SD card corrupts and looses partitions every time laptop sleeps with 12.2. This happens 100% of the time. The data is recoverable using “sgdisk -l” (I backed up the partion table). However I need to fix this to continue using SUSE.
I have an unusual requirement and install. I need to run on the SDCard and not a hard drive or SSD. Please don’t suggest changing hardware. To this end root, home, and swap partitions are on the SDCard. To allow booting the /boot partition is on a micro usb key, which remains in the side of the laptop. The system boots up and runs fine after adding the mmc_core module to the initfs. However if it goes to sleep for any reason (lid close, commanded, or inactivity timer) it will not reboot because the partion table on the SDCard is lost.
I have experimented with setting the cards not removable with the kernel parameter mmc_core.removable=0 however this does not seem to work reliably. I verified that /sys/module/mmc_core/paratemers/removable shows “N” and it still corrupts.
What kernel modules are reauired to properly read the SDCARD?
How can I reliably prevent the system from unmounting the SD Card on sleep?
Thanks for you help this issue is killing me. I’ve spent over 100 hours trying to get suse to run on my system and I still dont have it working reliably.
On 2012-11-14 15:26, jm001 wrote:
>
> SD card corrupts and looses partitions every time laptop sleeps with
> 12.2. This happens 100% of the time. The data is recoverable using
> “sgdisk -l” (I backed up the partion table). However I need to fix this
> to continue using SUSE.
→ bugzilla.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
Sorry I don’t understand your post. Are you reccomending I search buglizza? Are you recommending I write a bug?
I think the intended operation is to unmount SD on sleep. There may be a bug with the unsafe remove but I’m not sure I’m using it right.
On 11/15/2012 01:16 AM, jm001 wrote:
> Are you reccomending I search
> buglizza? Are you recommending I write a bug?
i can’t speak for Carlos, but if you well search bugzilla and fail to
find a relevant bug then he must have been suggesting you log your bug.
(that is, he may remember that this situation already has a bug
against it, and there is a usable workaround noted in the bug!)
On 2012-11-15 08:54, dd wrote:
> On 11/15/2012 01:16 AM, jm001 wrote:
>> Are you reccomending I search
>> buglizza? Are you recommending I write a bug?
>
> i can’t speak for Carlos, but if you well search bugzilla and fail to
> find a relevant bug then he must have been suggesting you log your bug.
> (that is, he may remember that this situation already has a bug
> against it, and there is a usable workaround noted in the bug!)
>
> **
Yes, I was busy.
Yes I meant to just use Bugzilla. Search for a similar bug, and if not
found, report the issue. The issue looks to me to be a bug somewhere.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
Could it be that the fact that system and /boot reside on one device, swap on the other?
And, where does GRUB reside?
I had to create a setup like this one for somebody, which works fine ( i.e. haven’t heard complaints ), on 2 devices, a 16GB and a 32GB. At that time I did not even think about it and created swap and “/” on the 16GB, mounted the single parition on the 32GB on /home. And made sure GRUB was installed on the 16GB. Worked like a charm, done in under 2 hours incl. updates and Packman addition.