initscripts 2009.11-1, device-mapper/lvm2 2.02.54-2, udev 146-3
 

On Fri, Nov 6, 2009 at 6:54 PM, Thomas Bächler <thomas@archlinux.org> wrote:
> This upload brings some changes to all device-mapper based applications and
> should be the start of fixing some race conditions (also needs a fix to
> devicekit's udev rules).
>
> For this email, the lvm and udev changes are more important: During the
> initial udevtrigger, the STARTUP=1 environment variable will be set in all
> uevents. This is used by the new device-mapper/lvm2 upstream udev rules
> (which replace our homegrown and outdated rule) which have to distinguish
> between the initial triggering and later execution of uevents.
>
> The device-mapper rules also set some flags that will tell other rules to
> ignore a dm device (these flags have to be respected in devicekit to solve
> some race conditions). The only change in udev is that the I removed a hunk
> from the arch patch that un-ignored dm-* devices: Peristent names for those
> are now handles in the device-mapper rules, respecting the ignore flags and
> thus also avoiding weird races conditions.
>
> If you found this confusing, ask, otherwise please test and sign off.
> (Tested and working on x86_64 with encryption and lvm2)
>
>

FYI: There's a problem with lilo and the new device-mapper when your
/boot is on a LVM volume. Basically, lilo fails to load the kernel.
Thomas and I are looking into it.

Eric

initscripts 2009.11-1, device-mapper/lvm2 2.02.54-2, udev 146-3
 

On Fri, Nov 6, 2009 at 9:11 PM, Eric Bélanger <snowmaniscool@gmail.com> wrote:
> On Fri, Nov 6, 2009 at 6:54 PM, Thomas Bächler <thomas@archlinux.org> wrote:
>> This upload brings some changes to all device-mapper based applications and
>> should be the start of fixing some race conditions (also needs a fix to
>> devicekit's udev rules).
>>
>> For this email, the lvm and udev changes are more important: During the
>> initial udevtrigger, the STARTUP=1 environment variable will be set in all
>> uevents. This is used by the new device-mapper/lvm2 upstream udev rules
>> (which replace our homegrown and outdated rule) which have to distinguish
>> between the initial triggering and later execution of uevents.
>>
>> The device-mapper rules also set some flags that will tell other rules to
>> ignore a dm device (these flags have to be respected in devicekit to solve
>> some race conditions). The only change in udev is that the I removed a hunk
>> from the arch patch that un-ignored dm-* devices: Peristent names for those
>> are now handles in the device-mapper rules, respecting the ignore flags and
>> thus also avoiding weird races conditions.
>>
>> If you found this confusing, ask, otherwise please test and sign off.
>> (Tested and working on x86_64 with encryption and lvm2)
>>
>>
>
> FYI: There's a problem with lilo and the new device-mapper when your
> /boot is on a LVM volume. Basically, lilo fails to load the kernel.
> Thomas and I are looking into it.
>
> Eric
>

I've just pushed a fixed lilo-22.8-4 in testing. I only tested the i686 package.

BTW, I'm also having problems with my x86_64 system which use raid and lvm2. :(
It can't find the root device so I'm dropped in the initramfs fallback
shell. I suspect that the problem is related to the initramfs as I
can boot with the fallback image.

initscripts 2009.11-1, device-mapper/lvm2 2.02.54-2, udev 146-3
 

On Sat, Nov 7, 2009 at 01:54, Thomas Bächler <thomas@archlinux.org> wrote:
> This upload brings some changes to all device-mapper based applications and
> should be the start of fixing some race conditions (also needs a fix to
> devicekit's udev rules).
>
> Let's start with initscripts. There some outstanding issues as always, but
> we should rather release than wait for me to think about all of them:
>
> Aaron Griffin (7):
> Â* Â* Â*Replace e with 33 for printfs
> Â* Â* Â*network: Use real interface name for ifconfig up
> Â* Â* Â*Always shutdown daemons, regardless of runlevel
> Â* Â* Â*network: support early 'up' for dhcp interfaces
> Â* Â* Â*Add hvc0 to inittab (commented by default)
> Â* Â* Â*Empty mtab instead of deleting it
> Â* Â* Â*Replace 'truncate file' logic in case of noclobber
>
> Jim Pryor (1):
> Â* Â* Â*Fix stty/tput usage for columns again
>
> Thomas Bächler (12):
> Â* Â* Â*Add size and noexec mount options to /dev, change "none" to "udev"
> Â* Â* Â*Use nosuid instead of noexec on /dev
> Â* Â* Â*Remove uevent processing benchmark, as it is useless now that we
> actually do work between trigger and settle
> Â* Â* Â*Remove support for static filesystem and make code in rc.single look
> like the code in rc.sysinit/rc.shutdown
> Â* Â* Â*Implement a hook-system that allows to add custom code to the
> initscripts at certain places
> Â* Â* Â*Make the hook-system a bit more usable
> Â* Â* Â*Only load rtc-cmos if rc.conf has proper clock settings
> Â* Â* Â*Disable special handling of ACPI modules
> Â* Â* Â*Remove code to create md* devices manually, this is done by udev these
> days
> Â* Â* Â*Set STARTUP=1 environment variable on initial udev triggering, required
> by latest device-mapper rules
> Â* Â* Â*Correct last commit: Unset the STARTUP=1 variable in udevadm control
> only after udev has settled
> Â* Â* Â*Remove redundant call to vgscan - this also fixes a "wrong" symlink in
> combination with the new udev-based device mapper node creation
>
> First I'd like to highlight the hooks-system. It can be used by splash
> daemons or live systems or any user to hook in special functions during the
> boot process. It is not really documented, there's some comments in
> rc.functions on using it.
>
> For this email, the lvm and udev changes are more important: During the
> initial udevtrigger, the STARTUP=1 environment variable will be set in all
> uevents. This is used by the new device-mapper/lvm2 upstream udev rules
> (which replace our homegrown and outdated rule) which have to distinguish
> between the initial triggering and later execution of uevents.
>
> The device-mapper rules also set some flags that will tell other rules to
> ignore a dm device (these flags have to be respected in devicekit to solve
> some race conditions). The only change in udev is that the I removed a hunk
> from the arch patch that un-ignored dm-* devices: Peristent names for those
> are now handles in the device-mapper rules, respecting the ignore flags and
> thus also avoiding weird races conditions.
>
> If you found this confusing, ask, otherwise please test and sign off.
> (Tested and working on x86_64 with encryption and lvm2)

I hit a weird bug.
When my USB 3G modem is plugged in during boot
I get this error for every encrypted partition except /:
Command failed: /dev/mapper/temporary-cryptsetup-1345 open filed: No
such file or directory
When the modem is not plugged during boot everything is okay.

Ideas?

--
Roman Kyrylych (Ð*оман Кирилич)

initscripts 2009.11-1, device-mapper/lvm2 2.02.54-2, udev 146-3
 

Roman Kyrylych schrieb:

I hit a weird bug.
When my USB 3G modem is plugged in during boot
I get this error for every encrypted partition except /:
Command failed: /dev/mapper/temporary-cryptsetup-1345 open filed: No
such file or directory
When the modem is not plugged during boot everything is okay.

Ideas?


None, please take part in
http://bugs.archlinux.org/task/16735

initscripts 2009.11-1, device-mapper/lvm2 2.02.54-2, udev 146-3
 

On Fri, Nov 6, 2009 at 6:54 PM, Thomas Bächler <thomas@archlinux.org> wrote:
> This upload brings some changes to all device-mapper based applications and
> should be the start of fixing some race conditions (also needs a fix to
> devicekit's udev rules).
>
> Let's start with initscripts. There some outstanding issues as always, but
> we should rather release than wait for me to think about all of them:
>
> Aaron Griffin (7):
> * * *Replace e with 33 for printfs
> * * *network: Use real interface name for ifconfig up
> * * *Always shutdown daemons, regardless of runlevel
> * * *network: support early 'up' for dhcp interfaces
> * * *Add hvc0 to inittab (commented by default)
> * * *Empty mtab instead of deleting it
> * * *Replace 'truncate file' logic in case of noclobber
>
> Jim Pryor (1):
> * * *Fix stty/tput usage for columns again
>
> Thomas Bächler (12):
> * * *Add size and noexec mount options to /dev, change "none" to "udev"
> * * *Use nosuid instead of noexec on /dev
> * * *Remove uevent processing benchmark, as it is useless now that we
> actually do work between trigger and settle
> * * *Remove support for static filesystem and make code in rc.single look
> like the code in rc.sysinit/rc.shutdown
> * * *Implement a hook-system that allows to add custom code to the
> initscripts at certain places
> * * *Make the hook-system a bit more usable
> * * *Only load rtc-cmos if rc.conf has proper clock settings
> * * *Disable special handling of ACPI modules
> * * *Remove code to create md* devices manually, this is done by udev these
> days
> * * *Set STARTUP=1 environment variable on initial udev triggering, required
> by latest device-mapper rules
> * * *Correct last commit: Unset the STARTUP=1 variable in udevadm control
> only after udev has settled
> * * *Remove redundant call to vgscan - this also fixes a "wrong" symlink in
> combination with the new udev-based device mapper node creation
>
> First I'd like to highlight the hooks-system. It can be used by splash
> daemons or live systems or any user to hook in special functions during the
> boot process. It is not really documented, there's some comments in
> rc.functions on using it.
>
> For this email, the lvm and udev changes are more important: During the
> initial udevtrigger, the STARTUP=1 environment variable will be set in all
> uevents. This is used by the new device-mapper/lvm2 upstream udev rules
> (which replace our homegrown and outdated rule) which have to distinguish
> between the initial triggering and later execution of uevents.
>
> The device-mapper rules also set some flags that will tell other rules to
> ignore a dm device (these flags have to be respected in devicekit to solve
> some race conditions). The only change in udev is that the I removed a hunk
> from the arch patch that un-ignored dm-* devices: Peristent names for those
> are now handles in the device-mapper rules, respecting the ignore flags and
> thus also avoiding weird races conditions.
>
> If you found this confusing, ask, otherwise please test and sign off.
> (Tested and working on x86_64 with encryption and lvm2)
>
>

FYI: I 've updated lvm2 and device-mapper in testing.

Changes:
- Minor upstream update
- Fix booting from LVM if a mirror or snapshot volume is present