Re: mdadm serious does wrong partition changes
On Sun, 2017-12-03 at 14:09 +0100, Frans van Berckel wrote:
> On Sun, 2017-12-03 at 13:32 +0100, John Paul Adrian Glaubitz wrote:
> >
> > I don't have any of these issues but I also don't understand your
> > strange RAID setup. We have setup mdraid on three of the sparc64
> > machines we have and I never saw any issues.
>
> I have created this md0 while setup as well. That goes on and boots.
>
> But what's the better way of adding more raid partitions to a running
> server? Is there a Debian specific, and anywhere documented?
>
> > I also think it's extremely unlikely that the mdraid code has such
> > fundamental issues. This is widely used code which doesn't see too
> > many changes often.
>
> I am aware of, so lets us see, your methode helps preventing this.
There are deferences for the version mdraid superblock position. When
are version 1.1 and 1.2 released?
Sub-Version Superblock position on device
0.9 At the end of the device
1.0 At the end of the device
1.1 At the beginning of the device
1.2 4K from the beginning of the device
https://raid.wiki.kernel.org/index.php/RAID_superblock_formats
So let me find out the problem still occurs with one of the other
versions.
My idea, maybe sdc1 and sdd1 are handling deferent, then sdc2 and sdd2
in relation to Sun disk label. Because that's what I am using and is
the one that corrupted my Sun table on sdc and sdd. And my md0 is
created on sda2 and sdb2, without any problem.
If you are running mdraid, are you willing to post your # lsblk
Thanks,
Frans van Berckel
Reply to: