[PATCH 1/1] Don't handle change event against raw devices

[PATCH 1/1] Don't handle change event against raw devices

[Xiao Ni]

Signed-off-by: Xiao Ni <xni@redhat.com>
---
 udev-md-raid-assembly.rules | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/udev-md-raid-assembly.rules b/udev-md-raid-assembly.rules
index 39b4344b8592..748ea05dadaa 100644
--- a/udev-md-raid-assembly.rules
+++ b/udev-md-raid-assembly.rules
@@ -11,6 +11,11 @@ SUBSYSTEM!="block", GOTO="md_inc_end"
 ENV{SYSTEMD_READY}=="0", GOTO="md_inc_end"
 
 # handle potential components of arrays (the ones supported by md)
+# For member devices which are md/dm devices, we don't need to
+# handle add event. Because md/dm devices need to do some init jobs.
+# Then the change event happens.
+# When adding md/dm devices, ID_FS_TYPE only be linux_raid_member
+# after change event happens.
 ENV{ID_FS_TYPE}=="linux_raid_member", GOTO="md_inc"
 
 # "noiswmd" on kernel command line stops mdadm from handling
@@ -28,6 +33,11 @@ GOTO="md_inc_end"
 
 LABEL="md_inc"
 
+# We only handle add event on raw disks. If we handle change event on raw disk,
+# the tool parted can't change partition table unless clear superblock on
+# member disks
+ACTION=="change", KERNEL!="dm-*|md*", GOTO="md_inc_end"
+
 # remember you can limit what gets auto/incrementally assembled by
 # mdadm.conf(5)'s 'AUTO' and selectively whitelist using 'ARRAY'
 ACTION!="remove", IMPORT{program}="BINDIR/mdadm --incremental --export $devnode --offroot $env{DEVLINKS}"
-- 
2.32.0 (Apple Git-132)

Re: [PATCH 1/1] Don't handle change event against raw devices

[Paul Menzel]

Dear Xiao,


Thank you for the patch.

Am 30.12.22 um 10:07 schrieb Xiao Ni:

It’d be great if you described the problem.

> Signed-off-by: Xiao Ni <xni@redhat.com>
> ---
>   udev-md-raid-assembly.rules | 10 ++++++++++
>   1 file changed, 10 insertions(+)
> 
> diff --git a/udev-md-raid-assembly.rules b/udev-md-raid-assembly.rules
> index 39b4344b8592..748ea05dadaa 100644
> --- a/udev-md-raid-assembly.rules
> +++ b/udev-md-raid-assembly.rules
> @@ -11,6 +11,11 @@ SUBSYSTEM!="block", GOTO="md_inc_end"
>   ENV{SYSTEMD_READY}=="0", GOTO="md_inc_end"
>   
>   # handle potential components of arrays (the ones supported by md)
> +# For member devices which are md/dm devices, we don't need to
> +# handle add event. Because md/dm devices need to do some init jobs.
> +# Then the change event happens.
> +# When adding md/dm devices, ID_FS_TYPE only be linux_raid_member

A verb is missing. Maybe: … can only be …

> +# after change event happens.
>   ENV{ID_FS_TYPE}=="linux_raid_member", GOTO="md_inc"
>   
>   # "noiswmd" on kernel command line stops mdadm from handling
> @@ -28,6 +33,11 @@ GOTO="md_inc_end"
>   
>   LABEL="md_inc"
>   
> +# We only handle add event on raw disks. If we handle change event on raw disk,
> +# the tool parted can't change partition table unless clear superblock on

1.  *the* partition table
2.  Please excuse my ignorance, but what is a “clear superblock”?

> +# member disks

Add a dot/period at the end of sentences?

> +ACTION=="change", KERNEL!="dm-*|md*", GOTO="md_inc_end"
> +
>   # remember you can limit what gets auto/incrementally assembled by
>   # mdadm.conf(5)'s 'AUTO' and selectively whitelist using 'ARRAY'
>   ACTION!="remove", IMPORT{program}="BINDIR/mdadm --incremental --export $devnode --offroot $env{DEVLINKS}"

Re: [PATCH 1/1] Don't handle change event against raw devices

[Xiao Ni]

On Fri, Dec 30, 2022 at 6:20 PM Paul Menzel <pmenzel@molgen.mpg.de> wrote:
>
> Dear Xiao,
>
>
> Thank you for the patch.
>
> Am 30.12.22 um 10:07 schrieb Xiao Ni:
>
> It’d be great if you described the problem.

Hi Paul

Thanks. I'll describe more in next version.
>
> > Signed-off-by: Xiao Ni <xni@redhat.com>
> > ---
> >   udev-md-raid-assembly.rules | 10 ++++++++++
> >   1 file changed, 10 insertions(+)
> >
> > diff --git a/udev-md-raid-assembly.rules b/udev-md-raid-assembly.rules
> > index 39b4344b8592..748ea05dadaa 100644
> > --- a/udev-md-raid-assembly.rules
> > +++ b/udev-md-raid-assembly.rules
> > @@ -11,6 +11,11 @@ SUBSYSTEM!="block", GOTO="md_inc_end"
> >   ENV{SYSTEMD_READY}=="0", GOTO="md_inc_end"
> >
> >   # handle potential components of arrays (the ones supported by md)
> > +# For member devices which are md/dm devices, we don't need to
> > +# handle add event. Because md/dm devices need to do some init jobs.
> > +# Then the change event happens.
> > +# When adding md/dm devices, ID_FS_TYPE only be linux_raid_member
>
> A verb is missing. Maybe: … can only be …
>
> > +# after change event happens.
> >   ENV{ID_FS_TYPE}=="linux_raid_member", GOTO="md_inc"
> >
> >   # "noiswmd" on kernel command line stops mdadm from handling
> > @@ -28,6 +33,11 @@ GOTO="md_inc_end"
> >
> >   LABEL="md_inc"
> >
> > +# We only handle add event on raw disks. If we handle change event on raw disk,
> > +# the tool parted can't change partition table unless clear superblock on
>
> 1.  *the* partition table
> 2.  Please excuse my ignorance, but what is a “clear superblock”?

It means clearing superblock with command mdadm --zero-superblock

>
> > +# member disks
>
> Add a dot/period at the end of sentences?
>
> > +ACTION=="change", KERNEL!="dm-*|md*", GOTO="md_inc_end"
> > +
> >   # remember you can limit what gets auto/incrementally assembled by
> >   # mdadm.conf(5)'s 'AUTO' and selectively whitelist using 'ARRAY'
> >   ACTION!="remove", IMPORT{program}="BINDIR/mdadm --incremental --export $devnode --offroot $env{DEVLINKS}"
>


Best Regards
Xiao

Re: [PATCH 1/1] Don't handle change event against raw devices

[Jes Sorensen]

On 12/30/22 05:37, Xiao Ni wrote:
> On Fri, Dec 30, 2022 at 6:20 PM Paul Menzel <pmenzel@molgen.mpg.de> wrote:
>>
>> Dear Xiao,
>>
>>
>> Thank you for the patch.
>>
>> Am 30.12.22 um 10:07 schrieb Xiao Ni:
>>
>> It’d be great if you described the problem.
> 
> Hi Paul
> 
> Thanks. I'll describe more in next version.
>>

Hi Xiao,

Did you post an updated version?

Thanks,
Jes

Re: [PATCH 1/1] Don't handle change event against raw devices

[Xiao Ni]

在 2023/1/4 下午11:48, Jes Sorensen 写道:
> On 12/30/22 05:37, Xiao Ni wrote:
>> On Fri, Dec 30, 2022 at 6:20 PM Paul Menzel <pmenzel@molgen.mpg.de> wrote:
>>> Dear Xiao,
>>>
>>>
>>> Thank you for the patch.
>>>
>>> Am 30.12.22 um 10:07 schrieb Xiao Ni:
>>>
>>> It’d be great if you described the problem.
>> Hi Paul
>>
>> Thanks. I'll describe more in next version.
> Hi Xiao,
>
> Did you post an updated version?
>
> Thanks,
> Jes
>
>
>
Hi Jes

I'll send the new version soon.

Regards

Xiao