Hi,
Just installed F9 preview and updated to latest. Rebooted everything was fine. Today, I booted the system again, played with it for a few hours, then I rebooted to test the "i915.modeset" (which is hyper cool thanks Now when the machine came up, I got:

Primary superblock features different from backup: Check forced
This was on the root filesystem. It checked the FS, and finally said "** Rebooting Linux **", at this stage I was fairly scared .. it looked like a serious problem!

However, Fedora rebooted and it came up just fine!

Well, I have no idea what just happened

PS: Why did F9 disable Synaptics touchpad tapping!! This is going to make a lot of Laptop owners angry! At least it made me. Considering Windows and older Fedoras had it enabled, everyone is going to be not too happy about the change I guess. Anyway, if there's infinite wisdom about this, could we at least get release note tips on how to enable it back ?


Thanks for the awesome work .. wifi works out of the box, Alsa too, correct wide screen, kernel mode setting ... etc Fedora is getting too good

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Ahmed Kamal wrote:
> Hi,
> Just installed F9 preview and updated to latest. Rebooted everything was
> fine. Today, I booted the system again, played with it for a few hours,
> then I rebooted to test the "i915.modeset" (which is hyper cool thanks
> Now when the machine came up, I got:
> Primary superblock features different from backup: Check forced
> This was on the root filesystem. It checked the FS, and finally said "**
> Rebooting Linux **", at this stage I was fairly scared .. it looked like
> a serious problem!
> However, Fedora rebooted and it came up just fine!
>
> Well, I have no idea what just happened

It should be fine.

Most recent e2fsprogs enforces that feature flags in backups match those
on the primary.

Problem is, the kernel sets things on the fly, but does not update the
backups. This is ... annoying... because then it causes the full fsck.

I'm a little surprised that it did check though, because for now I have
F9 carrying a patch to actually *disable* this test, because otherwise
any F8->F9 upgrade would cause a full fsck just for this reason, which I
think is overly aggressive.

But, anyway, you should be fine now, and it shouldn't happen again.

-Eric

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Just out of curiosity, could you let me know what kind of changes the kernel makes to the superblock without asking first! That sounds interesting

On Sun, May 11, 2008 at 12:59 AM, Eric Sandeen <sandeen@redhat.com> wrote:

Ahmed Kamal wrote:

> Hi,

> Just installed F9 preview and updated to latest. Rebooted everything was

> fine. Today, I booted the system again, played with it for a few hours,

> then I rebooted to test the "i915.modeset" (which is hyper cool thanks

> Now when the machine came up, I got:

> Primary superblock features different from backup: Check forced

> This was on the root filesystem. It checked the FS, and finally said "**

> Rebooting Linux **", at this stage I was fairly scared .. it looked like

> a serious problem!

> However, Fedora rebooted and it came up just fine!

>

> Well, I have no idea what just happened



It should be fine.



Most recent e2fsprogs enforces that feature flags in backups match those

on the primary.



Problem is, the kernel sets things on the fly, but does not update the

backups. *This is ... annoying... because then it causes the full fsck.



I'm a little surprised that it did check though, because for now I have

F9 carrying a patch to actually *disable* this test, because otherwise

any F8->F9 upgrade would cause a full fsck just for this reason, which I

think is overly aggressive.



But, anyway, you should be fine now, and it shouldn't happen again. *



-Eric



--

fedora-devel-list mailing list

fedora-devel-list@redhat.com

https://www.redhat.com/mailman/listinfo/fedora-devel-list



--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Ahmed Kamal wrote:
> Just out of curiosity, could you let me know what kind of changes the
> kernel makes to the superblock without asking first! That sounds interesting

sure, things like the first time a "large" (> 2G file) is written it
will set a flag to that effect.

Or the first time an extent-based file is created on ext4.

We've talked about removing these, actually, and requiring the flags to
be set at mkfs time, or via tune2fs, if you wish to use the feature,
rather than silently turning them on.

-Eric

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

bingo, I had created a virtualbox VM, which created a 8G file, that must have triggered the superblock changes. I feel much calmer now. Thank you

On Sun, May 11, 2008 at 1:15 AM, Eric Sandeen <sandeen@redhat.com> wrote:

Ahmed Kamal wrote:

> Just out of curiosity, could you let me know what kind of changes the

> kernel makes to the superblock without asking first! That sounds interesting



sure, things like the first time a "large" (> 2G file) is written it

will set a flag to that effect.



Or the first time an extent-based file is created on ext4.



We've talked about removing these, actually, and requiring the flags to

be set at mkfs time, or via tune2fs, if you wish to use the feature,

rather than silently turning them on.



-Eric



--

fedora-devel-list mailing list

fedora-devel-list@redhat.com

https://www.redhat.com/mailman/listinfo/fedora-devel-list



--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Sat, 2008-05-10 at 17:15 -0500, Eric Sandeen wrote:
> Ahmed Kamal wrote:
> > Just out of curiosity, could you let me know what kind of changes the
> > kernel makes to the superblock without asking first! That sounds interesting
>
> sure, things like the first time a "large" (> 2G file) is written it
> will set a flag to that effect.
>
> Or the first time an extent-based file is created on ext4.
>
> We've talked about removing these, actually, and requiring the flags to
> be set at mkfs time, or via tune2fs, if you wish to use the feature,
> rather than silently turning them on.

But make that the default for new filesystems pretty please kthxbye ;-).

Nils
--
Nils Philippsen / Red Hat / nphilipp@redhat.com
"Those who would give up Essential Liberty to purchase a little Temporary
Safety, deserve neither Liberty nor Safety." -- B. Franklin, 1759
PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Nils Philippsen wrote:
> On Sat, 2008-05-10 at 17:15 -0500, Eric Sandeen wrote:
>> Ahmed Kamal wrote:
>>> Just out of curiosity, could you let me know what kind of changes the
>>> kernel makes to the superblock without asking first! That sounds interesting
>> sure, things like the first time a "large" (> 2G file) is written it
>> will set a flag to that effect.
>>
>> Or the first time an extent-based file is created on ext4.
>>
>> We've talked about removing these, actually, and requiring the flags to
>> be set at mkfs time, or via tune2fs, if you wish to use the feature,
>> rather than silently turning them on.
>
> But make that the default for new filesystems pretty please kthxbye ;-).

Make which the default? But of course sane defaults would be chosen
for mkfs...

-Eric

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Hi.

On Sat, 10 May 2008 16:59:38 -0500, Eric Sandeen wrote

> Problem is, the kernel sets things on the fly, but does not update the
> backups. This is ... annoying... because then it causes the full
> fsck.

And why doesn't the kernel propagate those flags (on umount, for example)?

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Ralf Ertzinger wrote:
> Hi.
>
> On Sat, 10 May 2008 16:59:38 -0500, Eric Sandeen wrote
>
>> Problem is, the kernel sets things on the fly, but does not update the
>> backups. This is ... annoying... because then it causes the full
>> fsck.
>
> And why doesn't the kernel propagate those flags (on umount, for example)?

Well, if the kernel is going to set things on the fly (which IMHO it
shouldn't) then it should do it properly (from the perspective of fsck)
on all backup superblocks. As to why it doesn't? No good answer for you.

-Eric

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list