them by default. > That's usefull in case of when someone is using firstboot condition
? ? ? ? ?useful

> and doesn't want to have e.g getty services created automatically during
> system boot by systemd-getty-generator. Set them enabled by default to
> keep compatibility with old builds.
>
> Signed-off-by: Bartosz Bilas <[email protected] <mailto:[email protected]>>
> ---
>? package/systemd/Config.in? | 94
++++++++++++++++++++++++++++++++++++++
>? package/systemd/systemd.mk <http://systemd.mk> | 56
+++++++++++++++++++++++
>? 2 files changed, 150 insertions(+)
>
> diff --git a/package/systemd/Config.in b/package/systemd/Config.in
> index fadc1a32c8..052b69f4de 100644
> --- a/package/systemd/Config.in
> +++ b/package/systemd/Config.in
> @@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH >? ? ? ?default "x64"? ?if BR2_x86_64
>? ? ? ?depends on BR2_PACKAGE_SYSTEMD_BOOT
>
> +menu systemd-generators

?I don't like adding submenus. Instead, I'd just do it with a comment.

?However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In that
case, it could be done with a condition just under that option and the
generators would be indented. Although, maybe they don't only get executed on
firstboot...

No... generators are a core feature of systemd, it's not related to first-boot in any way that I know

?So, then I'd move this section to the end of Config.in, and use a comment
instead of a menu.

> +config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR > +? ? ?bool "systemd debug generator" > +? ? ?default y

?Although this one should default y for backward compatibility, I'm of a mind to
change the default here. It doesn't sound like something you want to have on by
default.

Despite its name, this is not a debug tool, but more of a rescue tool. It allows you to get a shell back on a
very broken system, or to prevent a unit from starting from the kernel-command-line

This should really be here all the time except on very specific use-cases, I don't think making it optional
is a good idea

> +? ? ?help
> +? ? ? ?systemd-debug-generator is for enabling a runtime debug > +? ? ? ?shell and masking specific units at boot.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
> +? ? ?bool "systemd fstab generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-fstab-generator is a generator that translates > +? ? ? ?/etc/fstab into native systemd units early at boot > +? ? ? ?and when configuration of the system manager is reloaded.
> +? ? ? ?This will instantiate mount and swap units as necessary.

?Makes me realize that we probably shouldn't have an fstab in skeleton-init-systemd...

fstab is supported in systemd and it is not a deprecated feature. It is the recommended?way to automatically mount filesystems at boot.

this one is really fundamental, and should really not be removed. lots of things might break.

there are a couple of reasons to use .mount units instead (and the fstab-generator simply
creates .mount units from /etc/fstab) but in the common case you are supposed to use /etc/fstab

so again, I don't think this is a good idea.

> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
> +? ? ?bool "systemd getty generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-getty-generator is a generator that automatically
> +? ? ? ?instantiates serial-getty at .service <mailto:[email protected]> on the kernel
> +? ? ? ?console(s), if they can function as ttys and are not > +? ? ? ?provided by the virtual console subsystem. It will also > +? ? ? ?instantiate serial-getty at .service <mailto:[email protected]> instances for
> +? ? ? ?virtualizer consoles, if execution in a virtualized > +? ? ? ?environment is detected.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html
> +

That one could be made optional, since it's about runtime-detection of the right way to start a getty (i.e figure out if the console is a serial tty, a normal tty, or a vconsole) and in the embedded case
we usually know what we have

note that it means manually enabling a service instead... which is probably harder

but otoh, it's typically very small and it does "the right thing" most of the time. So again, probably a
better fit for a post-image script

> +config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR > +? ? ?bool "systemd gpt auto generator" > +? ? ?default y

?For this one, I also doubt that we want to have this default on.

That one can probably be made optional, yes... it saves 34k approximately

> +? ? ?help
> +? ? ? ?systemd-gpt-auto-generator is a unit generator that > +? ? ? ?automatically discovers root, /home/, /srv/, the EFI > +? ? ? ?System Partition, the Extended Boot Loader Partition and
> +? ? ? ?swap partitions and creates mount and swap units for them,
> +? ? ? ?based on the partition type GUIDs of GUID partition tables
> +? ? ? ?(GPT). It implements the Discoverable Partitions > +? ? ? ?Specification.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR
> +? ? ?bool "systemd rc local generator"
> +? ? ?default y

?Same here. It's pretty useless in Buildroot context.

this one is already conditional to compiling sysV backward compatibility support...
which i think is always on in buildroot (neet to double-check that)

I think having an option to enable/disable sysV support would be more generic and more useful

> +? ? ?help
> +? ? ? ?systemd-rc-local-generator is a generator that checks > +? ? ? ?whether /etc/rc.local exists and is executable, > +? ? ? ?and if it is pulls the rc-local.service unit into the > +? ? ? ?boot process.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
> +? ? ?bool "systemd run generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-run-generator is for invoking commands specified
> +? ? ? ?on the kernel command line as system service. > +
> +
https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html
> +

This one could be optional... it's very usefull?for containers, less for full-blown systems

> +config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR > +? ? ?bool "systemd system update generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-system-update-generator is a generator that > +? ? ? ?automatically redirects the boot process to > +? ? ? ?system-update.target, if /system-update exists. > +
> +
https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
> +

this one could be optional

> +config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
> +? ? ?bool "systemd sysv generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-sysv-generator is a generator that creates wrapper
> +? ? ? ?.service units for SysV init scripts in /etc/init.d/* at
> +? ? ? ?boot and when configuration of the system manager is > +? ? ? ?reloaded. This will allow systemd(1) to support them > +? ? ? ?similarly to native units/
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
> +
> +endmenu
> +

Again, i'd rather have a generic way to disable the sysV support at compile time...

Quoted 6 lines

Ok so overall, I think most generators should stay in and only one or two should be made optional...
Sorry for the negative feedback, but I don't like adding complex compilation options that are not supported
upstream, especially when all generators together wait 272k on my debian

Until we don't decide about the installation machine-id file that patch doesn't make any sense due to the impossibility to run those generators at all.

Best
Bartek

Quoted 25 lines

The few that could legitimately be made optional are easy enough to remove in a post-image that I'm not sure
it's worth the trouble

i'd?be ok for a patch doing that, though, but only for the few ones that are worth it...

Cheers
Jeremy

--
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]> phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Quoted 38 lines
Quoted 8 lines

Twitter <https://twitter.com/GroupeSmile> Facebook
<https://www.facebook.com/smileopensource> LinkedIn
<https://www.linkedin.com/company/smile> Github
<https://github.com/Smile-SA>

D?couvrez l?univers Smile, rendez-vous sur smile.eu
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

Signature · 26 lines
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]>
phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook 
<https://www.facebook.com/smileopensource> LinkedIn 
<https://www.linkedin.com/company/smile> Github 
<https://github.com/Smile-SA>


D?couvrez l?univers Smile, rendez-vous sur smile.eu 
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191125/d68c75f9/attachment.html>

#7 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

That's very weird... I'm absolutely certain that generators are (supposed to) run on config reload and every reboot.
The files they create are stored under /run, so they are destroyed on every reboot.

Either you have some bug somewhere or you misdiagnosed something...

Do not that generators are run *very early* in the boot (before the unit files are even loaded) so stuff like journald are not around yet. The best way to check if they are run might be to create a tiny generator that saves the current date and time in /run or something like that...

Jeremy

Le lun. 25 nov. 2019 ? 17:55, Bartosz Bilas <[email protected]> a ?crit :

Quoted 343 lines

Hello,
On 25.11.2019 17:48, J?r?my ROSEN wrote:

I don't understand that remark...

generators are run at every boot and also when "systemctl daemon-reload" is called
They have nothing to do with the machine-id or the firstboot logic...

that's weird because when I was testing I noticed the generators were executing only once during the first new system instance boots up. I'll double-check that then.
Best
Bartek

Le lun. 25 nov. 2019 ? 17:44, Bartosz Bilas <[email protected]> a ?crit :

Quoted 290 lines

Hello guys,
On 17.11.2019 16:06, J?r?my ROSEN wrote:

overall I'm not very keen on adding options for the various generators...

there are special case where you want to get rid of a specific generator, but it's really special and probably better done in a post-image script

detailed answers below

Le dim. 17 nov. 2019 ? 15:12, Arnout Vandecappelle <[email protected]> a ?crit :

Quoted 31 lines · Bartosz Bilas
Quoted 3 lines

Allow the user to choose which systemd generator should be installed on the target instead of installing all of them by default. That's usefull in case of when someone is using firstboot condition

useful

Quoted 1 line

and doesn't want to have e.g getty services created automatically

during

Quoted 14 lines

system boot by systemd-getty-generator. Set them enabled by default to keep compatibility with old builds.

---
 package/systemd/Config.in  | 94 ++++++++++++++++++++++++++++++++++++++
 package/systemd/systemd.mk | 56 +++++++++++++++++++++++
 2 files changed, 150 insertions(+)
package/systemd/Config.in +0 -0
index fadc1a32c8..052b69f4de 100644

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 115

Quoted 4 lines
package/systemd/Config.in +1 -0
@@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
      default "x64"   if BR2_x86_64
      depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+menu systemd-generators

I don't like adding submenus. Instead, I'd just do it with a comment.

Quoted 8 lines

However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In
that
case, it could be done with a condition just under that option and the generators would be indented. Although, maybe they don't only get executed on
firstboot...

No... generators are a core feature of systemd, it's not related to first-boot in any way that I know

Quoted 8 lines

So, then I'd move this section to the end of Config.in, and use a
comment
instead of a menu.

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 119

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR
+     bool "systemd debug generator"
+     default y
Quoted 7 lines

Although this one should default y for backward compatibility, I'm of a
mind to
change the default here. It doesn't sound like something you want to have on by
default.

Despite its name, this is not a debug tool, but more of a rescue tool. It allows you to get a shell back on a
very broken system, or to prevent a unit from starting from the kernel-command-line

This should really be here all the time except on very specific use-cases, I don't think making it optional
is a good idea

Quoted 19 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 125

Quoted 5 lines
package/systemd/Config.in +5 -0
+     help
+       systemd-debug-generator is for enabling a runtime debug
+       shell and masking specific units at boot.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 133

Quoted 9 lines
package/systemd/Config.in +9 -0
+
+config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
+     bool "systemd fstab generator"
+     default y
+     help
+       systemd-fstab-generator is a generator that translates
+       /etc/fstab into native systemd units early at boot
+       and when configuration of the system manager is reloaded.
+       This will instantiate mount and swap units as necessary.

Makes me realize that we probably shouldn't have an fstab in
skeleton-init-systemd...

Quoted 1 line

fstab is supported in systemd and it is not a deprecated feature.

It is the recommended way to automatically mount filesystems at boot.

this one is really fundamental, and should really not be removed. lots of things might break.

there are a couple of reasons to use .mount units instead (and the fstab-generator simply
creates .mount units from /etc/fstab) but in the common case you are supposed to use /etc/fstab

so again, I don't think this is a good idea.

Quoted 20 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 134

Quoted 2 lines
package/systemd/Config.in +2 -0
+
+

https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html

Quoted 4 lines
package/systemd/Config.in +4 -0
+
+config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR
+     bool "systemd getty generator"
+     default y

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 150

Quoted 10 lines
package/systemd/Config.in +10 -0
+     help
+       systemd-getty-generator is a generator that automatically
+       instantiates serial-getty at .service on the kernel
+       console(s), if they can function as ttys and are not
+       provided by the virtual console subsystem. It will also
+       instantiate serial-getty at .service instances for
+       virtualizer consoles, if execution in a virtualized
+       environment is detected.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

That one could be made optional, since it's about runtime-detection of the right way to start a getty
(i.e figure out if the console is a serial tty, a normal tty, or a vconsole) and in the embedded case
we usually know what we have

note that it means manually enabling a service instead... which is probably harder

but otoh, it's typically very small and it does "the right thing" most of the time. So again, probably a
better fit for a post-image script

Quoted 7 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 153

Quoted 3 lines
package/systemd/Config.in +3 -0
+config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR
+     bool "systemd gpt auto generator"
+     default y

For this one, I also doubt that we want to have this default on.

That one can probably be made optional, yes... it saves 34k approximately

Quoted 11 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 164

Quoted 10 lines
package/systemd/Config.in +10 -0
+     help
+       systemd-gpt-auto-generator is a unit generator that
+       automatically discovers root, /home/, /srv/, the EFI
+       System Partition, the Extended Boot Loader Partition and
+       swap partitions and creates mount and swap units for them,
+       based on the partition type GUIDs of GUID partition tables
+       (GPT). It implements the Discoverable Partitions
+       Specification.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html

Quoted 8 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 167

Quoted 4 lines
package/systemd/Config.in +4 -0
+
+config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR
+     bool "systemd rc local generator"
+     default y

Same here. It's pretty useless in Buildroot context.

this one is already conditional to compiling sysV backward compatibility support...
which i think is always on in buildroot (neet to double-check that)

I think having an option to enable/disable sysV support would be more generic and more useful

Quoted 8 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 175

Quoted 7 lines
package/systemd/Config.in +7 -0
+     help
+       systemd-rc-local-generator is a generator that checks
+       whether /etc/rc.local exists and is executable,
+       and if it is pulls the rc-local.service unit into the
+       boot process.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 184

Quoted 9 lines
package/systemd/Config.in +9 -0
+
+config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
+     bool "systemd run generator"
+     default y
+     help
+       systemd-run-generator is for invoking commands specified
+       on the kernel command line as system service.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

This one could be optional... it's very usefull for containers, less for full-blown systems

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 194

Quoted 9 lines
package/systemd/Config.in +9 -0
+config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR
+     bool "systemd system update generator"
+     default y
+     help
+       systemd-system-update-generator is a generator that
+       automatically redirects the boot process to
+       system-update.target, if /system-update exists.
+
+

https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html

Quoted 1 line
package/systemd/Config.in +1 -0
+

this one could be optional

Quoted 12 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 206

Quoted 11 lines
package/systemd/Config.in +11 -0
+config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR
+     bool "systemd sysv generator"
+     default y
+     help
+       systemd-sysv-generator is a generator that creates wrapper
+       .service units for SysV init scripts in /etc/init.d/* at
+       boot and when configuration of the system manager is
+       reloaded. This will allow systemd(1) to support them
+       similarly to native units/
+
+

https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html

Quoted 4 lines

package/systemd/Config.in config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH · line 208

Quoted 3 lines
package/systemd/Config.in +3 -0
+
+endmenu
+

Again, i'd rather have a generic way to disable the sysV support at compile time...

Ok so overall, I think most generators should stay in and only one or two should be made optional...
Sorry for the negative feedback, but I don't like adding complex compilation options that are not supported
upstream, especially when all generators together wait 272k on my debian

Until we don't decide about the installation machine-id file that patch doesn't make any sense due to the impossibility to run those generators at all.

Best
Bartek

The few that could legitimately be made optional are easy enough to remove in a post-image that I'm not sure
it's worth the trouble

i'd be ok for a patch doing that, though, but only for the few ones that are worth it...

Cheers
Jeremy

--
[image: SMILE] <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone] +33 6 88 25 87 42
[image: url] http://www.smile.eu

Quoted 9 lines

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

--
[image: SMILE] <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone] +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

_______________________________________________
buildroot mailing listbuildroot at busybox.nethttp://lists.busybox.net/mailman/listinfo/buildroot

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191125/b36b8a36/attachment-0001.html>
#8 lore raw

PATCH 2/2 package/systemd: set systemd generators instalation as optional choice

Hi everyone,

I had finally a couple of time to double-check that and I agree with Jeremy that I was wrong therefore let's reject this patch as it turned out as an unthought idea.

Best
Bartek

Quoted 391 lines · J?r?my ROSEN

That's very weird... I'm absolutely certain that generators are (supposed to) run on config reload and every reboot.
The files they create are stored under /run, so they are destroyed on every reboot.

Either you have some bug somewhere or you misdiagnosed something...

Do not that generators are run *very early* in the boot (before the unit files are even loaded) so stuff like journald are not around yet. The best way to check if they are run might be to create a tiny generator that saves the current date and time in /run or something like that...

Jeremy

Le?lun. 25 nov. 2019 ??17:55, Bartosz Bilas <[email protected] <mailto:[email protected]>> a ?crit?:

    Hello,

    On 25.11.2019 17:48, J?r?my ROSEN wrote:
Quoted 5 lines

I don't understand that remark...

generators are run at every boot and also when "systemctl daemon-reload" is called
They have nothing to do with the machine-id or the firstboot logic...

that's weird because when I was testing I noticed the generators were executing only once during the first new system instance boots up. I'll double-check that then.

Best
Bartek

Quoted 357 lines
    Le?lun. 25 nov. 2019 ??17:44, Bartosz Bilas
    <b.bilas at grinn-global.com <mailto:[email protected]>> a
    ?crit?:

        Hello guys,

        On 17.11.2019 16:06, J?r?my ROSEN wrote:
Quoted 281 lines

overall I'm not very keen on adding options for the various generators...

there are special case where you want to get rid of a specific generator, but it's really special and probably better done in a post-image script

detailed answers below

Le?dim. 17 nov. 2019 ??15:12, Arnout Vandecappelle <arnout at mind.be <mailto:[email protected]>> a ?crit?:

On 17/11/2019 12:33, Bartosz Bilas wrote:
> Allow the user to choose which systemd generator should be installed
> on the target instead of installing all of them by default.
> That's usefull in case of when someone is using firstboot condition
? ? ? ? ?useful

> and doesn't want to have e.g getty services created automatically during
> system boot by systemd-getty-generator. Set them enabled by default to
> keep compatibility with old builds.
>
> Signed-off-by: Bartosz Bilas <[email protected] <mailto:[email protected]>>
> ---
>? package/systemd/Config.in? | 94
++++++++++++++++++++++++++++++++++++++
>? package/systemd/systemd.mk <http://systemd.mk> | 56
+++++++++++++++++++++++
>? 2 files changed, 150 insertions(+)
>
> diff --git a/package/systemd/Config.in b/package/systemd/Config.in
> index fadc1a32c8..052b69f4de 100644
> --- a/package/systemd/Config.in
> +++ b/package/systemd/Config.in
> @@ -112,6 +112,100 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
>? ? ? ?default "x64"? ?if BR2_x86_64
>? ? ? ?depends on BR2_PACKAGE_SYSTEMD_BOOT
>
> +menu systemd-generators

?I don't like adding submenus. Instead, I'd just do it with a comment.

?However, doesn't all this depend on BR2_PACKAGE_SYSTEMD_FIRSTBOOT? In that
case, it could be done with a condition just under that option and the
generators would be indented. Although, maybe they don't only get executed on
firstboot...

No... generators are a core feature of systemd, it's not related to first-boot in any way that I know

?So, then I'd move this section to the end of Config.in, and use a comment
instead of a menu.

> +config BR2_PACKAGE_SYSTEMD_DEBUG_GENERATOR > +? ? ?bool "systemd debug generator" > +? ? ?default y

?Although this one should default y for backward compatibility, I'm of a mind to
change the default here. It doesn't sound like something you want to have on by
default.

Despite its name, this is not a debug tool, but more of a rescue tool. It allows you to get a shell back on a very broken system, or to prevent a unit from starting from the kernel-command-line

This should really be here all the time except on very specific use-cases, I don't think making it optional is a good idea

> + ? ?help
> +? ? ? ?systemd-debug-generator is for enabling a runtime debug
> +? ? ? ?shell and masking specific units at boot. > +
> +
https://www.freedesktop.org/software/systemd/man/systemd-debug-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_FSTAB_GENERATOR
> +? ? ?bool "systemd fstab generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-fstab-generator is a generator that translates
> +? ? ? ?/etc/fstab into native systemd units early at boot > +? ? ? ?and when configuration of the system manager is reloaded.
> +? ? ? ?This will instantiate mount and swap units as necessary.

?Makes me realize that we probably shouldn't have an fstab in
skeleton-init-systemd...

fstab is supported in systemd and it is not a deprecated feature.
It is the recommended?way to automatically mount filesystems at boot.

this one is really fundamental, and should really not be removed. lots of things might break.

there are a couple of reasons to use .mount units instead (and the fstab-generator simply
creates .mount units from /etc/fstab) but in the common case you are supposed to use /etc/fstab

so again, I don't think this is a good idea.

> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-fstab-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_GETTY_GENERATOR > +? ? ?bool "systemd getty generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-getty-generator is a generator that automatically
> +? ? ? ?instantiates serial-getty at .service <mailto:[email protected]> on the kernel > +? ? ? ?console(s), if they can function as ttys and are not
> +? ? ? ?provided by the virtual console subsystem. It will also
> +? ? ? ?instantiate serial-getty at .service <mailto:[email protected]> instances for > +? ? ? ?virtualizer consoles, if execution in a virtualized
> +? ? ? ?environment is detected.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-getty-generator.html
> +

That one could be made optional, since it's about runtime-detection of the right way to start a getty (i.e figure out if the console is a serial tty, a normal tty, or a vconsole) and in the embedded case we usually know what we have

note that it means manually enabling a service instead... which is probably harder

but otoh, it's typically very small and it does "the right thing" most of the time. So again, probably a better fit for a post-image script

> +config BR2_PACKAGE_SYSTEMD_GPT_AUTO_GENERATOR > +? ? ?bool "systemd gpt auto generator" > +? ? ?default y

?For this one, I also doubt that we want to have this default on.

That one can probably be made optional, yes... it saves 34k approximately

> + ? ?help
> +? ? ? ?systemd-gpt-auto-generator is a unit generator that
> +? ? ? ?automatically discovers root, /home/, /srv/, the EFI
> +? ? ? ?System Partition, the Extended Boot Loader Partition and
> +? ? ? ?swap partitions and creates mount and swap units for them,
> +? ? ? ?based on the partition type GUIDs of GUID partition tables
> +? ? ? ?(GPT). It implements the Discoverable Partitions > +? ? ? ?Specification.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-gpt-auto-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_RC_LOCAL_GENERATOR > +? ? ?bool "systemd rc local generator"
> +? ? ?default y

?Same here. It's pretty useless in Buildroot context.

this one is already conditional to compiling sysV backward compatibility support...
which i think is always on in buildroot (neet to double-check that)

I think having an option to enable/disable sysV support would be more generic and more useful

> + ? ?help
> +? ? ? ?systemd-rc-local-generator is a generator that checks
> +? ? ? ?whether /etc/rc.local exists and is executable, > +? ? ? ?and if it is pulls the rc-local.service unit into the
> +? ? ? ?boot process.
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-rc-local-generator.html
> +
> +config BR2_PACKAGE_SYSTEMD_RUN_GENERATOR
> +? ? ?bool "systemd run generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-run-generator is for invoking commands specified
> +? ? ? ?on the kernel command line as system service. > +
> +
https://www.freedesktop.org/software/systemd/man/systemd-run-generator.html
> +

This one could be optional... it's very usefull?for containers, less for full-blown systems

> +config BR2_PACKAGE_SYSTEMD_SYSTEM_UPDATE_GENERATOR > +? ? ?bool "systemd system update generator" > +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-system-update-generator is a generator that
> +? ? ? ?automatically redirects the boot process to > +? ? ? ?system-update.target, if /system-update exists. > +
> +
https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
> +

this one could be optional

> +config BR2_PACKAGE_SYSTEMD_SYSV_GENERATOR > +? ? ?bool "systemd sysv generator"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?systemd-sysv-generator is a generator that creates wrapper
> +? ? ? ?.service units for SysV init scripts in /etc/init.d/* at
> +? ? ? ?boot and when configuration of the system manager is
> +? ? ? ?reloaded. This will allow systemd(1) to support them
> +? ? ? ?similarly to native units/
> +
> +
https://www.freedesktop.org/software/systemd/man/systemd-system-update-generator.html
> +
> +endmenu
> +

Again, i'd rather have a generic way to disable the sysV support at compile time...

Quoted 6 lines

Ok so overall, I think most generators should stay in and only one or two should be made optional...
Sorry for the negative feedback, but I don't like adding complex compilation options that are not supported upstream, especially when all generators together wait 272k on my debian

Until we don't decide about the installation machine-id file that patch doesn't make any sense due to the impossibility to run those generators at all.

Best
Bartek

Quoted 25 lines

The few that could legitimately be made optional are easy enough to remove in a post-image that I'm not sure it's worth the trouble

i'd?be ok for a patch doing that, though, but only for the few ones that are worth it...

Cheers
Jeremy

--
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]> phone +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Quoted 8 lines

Twitter <https://twitter.com/GroupeSmile> Facebook
<https://www.facebook.com/smileopensource> LinkedIn
<https://www.linkedin.com/company/smile> Github
<https://github.com/Smile-SA>

D?couvrez l?univers Smile, rendez-vous sur smile.eu
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

    -- 
    SMILE <http://www.smile.eu/>

    20 rue des Jardins
    92600 Asni?res-sur-Seine

    	
    *J?r?my ROSEN*
    Architecte technique

    email jeremy.rosen at smile.fr <mailto:[email protected]>
    phone? +33 6 88 25 87 42
    url http://www.smile.eu <http://www.smile.eu/>

    Twitter <https://twitter.com/GroupeSmile> Facebook
    <https://www.facebook.com/smileopensource> LinkedIn
    <https://www.linkedin.com/company/smile> Github
    <https://github.com/Smile-SA>
Quoted 32 lines
Quoted 7 lines

D?couvrez l?univers Smile, rendez-vous sur smile.eu
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

_______________________________________________
buildroot mailing list
buildroot at busybox.net <mailto:[email protected]>
http://lists.busybox.net/mailman/listinfo/buildroot

Signature · 21 lines
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]>
phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook 
<https://www.facebook.com/smileopensource> LinkedIn 
<https://www.linkedin.com/company/smile> Github 
<https://github.com/Smile-SA>


D?couvrez l?univers Smile, rendez-vous sur smile.eu 
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20200203/b959925e/attachment.html>

#9 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

In reply to Bartosz Bilas

Hi Bartosz,

Quoted 1 line · Bartosz Bilas

In case of e.g writable systems there is no neccesity to have

writeable necessity

Quoted 6 lines

pre-installed empty machine-id file on target because systemd generates that automatically during system boot.
Also in case of having empty machine-id file we are not able
to use a service with ConditionFirstBoot because systemd recognizes machine-id file in system therefore we can't detect new system instance boots up. Set this option as enable by default to keep compatibility with old builds.

Nice that you take a look at this! I'm immediately going to follow it up with
more feature requests though :-)

- If there is a writeable persistent partition, but /etc is not writeable, can we create a symlink and will systemd still create it correctly at first boot?

- Maybe if BR2_TARGET_GENERIC_REMOUNT_ROOTFS_RW is set, we should default it or even force it to no.

Regards,
Arnout

Quoted 48 lines
---
 package/systemd/Config.in  | 13 +++++++++++++
 package/systemd/systemd.mk |  2 ++
 2 files changed, 15 insertions(+)
package/systemd/Config.in +13 -0
index aef39abe27..fadc1a32c8 100644
@@ -112,6 +112,19 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
 	default "x64"   if BR2_x86_64
 	depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
+	bool "Install empty machine id file"
+	default y
+	help
+	  The /etc/machine-id file contains the unique machine ID
+	  of the local system that is set during installation or
+	  boot. The machine ID is a single newline-terminated,
+	  hexadecimal, 32-character, lowercase ID. When decoded from
+	  hexadecimal, this corresponds to a 16-byte/128-bit value.
+	  This ID may not be all zeros.
+
+	  https://www.freedesktop.org/software/systemd/man/machine-id.html
+
 config BR2_PACKAGE_SYSTEMD_JOURNAL_GATEWAY
 	bool "HTTP server for journal events"
 	select BR2_PACKAGE_LIBMICROHTTPD
package/systemd/systemd.mk +2 -0
index 92490eb86b..fc348fe120 100644
@@ -461,9 +461,11 @@ define SYSTEMD_INSTALL_INIT_HOOK
 		$(TARGET_DIR)/etc/systemd/system/multi-user.target.wants/remote-fs.target
 endef
 
+ifeq ($(BR2_PACKAGE_SYSTEMD_MACHINEID_FILE),y)
 define SYSTEMD_INSTALL_MACHINEID_HOOK
 	touch $(TARGET_DIR)/etc/machine-id
 endef
+endif
 
 SYSTEMD_POST_INSTALL_TARGET_HOOKS += \
 	SYSTEMD_INSTALL_TARGET_CRYPTSETUP \
#10 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

hmmm, I'm kinda torn on that one..

The only difference I see betweed no machine-id and an empty machine-id is systemd-firstboot and ConditionFirstBoot

if the file is empty or missing and systemd can write to it, systemd will set it up (not sure what happens for Arnout's symlink, that's worth testing) however, if the file exists but is empty we will skip systemd-firstboot

for context, systemd-firstboot does the following:
* set (unconditionnaly) a machine-id
* sets the locale (LANG= LC_MESSGES)
* sets the hostname
* sets the timezone
* sets the root password

systemd-firstboot can use values from its command line, but when launched on the target, the point is usually to prompt the user for answers. (as a side note, we can call it with --root=$DESTDIR once we have host-systemd)

So I would tend to consider systemd-firstboot as "an uncommon case" because prompting user on first boot is not something we see often in the embedded space

The good news is: we already know when systemd-firstboot is used because it's a config option of the systemd package.

ConditionFirstBoot is a bit more useful for us.. it allows to easily run a service only at first boot (or skip it on first boot)

we also have a big clue if the system is meant to be read-only via the "remount RW" option

So, putting it all together
* if we use systemd-firstboot, we don't want a machine-id
* if we have a read-only rootfs, we want an empty machine-id
* in other cases, we can deal with both, but it's better to have no machine-id
* I don't think that the concept of first boot makes any sense with a read-only rootfs...

Could something like that be enough ? can we trust "remount RW" ? maybe "remount RW" should be renamed "create a RW filesystem" and enable various tweaks related to RO vs RW

my two cents...

Le dim. 17 nov. 2019 ? 14:39, Arnout Vandecappelle <[email protected]> a ?crit :

Quoted 86 lines

Hi Bartosz,

Quoted 1 line · Bartosz Bilas

In case of e.g writable systems there is no neccesity to have

writeable necessity

Quoted 5 lines

pre-installed empty machine-id file on target because systemd generates that automatically during system boot.
Also in case of having empty machine-id file we are not able to use a service with ConditionFirstBoot because systemd recognizes machine-id file in system therefore we can't detect new system instance

boots up.

Quoted 1 line

Set this option as enable by default to keep compatibility with old

builds.

Nice that you take a look at this! I'm immediately going to follow it up
with
more feature requests though :-)

- If there is a writeable persistent partition, but /etc is not writeable, can
we create a symlink and will systemd still create it correctly at first boot?

- Maybe if BR2_TARGET_GENERIC_REMOUNT_ROOTFS_RW is set, we should default it or
even force it to no.

Regards,
Arnout

Quoted 36 lines
---
 package/systemd/Config.in  | 13 +++++++++++++
 package/systemd/systemd.mk |  2 ++
 2 files changed, 15 insertions(+)
package/systemd/Config.in +13 -0
index aef39abe27..fadc1a32c8 100644
@@ -112,6 +112,19 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
      default "x64"   if BR2_x86_64
      depends on BR2_PACKAGE_SYSTEMD_BOOT
 
+config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
+     bool "Install empty machine id file"
+     default y
+     help
+       The /etc/machine-id file contains the unique machine ID
+       of the local system that is set during installation or
+       boot. The machine ID is a single newline-terminated,
+       hexadecimal, 32-character, lowercase ID. When decoded from
+       hexadecimal, this corresponds to a 16-byte/128-bit value.
+       This ID may not be all zeros.
+
+       https://www.freedesktop.org/software/systemd/man/machine-id.html
+
 config BR2_PACKAGE_SYSTEMD_JOURNAL_GATEWAY
      bool "HTTP server for journal events"
      select BR2_PACKAGE_LIBMICROHTTPD
package/systemd/systemd.mk +0 -0
index 92490eb86b..fc348fe120 100644

package/systemd/systemd.mk define SYSTEMD_INSTALL_INIT_HOOK · line 461

Quoted 1 line
package/systemd/systemd.mk +0 -0
@@ -461,9 +461,11 @@ define SYSTEMD_INSTALL_INIT_HOOK
 

$(TARGET_DIR)/etc/systemd/system/multi-user.target.wants/remote-fs.target

package/systemd/systemd.mk define SYSTEMD_INSTALL_INIT_HOOK · line 471

Quoted 11 lines
package/systemd/systemd.mk +2 -0
 endef
 
+ifeq ($(BR2_PACKAGE_SYSTEMD_MACHINEID_FILE),y)
 define SYSTEMD_INSTALL_MACHINEID_HOOK
      touch $(TARGET_DIR)/etc/machine-id
 endef
+endif
 
 SYSTEMD_POST_INSTALL_TARGET_HOOKS += \
      SYSTEMD_INSTALL_TARGET_CRYPTSETUP \

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191117/45cea66a/attachment.html>
#11 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Quoted 4 lines · J?r?my ROSEN

hmmm, I'm kinda torn on that one..

The only difference I see betweed no machine-id and an empty machine-id is systemd-firstboot and ConditionFirstBoot

That's exactly the point I think.

I assumed, however, that ConditionFirstBoot was evaluated by systemd-firstboot,
but that's not the case then?

Quoted 6 lines

if the file is empty or missing and systemd can write to it, systemd will set it up (not sure what happens for Arnout's symlink, that's worth testing) however, if the file exists but is empty we will skip systemd-firstboot

for context, systemd-firstboot does the following:
* set (unconditionnaly) a machine-id

Are you saying that if systemd-firstboot it not enabled, no machine-id will be
generated?

Quoted 9 lines

* sets the locale (LANG= LC_MESSGES)
* sets the hostname
* sets the timezone
* sets the root password

systemd-firstboot can use values from its command line, but when launched on the target, the point is usually to prompt the user for answers.
(as a side note, we can call it with --root=$DESTDIR once we have host-systemd)

Quoted 8 lines

So I would tend to consider systemd-firstboot as "an uncommon case" because prompting?user on first boot is not something we see often in the embedded space

The good news is: we already know when systemd-firstboot is used because it's a config option of the systemd package.

ConditionFirstBoot is a bit more useful for us.. it allows to easily run a service only at first boot (or skip it on first boot)

Well, here are a few things you'd typically want to generate on first boot:

- machine id;
- ssh keys;
- various certificates;
- maybe a hashed root password based on a serial number of MAC address;
- maybe a random or hashed hostname;
- register on the phone-home server;
- resize the data partition based on actual size of the eMMC.

None of them interactively, of course.

So apparently systemd-firstboot can only do the first one. That's indeed not
very useful.

Quoted 2 lines

we also have a big clue if the system is meant to be read-only via the "remount RW" option

Yeah, but it's just a clue. You can have a writeable machine ID without a
remount RW option in a number of situations:

- custom skeleton
- fstab is overwritten by fs-overlay
- /etc is (bind)-mounted to a writeable partition
- /etc/machine-id is symlinked into a writeable partition
- root is squashfs (so it's not writeable even if RW is on)

Quoted 2 lines

So, putting it all together
* if we use systemd-firstboot, we don't want a machine-id

... at least, not an empty one :-)

Quoted 4 lines

* if we have a read-only rootfs, we want an empty machine-id
* in other cases, we can deal with both, but it's better to have no machine-id
* I don't think that the concept of first boot makes any sense with a read-only rootfs...

With readonly *rootfs* it certainly does, but not without any persistent
storage at all.

Quoted 4 lines

Could something like that be enough ? can we trust "remount RW" ? maybe "remount RW" should be renamed "create a RW filesystem" and enable various tweaks related to RO vs RW

As written above: no.

The problem is: we're not a distro. We leave too much freedom for the user to
integrate things in various ways to be able to make assumptions about what is the right way to do things. So, the only thing we can do is to give a decent out-of-the-box experience, and let the user figure out how to tweak things - possibly adding a config option for a common situation that is easily handled in a generic way. The other thing we can do is to provide documentation about the proper way to integrate things in different scenarios.

I'm starting to agree that this option is maybe not that great.

Actually, come to think of it, maybe it's better to never create the machine-id
at all. What is the effect (with a readonly rootfs) if machine-id doesn't exist? Does it block booting, does the boot take a little longer, or do we boot with a different machine-id all the time? And in the latter case, is there a real problem with that?

Regards,
Arnout

Quoted 113 lines

my two cents...

Le?dim. 17 nov. 2019 ??14:39, Arnout Vandecappelle <[email protected] <mailto:[email protected]>> a ?crit?:

?Hi Bartosz,

On 17/11/2019 12:33, Bartosz Bilas wrote:
> In case of e.g writable systems there is no neccesity to have ? ? ? ? ? ? ? ? ?writeable? ? ? ? ? ? ? ? ? ? necessity

> pre-installed empty machine-id file on target because systemd > generates that automatically during system boot.
> Also in case of having empty machine-id file we are not able
> to use a service with ConditionFirstBoot because systemd recognizes > machine-id file in system therefore we can't detect new system instance boots up.
> Set this option as enable by default to keep compatibility with old builds.

?Nice that you take a look at this! I'm immediately going to follow it up with more feature requests though :-)

- If there is a writeable persistent partition, but /etc is not writeable, can we create a symlink and will systemd still create it correctly at first boot?

- Maybe if BR2_TARGET_GENERIC_REMOUNT_ROOTFS_RW is set, we should default it or even force it to no.

?Regards,
?Arnout

>
> Signed-off-by: Bartosz Bilas <[email protected] <mailto:[email protected]>>
> ---
>? package/systemd/Config.in? | 13 +++++++++++++
>? package/systemd/systemd.mk <http://systemd.mk> |? 2 ++
>? 2 files changed, 15 insertions(+)
>
> diff --git a/package/systemd/Config.in b/package/systemd/Config.in > index aef39abe27..fadc1a32c8 100644
> --- a/package/systemd/Config.in
> +++ b/package/systemd/Config.in
> @@ -112,6 +112,19 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH >? ? ? ?default "x64"? ?if BR2_x86_64
>? ? ? ?depends on BR2_PACKAGE_SYSTEMD_BOOT
>?
> +config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
> +? ? ?bool "Install empty machine id file"
> +? ? ?default y
> +? ? ?help
> +? ? ? ?The /etc/machine-id file contains the unique machine ID > +? ? ? ?of the local system that is set during installation or > +? ? ? ?boot. The machine ID is a single newline-terminated, > +? ? ? ?hexadecimal, 32-character, lowercase ID. When decoded from > +? ? ? ?hexadecimal, this corresponds to a 16-byte/128-bit value. > +? ? ? ?This ID may not be all zeros.
> +
> +? ? ? ?https://www.freedesktop.org/software/systemd/man/machine-id.html > +
>? config BR2_PACKAGE_SYSTEMD_JOURNAL_GATEWAY
>? ? ? ?bool "HTTP server for journal events"
>? ? ? ?select BR2_PACKAGE_LIBMICROHTTPD
> diff --git a/package/systemd/systemd.mk <http://systemd.mk>
b/package/systemd/systemd.mk <http://systemd.mk>
> index 92490eb86b..fc348fe120 100644
> --- a/package/systemd/systemd.mk <http://systemd.mk>
> +++ b/package/systemd/systemd.mk <http://systemd.mk>
> @@ -461,9 +461,11 @@ define SYSTEMD_INSTALL_INIT_HOOK
>? ? ? ? ? ? ? ?$(TARGET_DIR)/etc/systemd/system/multi-user.target.wants/remote-fs.target >? endef
>?
> +ifeq ($(BR2_PACKAGE_SYSTEMD_MACHINEID_FILE),y)
>? define SYSTEMD_INSTALL_MACHINEID_HOOK
>? ? ? ?touch $(TARGET_DIR)/etc/machine-id
>? endef
> +endif
>?
>? SYSTEMD_POST_INSTALL_TARGET_HOOKS += \
>? ? ? ?SYSTEMD_INSTALL_TARGET_CRYPTSETUP \
>
_______________________________________________
buildroot mailing list
buildroot at busybox.net <mailto:[email protected]>
http://lists.busybox.net/mailman/listinfo/buildroot

Signature · 20 lines
SMILE? <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email?jeremy.rosen at smile.fr <mailto:[email protected]>?
phone? +33 6 88 25 87 42?
url?http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile>?Facebook
<https://www.facebook.com/smileopensource>?LinkedIn
<https://www.linkedin.com/company/smile>?Github <https://github.com/Smile-SA>


D?couvrez l?univers Smile, rendez-vous sur smile.eu
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
#12 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Argh, each time I simplify a bit, it comes back to bite me :)

Le dim. 17 nov. 2019 ? 17:00, Arnout Vandecappelle <[email protected]> a ?crit :

Quoted 17 lines · J?r?my ROSEN
Quoted 3 lines

hmmm, I'm kinda torn on that one..

The only difference I see betweed no machine-id and an empty machine-id

is

Quoted 1 line

systemd-firstboot and ConditionFirstBoot

That's exactly the point I think.

I assumed, however, that ConditionFirstBoot was evaluated by
systemd-firstboot,
but that's not the case then?

no, all Condition* are handled by PID1. systemd-firstboot.service itself relies on ConditionFirstBoot to decide if the binary needs to be started

Quoted 5 lines
Quoted 1 line

if the file is empty or missing and systemd can write to it, systemd

will set it

Quoted 3 lines

up (not sure what happens for Arnout's symlink, that's worth testing) however, if the file exists but is empty we will skip systemd-firstboot

Quoted 8 lines
Quoted 2 lines

for context, systemd-firstboot does the following:
* set (unconditionnaly) a machine-id

Are you saying that if systemd-firstboot it not enabled, no machine-id
will be
generated?

no, it's a bit more complex, and i'd need to check the details.

my understanding is that systemd-firstboot will always override the machine-id whereas on a normal boot,
the machine-id will be used if available and generated/written if not

Quoted 96 lines
Quoted 6 lines

* sets the locale (LANG= LC_MESSGES)
* sets the hostname
* sets the timezone
* sets the root password

systemd-firstboot can use values from its command line, but when

launched on the

Quoted 2 lines

target, the point is usually to prompt the user for answers. (as a side note, we can call it with --root=$DESTDIR once we have

host-systemd)

Quoted 2 lines

So I would tend to consider systemd-firstboot as "an uncommon case"

because

Quoted 1 line

prompting user on first boot is not something we see often in the

embedded space

Quoted 2 lines

The good news is: we already know when systemd-firstboot is used because

it's a

Quoted 3 lines

config option of the systemd package.

ConditionFirstBoot is a bit more useful for us.. it allows to easily run

a

Quoted 1 line

service only at first boot (or skip it on first boot)

Well, here are a few things you'd typically want to generate on first
boot:

- machine id;
- ssh keys;
- various certificates;
- maybe a hashed root password based on a serial number of MAC address;
- maybe a random or hashed hostname;
- register on the phone-home server;
- resize the data partition based on actual size of the eMMC.

None of them interactively, of course.

So apparently systemd-firstboot can only do the first one. That's indeed
not
very useful.

Quoted 1 line

we also have a big clue if the system is meant to be read-only via the

"remount

Quoted 1 line

RW" option

Yeah, but it's just a clue. You can have a writeable machine ID without a
remount RW option in a number of situations:

- custom skeleton
- fstab is overwritten by fs-overlay
- /etc is (bind)-mounted to a writeable partition
- /etc/machine-id is symlinked into a writeable partition
- root is squashfs (so it's not writeable even if RW is on)

Quoted 2 lines

So, putting it all together
* if we use systemd-firstboot, we don't want a machine-id

... at least, not an empty one :-)

Quoted 2 lines

* if we have a read-only rootfs, we want an empty machine-id
* in other cases, we can deal with both, but it's better to have no

machine-id

Quoted 1 line

* I don't think that the concept of first boot makes any sense with a

read-only

Quoted 1 line

rootfs...

With readonly *rootfs* it certainly does, but not without any persistent
storage at all.

Quoted 3 lines

Could something like that be enough ? can we trust "remount RW" ? maybe "remount RW" should be renamed "create a RW filesystem" and enable

various

Quoted 1 line

tweaks related to RO vs RW

As written above: no.

The problem is: we're not a distro. We leave too much freedom for the
user to
integrate things in various ways to be able to make assumptions about what is
the right way to do things. So, the only thing we can do is to give a decent
out-of-the-box experience, and let the user figure out how to tweak things -
possibly adding a config option for a common situation that is easily handled in
a generic way. The other thing we can do is to provide documentation about the
proper way to integrate things in different scenarios.

I'm starting to agree that this option is maybe not that great.

Quoted 9 lines

Actually, come to think of it, maybe it's better to never create the
machine-id
at all. What is the effect (with a readonly rootfs) if machine-id doesn't exist?
Does it block booting, does the boot take a little longer, or do we boot with a
different machine-id all the time? And in the latter case, is there a real problem with that?

on a RO filesystem
* if /etc/machine-id exists and contains a machine-id, systemd will use it
* if it's an empty file, systemd will bind-mount a file with a machine-id over the file
* if the file does not exist, systemd can't bind mount over the file... systemd itself should be fine, but other services might not, it's hard to tell

So the recommended way for a read-only rootfs is to have that empty file, which is why this is what currently have.
it works both for RW and RO filesystems..

A longer-term solution would be to add a patch upstream to allow to specify in a systemd configuration file what is
the file to check for first boot instead of /etc/machine-id.

I can look into that, but for the short term, i'd rather break ConditionFirstBoot than read-only rootfs

so... not sure what to advise at that point. that choice is something a bit too specialized for a config option
especially if we plan to drop it in the future

so not sure where to go...

Quoted 129 lines

Regards,
Arnout

Quoted 13 lines

my two cents...

Le dim. 17 nov. 2019 ? 14:39, Arnout Vandecappelle <[email protected] <mailto:[email protected]>> a ?crit :

Hi Bartosz,

On 17/11/2019 12:33, Bartosz Bilas wrote:
> In case of e.g writable systems there is no neccesity to have
writeable necessity

Quoted 5 lines

> pre-installed empty machine-id file on target because systemd > generates that automatically during system boot.
> Also in case of having empty machine-id file we are not able > to use a service with ConditionFirstBoot because systemd recognizes > machine-id file in system therefore we can't detect new system

instance

Quoted 2 lines

boots up.
> Set this option as enable by default to keep compatibility with

old builds.

Quoted 2 lines

Nice that you take a look at this! I'm immediately going to follow

it up with

Quoted 3 lines

more feature requests though :-)

- If there is a writeable persistent partition, but /etc is not

writeable, can

Quoted 1 line

we create a symlink and will systemd still create it correctly at

first boot?

Quoted 2 lines

- Maybe if BR2_TARGET_GENERIC_REMOUNT_ROOTFS_RW is set, we should

default it or

Quoted 22 lines

even force it to no.

Regards,
Arnout

>
> Signed-off-by: Bartosz Bilas <[email protected] <mailto:[email protected]>>
> ---
> package/systemd/Config.in | 13 +++++++++++++
> package/systemd/systemd.mk <http://systemd.mk> | 2 ++ > 2 files changed, 15 insertions(+)
>
> diff --git a/package/systemd/Config.in b/package/systemd/Config.in > index aef39abe27..fadc1a32c8 100644
> --- a/package/systemd/Config.in
> +++ b/package/systemd/Config.in
> @@ -112,6 +112,19 @@ config BR2_PACKAGE_SYSTEMD_BOOT_EFI_ARCH
> default "x64" if BR2_x86_64
> depends on BR2_PACKAGE_SYSTEMD_BOOT
>

Quoted 12 lines

> +config BR2_PACKAGE_SYSTEMD_MACHINEID_FILE
> + bool "Install empty machine id file"
> + default y
> + help
> + The /etc/machine-id file contains the unique machine ID
> + of the local system that is set during installation or
> + boot. The machine ID is a single newline-terminated,
> + hexadecimal, 32-character, lowercase ID. When decoded from
> + hexadecimal, this corresponds to a 16-byte/128-bit value.
> + This ID may not be all zeros.
> +
> +

https://www.freedesktop.org/software/systemd/man/machine-id.html

Quoted 12 lines

<http://systemd.mk>

Quoted 5 lines
package/systemd/Config.in +1 -0
+
 config BR2_PACKAGE_SYSTEMD_JOURNAL_GATEWAY
      bool "HTTP server for journal events"
      select BR2_PACKAGE_LIBMICROHTTPD
<http://systemd.mk> renamed from package/systemd/systemd.mk +0 -0
    b/package/systemd/systemd.mk <http://systemd.mk>
    > index 92490eb86b..fc348fe120 100644
    > --- a/package/systemd/systemd.mk <http://systemd.mk>
    > +++ b/package/systemd/systemd.mk <http://systemd.mk>
    > @@ -461,9 +461,11 @@ define SYSTEMD_INSTALL_INIT_HOOK
    >

$(TARGET_DIR)/etc/systemd/system/multi-user.target.wants/remote-fs.target

Quoted 28 lines

<http://systemd.mk>

Quoted 11 lines

endef

<http://systemd.mk> renamed from package/systemd/systemd.mk +2 -0
+ifeq ($(BR2_PACKAGE_SYSTEMD_MACHINEID_FILE),y)
 define SYSTEMD_INSTALL_MACHINEID_HOOK
      touch $(TARGET_DIR)/etc/machine-id
 endef
+endif
 
 SYSTEMD_POST_INSTALL_TARGET_HOOKS += \
      SYSTEMD_INSTALL_TARGET_CRYPTSETUP \

_______________________________________________
buildroot mailing list
buildroot at busybox.net <mailto:[email protected]>
http://lists.busybox.net/mailman/listinfo/buildroot

--
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

*J?r?my ROSEN*
Architecte technique

Quoted 7 lines

email jeremy.rosen at smile.fr <mailto:[email protected]> phone +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook
<https://www.facebook.com/smileopensource> LinkedIn
<https://www.linkedin.com/company/smile> Github <

https://github.com/Smile-SA>

Quoted 4 lines

D?couvrez l?univers Smile, rendez-vous sur smile.eu <

https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature

Quoted 1 line
Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191117/ac2ae46a/attachment.html>
#13 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Hi Je?re?my,

Thanks for your input!

Quoted 15 lines · J?r?my ROSEN

Argh, each time I simplify a bit, it comes back to bite me :)

Le?dim. 17 nov. 2019 ??17:00, Arnout Vandecappelle <[email protected] <mailto:[email protected]>> a ?crit?:

On 17/11/2019 15:19, J?r?my ROSEN wrote:
> hmmm, I'm kinda torn on that one..
>
> The only difference I see betweed no machine-id and an empty machine-id is > systemd-firstboot and ConditionFirstBoot

?That's exactly the point I think.

Quoted 8 lines

?I assumed, however, that ConditionFirstBoot was evaluated by systemd-firstboot, but that's not the case then?

no, all Condition* are handled by PID1.? systemd-firstboot.service itself relies on ConditionFirstBoot to decide if the binary needs to be started

Of course, that makes a lot of sense if you think about it. I should do more of
that :-)

Quoted 8 lines
Quoted 1 line

if the file is empty or missing and systemd can write to it, systemd will

set it
> up (not sure what happens for Arnout's symlink, that's worth testing) > however, if the file exists but is empty we will skip systemd-firstboot >
> for context, systemd-firstboot does the following:
> * set (unconditionnaly) a machine-id

Quoted 8 lines

?Are you saying that if systemd-firstboot it not enabled, no machine-id will be generated?

no, it's a bit more complex, and i'd need to check the details.

my understanding is that systemd-firstboot will always override the machine-id whereas on a normal boot,?
the machine-id will be used if available and generated/written if not

That also makes sense.

[snip]

Quoted 11 lines

?Actually, come to think of it, maybe it's better to never create the machine-id at all. What is the effect (with a readonly rootfs) if machine-id doesn't exist? Does it block booting, does the boot take a little longer, or do we boot with a different machine-id all the time? And in the latter case, is there a real problem with that?

on a RO filesystem
* if /etc/machine-id exists and contains a machine-id, systemd will use it
* if it's an empty file, systemd will bind-mount a file with a machine-id over the file

Where I suppose the machine-id is regenerated on every boot.

Quoted 2 lines

* if the file does not exist, systemd can't bind mount over the file... systemd itself should be fine, but other services might not, it's hard to tell

That's not good...

Quoted 4 lines

So the recommended?way for a read-only rootfs is to have that empty file, which is why this is what currently have.
it works both for RW and RO filesystems..

Another option would be to make it a symlink to a file in tmpfs, but that
doesn't help because the net effect is the same: machine-id is regenerated on every boot :-(

Quoted 3 lines

A longer-term solution would be to add a patch upstream to allow to specify in a systemd configuration file what is?
the file to check for first boot instead of /etc/machine-id.

No, that makes very little sense. If a system has any kind of persistency, then
machine-id should definitely be one of the things that is persisted, so it's a perfect flag for firstboot.

Upstream's vision is that /etc should be persistent and empty on factory reset
[1]. But we're far away from having an empty /etc in Buildroot...

Quoted 8 lines

I can look into that, but for the short term, i'd?rather break ConditionFirstBoot than read-only rootfs

so... not sure what to advise at that point. that choice is something a bit too specialized for a config option
especially if we plan to drop it in the future

so not sure where to go...

As is often the case: it looks to me that the status quo is the best we can do :-)

Regards,
Arnout

[1] http://0pointer.net/blog/projects/stateless.html

[snip]

#14 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Quoted 23 lines

[snip]

Quoted 1 line

Actually, come to think of it, maybe it's better to never create

the machine-id

Quoted 1 line

at all. What is the effect (with a readonly rootfs) if machine-id

doesn't exist?

Quoted 1 line

Does it block booting, does the boot take a little longer, or do we

boot with a

Quoted 1 line

different machine-id all the time? And in the latter case, is there

a real

Quoted 5 lines

problem with that?

on a RO filesystem
* if /etc/machine-id exists and contains a machine-id, systemd will use

it

Quoted 1 line

* if it's an empty file, systemd will bind-mount a file with a

machine-id over

Quoted 1 line

the file

Where I suppose the machine-id is regenerated on every boot.

Quoted 9 lines

yes, which is fine, that's the intended behaviour

Quoted 1 line

* if the file does not exist, systemd can't bind mount over the file...

systemd

Quoted 1 line

itself should be fine, but other services might not, it's hard to tell

That's not good...

I just tested, and systemd logs an error message that it can't boot in that case.

more precisely, when systemd starts
* either /etc is readwrite
* or it should have a /etc/machine-id
so we need the file, since our rootfs is read-only when systemd starts
* it's just an error message, everything turns out fine once we remount rw and we save the machine-id only
* sounds like a bug from upstream... read-only rootfs and read-only rootfs that become RW are supposed to be supported

Quoted 13 lines
Quoted 3 lines

So the recommended way for a read-only rootfs is to have that empty

file, which

Quoted 2 lines

is why this is what currently have.
it works both for RW and RO filesystems..

Another option would be to make it a symlink to a file in tmpfs, but that
doesn't help because the net effect is the same: machine-id is regenerated on
every boot :-(

regenerated every boot is fine, i'm just not sure in what way a symlink to a tmpfs would be different/better than what systemd currently does

symlink to a r/w partition makes sense, but that would not work for ConditionFirstBoot

Quoted 6 lines
Quoted 2 lines

A longer-term solution would be to add a patch upstream to allow to

specify in a

Quoted 2 lines

systemd configuration file what is
the file to check for first boot instead of /etc/machine-id.

Quoted 7 lines

No, that makes very little sense. If a system has any kind of
persistency, then
machine-id should definitely be one of the things that is persisted, so it's a
perfect flag for firstboot.

no because it makes sense to have it be a symlink to a RW file... in which

case it exists even in first boot..

Quoted 5 lines

Upstream's vision is that /etc should be persistent and empty on factory
reset
[1]. But we're far away from having an empty /etc in Buildroot...

yeah, but I think the assumption empty /etc == first boot is wrong...

in the real world, the assumption is no /etc/machine-id == first boot makes more sense, but it doesn't work for us.
No... upstream patch is really the way to go (and i'm already working on it :P )

Quoted 6 lines
Quoted 6 lines

I can look into that, but for the short term, i'd rather break ConditionFirstBoot than read-only rootfs

so... not sure what to advise at that point. that choice is something a

Quoted 9 lines

bit too

Quoted 4 lines

specialized for a config option
especially if we plan to drop it in the future

so not sure where to go...

As is often the case: it looks to me that the status quo is the best we
can do :-)

it could be slightly improved by removing /etc/machine-id when systemd-firstboot is selected, but I agree...
Until we can change and configure how to detect first-boot, we are kinda stuck...

Quoted 8 lines

Regards,
Arnout

[1] http://0pointer.net/blog/projects/stateless.html

[snip]

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191117/07e9350f/attachment.html>
#15 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

In reply to Arnout Vandecappelle

Hello,

Quoted 7 lines · On Sun, 17 Nov 2019 17:00:05 +0100 Arnout Vandecappelle <[email protected]> wrote:
Quoted 3 lines

Could something like that be enough ? can we trust "remount RW" ? maybe "remount RW" should be renamed "create a RW filesystem" and enable various tweaks related to RO vs RW

As written above: no.

The problem is: we're not a distro.

Agreed.

Quoted 9 lines

We leave too much freedom for the user to
integrate things in various ways to be able to make assumptions about what is the right way to do things. So, the only thing we can do is to give a decent out-of-the-box experience, and let the user figure out how to tweak things - possibly adding a config option for a common situation that is easily handled in a generic way. The other thing we can do is to provide documentation about the proper way to integrate things in different scenarios.

I'm starting to agree that this option is maybe not that great.

But I would in fact not come to the same conclusion. Having this empty machine-id file is useless and causes problems when the filesystem is R/W. So for the sake of supporting the R/O case (for which we create this empty machine-id file), we make the R/W experience less good.

So I'd say that the right approach is to not do too much integration by precisely having the option proposed by Bartosz, with many the tweak that it should default y if rootfs is really, and default disable otherwise, but while still being an option that the user can tweak, because as you rightfully explained, the RW/RO remount option is just a clue, not a definitive answer on whether /etc is writable or not.

To me, having this option matches the Buildroot way: we are not a distro, we don't enforce how the system should work, so we provide the appropriate options, while making sure the option has the most sensible default values.

Generally speaking, Buildroot kind of supports "out of the box" two use cases:

(1) The root filesystem is completely read-write.

(2) The root filesystem is completely read-only, and all files that need to be written are stored in tmpfs, and therefore are volatile.

I.e, we do not have any explicit support for what is I guess a much more common use case than (2):

(3) The root filesystem is completely read-only, but there is another read-write partition somewhere that stores the information that can change but needs to be persistent (user configuration, etc.)

Since we don't have explicit support for (3), there is no way we can properly support machine-id and ConditionFirstBoot in the case of (2), because there's nowhere we can store /etc/machine-id.

So the best we can do is in the case of (2), default to creating an empty /etc/machine-id, while giving the possibility for the user implementing (3) in its own way, to disable the creation of the empty /etc/machine-id.

Best regards,

Thomas

Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
#16 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

As a side-note, I am working with upstream to have a better support of (3) : https://github.com/systemd/systemd/pull/14059

I am a bit cautious about the new config option because it seems too "advanced" for a config option (for me, it's something that should be set in a post-image or overlay) but that's open to discussion.

Please wait a little before applying this patch, if that's the way buildroot wants to go, so my pull-request above is solved and we might backport it to ease our transition.

Cheers
J?r?my

Le mar. 19 nov. 2019 ? 09:40, Thomas Petazzoni <[email protected]> a ?crit :

Quoted 80 lines

Hello,

Quoted 2 lines · On Sun, 17 Nov 2019 17:00:05 +0100 Arnout Vandecappelle <[email protected]> wrote:
Quoted 2 lines

Could something like that be enough ? can we trust "remount RW" ? maybe "remount RW" should be renamed "create a RW filesystem" and

enable various

Quoted 5 lines
Quoted 1 line

tweaks related to RO vs RW

As written above: no.

The problem is: we're not a distro.

Agreed.

Quoted 2 lines

We leave too much freedom for the user to
integrate things in various ways to be able to make assumptions about

what is

Quoted 1 line

the right way to do things. So, the only thing we can do is to give a

decent

Quoted 1 line

out-of-the-box experience, and let the user figure out how to tweak

things -

Quoted 1 line

possibly adding a config option for a common situation that is easily

handled in

Quoted 1 line

a generic way. The other thing we can do is to provide documentation

about the

Quoted 3 lines

proper way to integrate things in different scenarios.

I'm starting to agree that this option is maybe not that great.

But I would in fact not come to the same conclusion. Having this empty machine-id file is useless and causes problems when the filesystem is R/W. So for the sake of supporting the R/O case (for which we create this empty machine-id file), we make the R/W experience less good.

So I'd say that the right approach is to not do too much integration by precisely having the option proposed by Bartosz, with many the tweak that it should default y if rootfs is really, and default disable otherwise, but while still being an option that the user can tweak, because as you rightfully explained, the RW/RO remount option is just a clue, not a definitive answer on whether /etc is writable or not.

To me, having this option matches the Buildroot way: we are not a distro, we don't enforce how the system should work, so we provide the appropriate options, while making sure the option has the most sensible default values.

Generally speaking, Buildroot kind of supports "out of the box" two use cases:

(1) The root filesystem is completely read-write.

(2) The root filesystem is completely read-only, and all files that need to be written are stored in tmpfs, and therefore are volatile.

I.e, we do not have any explicit support for what is I guess a much more common use case than (2):

(3) The root filesystem is completely read-only, but there is another read-write partition somewhere that stores the information that can change but needs to be persistent (user configuration, etc.)

Since we don't have explicit support for (3), there is no way we can properly support machine-id and ConditionFirstBoot in the case of (2), because there's nowhere we can store /etc/machine-id.

So the best we can do is in the case of (2), default to creating an empty /etc/machine-id, while giving the possibility for the user implementing (3) in its own way, to disable the creation of the empty /etc/machine-id.

Best regards,

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20191119/4d5e4cd1/attachment.html>
#17 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Hello everyone,

let's reject this patch due to possible upstream solution as Jeremy mentioned.

Best
Bartek

Quoted 96 lines · J?r?my ROSEN

As a side-note, I am working with upstream to have a better support of
(3) : https://github.com/systemd/systemd/pull/14059

I am a bit cautious about the new config option because it seems too "advanced" for a config option (for me, it's something that should be set in a post-image or overlay) but that's open to discussion.

Please wait a little before applying this patch, if that's the way buildroot wants to go, so my pull-request above is solved and we might backport it to ease our transition.

Cheers
J?r?my

Le?mar. 19 nov. 2019 ??09:40, Thomas Petazzoni <thomas.petazzoni at bootlin.com <mailto:[email protected]>> a ?crit?:

Hello,

On Sun, 17 Nov 2019 17:00:05 +0100
Arnout Vandecappelle <arnout at mind.be <mailto:[email protected]>> wrote:

> > Could something like that be enough ? can we trust "remount RW" ? > > maybe "remount RW" should be renamed "create a RW filesystem" and enable various
> > tweaks related to RO vs RW
>
>? As written above: no.
>
>? The problem is: we're not a distro.

Agreed.

> We leave too much freedom for the user to
> integrate things in various ways to be able to make assumptions about what is
> the right way to do things. So, the only thing we can do is to give a decent
> out-of-the-box experience, and let the user figure out how to tweak things -
> possibly adding a config option for a common situation that is easily handled in
> a generic way. The other thing we can do is to provide documentation about the
> proper way to integrate things in different scenarios. >
>? I'm starting to agree that this option is maybe not that great.

But I would in fact not come to the same conclusion. Having this empty machine-id file is useless and causes problems when the filesystem is R/W. So for the sake of supporting the R/O case (for which we create this empty machine-id file), we make the R/W experience less good.

So I'd say that the right approach is to not do too much integration by
precisely having the option proposed by Bartosz, with many the tweak that it should default y if rootfs is really, and default disable otherwise, but while still being an option that the user can tweak, because as you rightfully explained, the RW/RO remount option is just a
clue, not a definitive answer on whether /etc is writable or not.

To me, having this option matches the Buildroot way: we are not a distro, we don't enforce how the system should work, so we provide the appropriate options, while making sure the option has the most sensible
default values.

Generally speaking, Buildroot kind of supports "out of the box" two use
cases:

?(1) The root filesystem is completely read-write.

?(2) The root filesystem is completely read-only, and all files that need
? ? ?to be written are stored in tmpfs, and therefore are volatile.

I.e, we do not have any explicit support for what is I guess a much more common use case than (2):

?(3) The root filesystem is completely read-only, but there is another ? ? ?read-write partition somewhere that stores the information that can
? ? ?change but needs to be persistent (user configuration, etc.)

Since we don't have explicit support for (3), there is no way we can properly support machine-id and ConditionFirstBoot in the case of (2), because there's nowhere we can store /etc/machine-id.

So the best we can do is in the case of (2), default to creating an empty /etc/machine-id, while giving the possibility for the user implementing (3) in its own way, to disable the creation of the empty /etc/machine-id.

Quoted 32 lines

Best regards,

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

Signature · 21 lines
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]>
phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook 
<https://www.facebook.com/smileopensource> LinkedIn 
<https://www.linkedin.com/company/smile> Github 
<https://github.com/Smile-SA>


D?couvrez l?univers Smile, rendez-vous sur smile.eu 
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20200203/3c4a0a7b/attachment.html>

#18 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Hi everyone,

I had finally a couple of time to double-check that and I agree with Jeremy that I was wrong therefore let's reject this patch as it turned out as an unthought idea.

Best
Bartek

Quoted 138 lines · Bartosz Bilas

Hello everyone,

let's reject this patch due to possible upstream solution as Jeremy mentioned.

Best
Bartek

Quoted 130 lines · J?r?my ROSEN

As a side-note, I am working with upstream to have a better support of (3) : https://github.com/systemd/systemd/pull/14059

I am a bit cautious about the new config option because it seems too "advanced" for a config option (for me, it's something that should be set in a post-image or overlay) but that's open to discussion.

Please wait a little before applying this patch, if that's the way buildroot wants to go, so my pull-request above is solved and we might backport it to ease our transition.

Cheers
J?r?my

Le?mar. 19 nov. 2019 ??09:40, Thomas Petazzoni <thomas.petazzoni at bootlin.com <mailto:[email protected]>> a ?crit?:

Hello,

On Sun, 17 Nov 2019 17:00:05 +0100
Arnout Vandecappelle <arnout at mind.be <mailto:[email protected]>> wrote:

> > Could something like that be enough ? can we trust "remount RW" ? > > maybe "remount RW" should be renamed "create a RW filesystem" and enable various
> > tweaks related to RO vs RW
>
>? As written above: no.
>
>? The problem is: we're not a distro.

Agreed.

> We leave too much freedom for the user to
> integrate things in various ways to be able to make assumptions about what is
> the right way to do things. So, the only thing we can do is to give a decent
> out-of-the-box experience, and let the user figure out how to tweak things -
> possibly adding a config option for a common situation that is easily handled in
> a generic way. The other thing we can do is to provide documentation about the
> proper way to integrate things in different scenarios. >
>? I'm starting to agree that this option is maybe not that great.

But I would in fact not come to the same conclusion. Having this empty
machine-id file is useless and causes problems when the filesystem is R/W. So for the sake of supporting the R/O case (for which we create this empty machine-id file), we make the R/W experience less good.

So I'd say that the right approach is to not do too much integration by
precisely having the option proposed by Bartosz, with many the tweak that it should default y if rootfs is really, and default disable otherwise, but while still being an option that the user can tweak, because as you rightfully explained, the RW/RO remount option is just a
clue, not a definitive answer on whether /etc is writable or not.

To me, having this option matches the Buildroot way: we are not a distro, we don't enforce how the system should work, so we provide the
appropriate options, while making sure the option has the most sensible
default values.

Generally speaking, Buildroot kind of supports "out of the box" two use
cases:

?(1) The root filesystem is completely read-write.

?(2) The root filesystem is completely read-only, and all files that need
? ? ?to be written are stored in tmpfs, and therefore are volatile.

I.e, we do not have any explicit support for what is I guess a much more common use case than (2):

?(3) The root filesystem is completely read-only, but there is another
? ? ?read-write partition somewhere that stores the information that can
? ? ?change but needs to be persistent (user configuration, etc.)

Since we don't have explicit support for (3), there is no way we can properly support machine-id and ConditionFirstBoot in the case of (2),
because there's nowhere we can store /etc/machine-id.

So the best we can do is in the case of (2), default to creating an empty /etc/machine-id, while giving the possibility for the user implementing (3) in its own way, to disable the creation of the empty /etc/machine-id.

Best regards,

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

Signature · 17 lines
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]>
phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook 
<https://www.facebook.com/smileopensource> LinkedIn 
<https://www.linkedin.com/company/smile> Github 
<https://github.com/Smile-SA>
Quoted 7 lines
Quoted 2 lines

D?couvrez l?univers Smile, rendez-vous sur smile.eu
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20200203/e21e22b7/attachment-0001.html>

#19 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

Nah, sorry guys - I've been mistaken for the email so please ignore the previous message then.

Best
Bartek

Quoted 161 lines · Bartosz Bilas

Hi everyone,

I had finally a couple of time to double-check that and I agree with Jeremy that I was wrong therefore let's reject this patch as it turned out as an unthought idea.

Best
Bartek

Quoted 142 lines · Bartosz Bilas

Hello everyone,

let's reject this patch due to possible upstream solution as Jeremy mentioned.

Best
Bartek

Quoted 134 lines · J?r?my ROSEN

As a side-note, I am working with upstream to have a better support of (3) : https://github.com/systemd/systemd/pull/14059

I am a bit cautious about the new config option because it seems too "advanced" for a config option (for me, it's something that should be set
in a post-image or overlay) but that's open to discussion.

Please wait a little before applying this patch, if that's the way buildroot wants to go, so my pull-request above is solved and we might backport it to ease our transition.

Cheers
J?r?my

Le?mar. 19 nov. 2019 ??09:40, Thomas Petazzoni <thomas.petazzoni at bootlin.com <mailto:[email protected]>> a ?crit?:

Hello,

On Sun, 17 Nov 2019 17:00:05 +0100
Arnout Vandecappelle <arnout at mind.be <mailto:[email protected]>> wrote:

> > Could something like that be enough ? can we trust "remount RW" ?
> > maybe "remount RW" should be renamed "create a RW filesystem" and enable various
> > tweaks related to RO vs RW
>
>? As written above: no.
>
>? The problem is: we're not a distro.

Agreed.

> We leave too much freedom for the user to
> integrate things in various ways to be able to make assumptions about what is
> the right way to do things. So, the only thing we can do is to give a decent
> out-of-the-box experience, and let the user figure out how to tweak things -
> possibly adding a config option for a common situation that is easily handled in
> a generic way. The other thing we can do is to provide documentation about the
> proper way to integrate things in different scenarios. >
>? I'm starting to agree that this option is maybe not that great.

But I would in fact not come to the same conclusion. Having this empty
machine-id file is useless and causes problems when the filesystem is
R/W. So for the sake of supporting the R/O case (for which we create this empty machine-id file), we make the R/W experience less good.

So I'd say that the right approach is to not do too much integration by
precisely having the option proposed by Bartosz, with many the tweak that it should default y if rootfs is really, and default disable otherwise, but while still being an option that the user can tweak, because as you rightfully explained, the RW/RO remount option is just a
clue, not a definitive answer on whether /etc is writable or not.

To me, having this option matches the Buildroot way: we are not a distro, we don't enforce how the system should work, so we provide the
appropriate options, while making sure the option has the most sensible
default values.

Generally speaking, Buildroot kind of supports "out of the box" two use
cases:

?(1) The root filesystem is completely read-write.

?(2) The root filesystem is completely read-only, and all files that need
? ? ?to be written are stored in tmpfs, and therefore are volatile.

I.e, we do not have any explicit support for what is I guess a much more common use case than (2):

?(3) The root filesystem is completely read-only, but there is another
? ? ?read-write partition somewhere that stores the information that can
? ? ?change but needs to be persistent (user configuration, etc.)

Since we don't have explicit support for (3), there is no way we can properly support machine-id and ConditionFirstBoot in the case of (2),
because there's nowhere we can store /etc/machine-id.

So the best we can do is in the case of (2), default to creating an empty /etc/machine-id, while giving the possibility for the user implementing (3) in its own way, to disable the creation of the empty
/etc/machine-id.

Best regards,

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

Signature · 17 lines
SMILE <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine

	
*J?r?my ROSEN*
Architecte technique

email jeremy.rosen at smile.fr <mailto:[email protected]>
phone? +33 6 88 25 87 42
url http://www.smile.eu <http://www.smile.eu/>

Twitter <https://twitter.com/GroupeSmile> Facebook 
<https://www.facebook.com/smileopensource> LinkedIn 
<https://www.linkedin.com/company/smile> Github 
<https://github.com/Smile-SA>
Quoted 7 lines
Quoted 2 lines

D?couvrez l?univers Smile, rendez-vous sur smile.eu
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Quoted 4 lines

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20200203/40ed1c57/attachment.html>

#20 lore raw

PATCH 1/2 package/systemd: set machine id file instalation as optional choice

So don't drop your patch ?

I still think that the approach I proposed upstream is the best one but I am pessimistic about the chances of upsreaming...
I'll ping the issue, but don't hold your breath.

If this is rejected upstream, we might want to keep the patch locally, but that's a subject to discuss when we reach that point.

Le lun. 3 f?vr. 2020 ? 19:41, Bartosz Bilas <[email protected]> a ?crit :

Quoted 155 lines

Nah, sorry guys - I've been mistaken for the email so please ignore the previous message then.

Best
Bartek
On 03.02.2020 19:34, Bartosz Bilas wrote:

Hi everyone,

I had finally a couple of time to double-check that and I agree with Jeremy that I was wrong therefore let's reject this patch as it turned out as an unthought idea.

Best
Bartek
On 03.02.2020 19:28, Bartosz Bilas wrote:

Hello everyone,

let's reject this patch due to possible upstream solution as Jeremy mentioned.
Best
Bartek
On 19.11.2019 11:15, J?r?my ROSEN wrote:

As a side-note, I am working with upstream to have a better support of (3) : https://github.com/systemd/systemd/pull/14059

I am a bit cautious about the new config option because it seems too "advanced" for a config option (for me, it's something that should be set in a post-image or overlay) but that's open to discussion.

Please wait a little before applying this patch, if that's the way buildroot wants to go, so my pull-request above is solved and we might backport it to ease our transition.

Cheers
J?r?my

Le mar. 19 nov. 2019 ? 09:40, Thomas Petazzoni <
thomas.petazzoni at bootlin.com> a ?crit :

Quoted 72 lines

Hello,

Quoted 2 lines · On Sun, 17 Nov 2019 17:00:05 +0100 Arnout Vandecappelle <[email protected]> wrote:
Quoted 2 lines

Could something like that be enough ? can we trust "remount RW" ? maybe "remount RW" should be renamed "create a RW filesystem" and

enable various

Quoted 5 lines
Quoted 1 line

tweaks related to RO vs RW

As written above: no.

The problem is: we're not a distro.

Agreed.

Quoted 2 lines

We leave too much freedom for the user to
integrate things in various ways to be able to make assumptions about

what is

Quoted 1 line

the right way to do things. So, the only thing we can do is to give a

decent

Quoted 1 line

out-of-the-box experience, and let the user figure out how to tweak

things -

Quoted 1 line

possibly adding a config option for a common situation that is easily

handled in

Quoted 1 line

a generic way. The other thing we can do is to provide documentation

about the

Quoted 3 lines

proper way to integrate things in different scenarios.

I'm starting to agree that this option is maybe not that great.

But I would in fact not come to the same conclusion. Having this empty machine-id file is useless and causes problems when the filesystem is R/W. So for the sake of supporting the R/O case (for which we create this empty machine-id file), we make the R/W experience less good.

So I'd say that the right approach is to not do too much integration by precisely having the option proposed by Bartosz, with many the tweak that it should default y if rootfs is really, and default disable otherwise, but while still being an option that the user can tweak, because as you rightfully explained, the RW/RO remount option is just a clue, not a definitive answer on whether /etc is writable or not.

To me, having this option matches the Buildroot way: we are not a distro, we don't enforce how the system should work, so we provide the appropriate options, while making sure the option has the most sensible default values.

Generally speaking, Buildroot kind of supports "out of the box" two use cases:

(1) The root filesystem is completely read-write.

(2) The root filesystem is completely read-only, and all files that need to be written are stored in tmpfs, and therefore are volatile.

I.e, we do not have any explicit support for what is I guess a much more common use case than (2):

(3) The root filesystem is completely read-only, but there is another read-write partition somewhere that stores the information that can change but needs to be persistent (user configuration, etc.)

Since we don't have explicit support for (3), there is no way we can properly support machine-id and ConditionFirstBoot in the case of (2), because there's nowhere we can store /etc/machine-id.

So the best we can do is in the case of (2), default to creating an empty /etc/machine-id, while giving the possibility for the user implementing (3) in its own way, to disable the creation of the empty /etc/machine-id.

Quoted 8 lines

Best regards,

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

--
[image: SMILE] <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone] +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>

_______________________________________________
buildroot mailing listbuildroot at busybox.nethttp://lists.busybox.net/mailman/listinfo/buildroot

_______________________________________________
buildroot mailing listbuildroot at busybox.nethttp://lists.busybox.net/mailman/listinfo/buildroot

Signature · 21 lines
[image: SMILE]  <http://www.smile.eu/>

20 rue des Jardins
92600 Asni?res-sur-Seine
*J?r?my ROSEN*
Architecte technique

[image: email] jeremy.rosen at smile.fr
[image: phone]  +33 6 88 25 87 42
[image: url] http://www.smile.eu

[image: Twitter] <https://twitter.com/GroupeSmile> [image: Facebook]
<https://www.facebook.com/smileopensource> [image: LinkedIn]
<https://www.linkedin.com/company/smile> [image: Github]
<https://github.com/Smile-SA>

[image: D?couvrez l?univers Smile, rendez-vous sur smile.eu]
<https://www.smile.eu/fr/publications/livres-blancs/yocto?utm_source=signature&utm_medium=email&utm_campaign=signature>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20200205/d16b4a3f/attachment.html>
#21 lore raw
In reply to Bartosz Bilas

Hello Bartosz,

Quoted 8 lines · On Sun, 17 Nov 2019 12:33:44 +0100 Bartosz Bilas <[email protected]>…

In case of e.g writable systems there is no neccesity to have pre-installed empty machine-id file on target because systemd generates that automatically during system boot.
Also in case of having empty machine-id file we are not able
to use a service with ConditionFirstBoot because systemd recognizes machine-id file in system therefore we can't detect new system instance boots up. Set this option as enable by default to keep compatibility with old builds.

Quoted 5 lines
---
 package/systemd/Config.in  | 13 +++++++++++++
 package/systemd/systemd.mk |  2 ++
 2 files changed, 15 insertions(+)

This series of two patches has been sitting on the mailing list for a long time. The feedback of Jérémy was not very positive (i.e quite negative) and you said yourself on this patch "I had finally a couple of time to double-check that and I agree with Jeremy that I was wrong therefore let's reject this patch as it turned out as an unthought idea.", but then you said you were mistaken, Jeremy asked for clarifications which were never given.

So at this point, I'm going to mark those two patches as Rejected. However, if you believe there is still something to improve, do not hesitate to reopen the discussion.

Thanks a lot!

Thomas

Signature · 7 lines
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
_______________________________________________
buildroot mailing list
[email protected]
http://lists.busybox.net/mailman/listinfo/buildroot