ubifs became broken on contigous power-fails

[Alexander Pazdnikov]

В сообщении от Суббота 12 июня 2010 20:22:55 автор Artem Bityutskiy написал:
> On Mon, 2010-05-24 at 12:56 +0300, Artem Bityutskiy wrote:
> > On Mon, 2010-05-24 at 15:50 +0600, Alexander Pazdnikov wrote:
> > > > Could you enable debugging messages when mounging it after power
> > > > cuts? The messages generate a lot of noise, so you need to disable
> > > > them before starting the real test. The idea is to get UBIFS
> > > > debugging messages when errors occur.
> > >
> > > Sorry, I've hurried up, and after applying your pathes reflash all of
> > > the devices with new firmware.
> > > But I'll setup some devices with all firmware to catch this issue
> > > again.
> > >
> > > > I never had time to really dig this and fix. Would be nice if you
> > > > could do this because I will unlikely have time in the nearest
> > > > future, but I can help you.
> > >
> > > Thank you, I'll continue testings and can supply with debug
> > > information.
> > >
> > > > One think you can do is to make sure you do not fill UBIFS
> > > > completely, then you will avoid hitting these budgeting bugs, also
> > > > known as ENOSPC issues (I think btrfs uses this terminology).
> > >
> > > In the issue above there is 11% used space and 89% free space.
> > > I've got this issue only when mounting fs in RO and then remounting it
> > > in RW.
> >
> > OK. Try my patch. Also, it is worth running your tests with UBIFS extra
> > self checks enabled.
> 
> Did it help?
> 
Artem, thank you very much, yes - it helped, all of 10 testing devices 
survived with repeatable powerfails (1,2,3,4,5,6,7,8,9,10 minutes in this 
sequence) for a week.
Now we leave 3 devices for this test.

Appologize me, please, I've not tested again for the RO-RW recovery on remount 
with this patch to supply you with debug info.

-- 
Best Regards,
Alexander Pazdnikov