Hi Brett, As far as I can see, the replace will finish successfully, but the numbers are quite confusing. Maybe the percentage at the end (after more than half a day or so) gives a hint how the calculation is done, I currently can't see it.
RAID1 and RAID10 have confusing output for commands 'btrfs filesystem usage' and especially for 'btrfs device usage' . At least for the latter, unallocated size was 2x too high I once saw/see, but I have not analysed it in detail. I had once a RAID5 fs and there I replaced devices via a add and then a remove . A replace was considered too risky, also RAID5 size calculation was not supported at that time. With a btrfs fi us I could see that things were going in the right direction, with a lot of patience (a week or so). With RAID10, I did a native replace and mostly looked at iostat and/or ksysguard and it went at expected speed (top media transferrates). The replace status output was correctly from 0-100% in my case, but it was same size replace. I did shrink the source first to max device size minus 2048 sectors and then did replace from a devid to a '/dev/sdX1'. In your case, a 'max' the size of the 8T disk after succesful replace is needed, otherwise 3T wont be used. /Henk On Fri, Oct 16, 2015 at 12:36 PM, Brett King <[email protected]> wrote: > Hi All, > I am noticing some strange numbers when replacing a disk under RAID1: > > [root@array ~]# btrfs replace status -1 /export/archive/ > 367.2% done, 0 write errs, 0 uncorr. read errs > > The filesystem is currently at 6x 6TB + 2x 5TB, replacing the first of each > 5TB with 8TB units: > > [root@array ~]# btrfs fi df /export/archive/ > Data, RAID1: total=20.90TiB, used=20.62TiB > System, RAID1: total=8.00MiB, used=2.91MiB > System, single: total=4.00MiB, used=0.00B > Metadata, RAID1: total=23.00GiB, used=22.13GiB > Metadata, single: total=8.00MiB, used=0.00B > GlobalReserve, single: total=512.00MiB, used=0.00B > [root@array ~]# btrfs fi sh /export/archive/ > Label: 'archive' uuid: 1b1f6bbb-0865-4752-bdaf-07ba30acb27e > Total devices 9 FS bytes used 20.64TiB > devid 0 size 4.55TiB used 4.55TiB path /dev/sdj > devid 1 size 5.46TiB used 5.46TiB path /dev/sdc > devid 2 size 5.46TiB used 5.46TiB path /dev/sdi > devid 3 size 5.46TiB used 5.46TiB path /dev/sdg > devid 4 size 5.46TiB used 5.46TiB path /dev/sdf > devid 5 size 4.55TiB used 4.55TiB path /dev/sdh > devid 6 size 4.55TiB used 4.55TiB path /dev/sdb > devid 7 size 5.46TiB used 5.46TiB path /dev/sde > devid 8 size 5.46TiB used 5.46TiB path /dev/sdd > > btrfs-progs v4.2.1 > [root@array ~]# btrfs device usage /export/archive/ > /dev/sdb, ID: 6 > Device size: 4.55TiB > Data,RAID1: 4.55TiB > Metadata,RAID1: 1.00GiB > Unallocated: 1.02MiB > > /dev/sdc, ID: 1 > Device size: 5.46TiB > Data,RAID1: 5.45TiB > Metadata,single: 8.00MiB > Metadata,RAID1: 10.00GiB > System,single: 4.00MiB > Unallocated: 5.02MiB > > /dev/sdd, ID: 8 > Device size: 5.46TiB > Data,RAID1: 5.46TiB > Metadata,RAID1: 1.00GiB > Unallocated: 1.02MiB > > /dev/sde, ID: 7 > Device size: 5.46TiB > Data,RAID1: 5.46TiB > Unallocated: 1.02MiB > > /dev/sdf, ID: 4 > Device size: 5.46TiB > Data,RAID1: 5.45TiB > Metadata,RAID1: 11.00GiB > System,RAID1: 8.00MiB > Unallocated: 1.02MiB > > /dev/sdg, ID: 3 > Device size: 5.46TiB > Data,RAID1: 5.45TiB > Metadata,RAID1: 11.00GiB > System,RAID1: 8.00MiB > Unallocated: 1.02MiB > > /dev/sdh, ID: 5 > Device size: 4.55TiB > Data,RAID1: 4.55TiB > Metadata,RAID1: 1.00GiB > Unallocated: 1.02MiB > > /dev/sdi, ID: 2 > Device size: 5.46TiB > Data,RAID1: 5.45TiB > Metadata,RAID1: 11.00GiB > Unallocated: 1.02MiB > > /dev/sdj, ID: 0 > Device size: 7.28TiB > Unallocated: 4.55TiB > > [root@array ~]# uname -r > 4.1.10-200.fc22.x86_64 > > Is this a problem ? And is there another way to see moved / total blocks from > the running replace job ? > > Thanks, > Brett. > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to [email protected] > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
