Re: RAID0 size over 2 TB

["azurIt" <azurit@pobox.sk>]

______________________________________________________________
> Od: Chris Murphy <lists@colorremedies.com>
> Komu: azurIt <azurit@pobox.sk>
> Dátum: 08.09.2014 20:27
> Predmet: Re: RAID0 size over 2 TB
>
> CC: linux-raid@vger.kernel.org
>On Sep 8, 2014, at 1:33 AM, azurIt <azurit@pobox.sk> wrote:
>> 
>> This is really strange, here is what cgdisk is reporting:
>> http://imgur.com/U8VnAdM
>
>Does your kernel have CONFIG_EFI_PARTITION=y ?
>
>Old versions of fdisk report the protective MBR, it does not show you GPT partition map. It's normal for the protective MBR to max out out 2TB because MBR encodes 32-bit 512 byte sectors on 512e devices. Your kernel appears to think the disk is 2TB, which suggests to me it's looking only at the MBR, meaning it doesn't have GPT support turned on.
>
>
>Chris Murphy--
>To unsubscribe from this list: send the line "unsubscribe linux-raid" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html



Ok so, problem is resolved. This is what happened:
 - i replaced 2x 2 TB drives with 2x 4 TB drives WITHOUT rebooting
 - i created 4 TB partition on every drive
 - kernel still thinks that partitions are 2 TB, probably because of some kind of partition table cache
 - when i added these partitions to RAID array, kernel didn't know they are 4 TB and, even, lightly corrupted them (cgdisk was reporting some kind of issues after i added partitions into RAID array, but everything was working ok)

The solution was to call command 'partprobe' (which is part of 'parted' package) after partitioning but BEFORE adding partitions into RAID array.

Thanks to everyone involved!

azur
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html