Skip to contentSkip to conversation

No space left after add device and balance

DISCUSSION Miloslav Hůla 3 participants 5 messages Sep 9, 2020 – Sep 15, 2020 Btrfs file system

lore
Share & tools
lore
#0 lore raw

Hello,

we are using btrfs RAID-10 (/data, 4.7TB) on a physical Supermicro server with Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz and 125GB of RAM.

We added two more drives:

btrfs device add /dev/sds /data
btrfs device add /dev/sdt /data

and than we run balance:

btrfs filesystem balance /data

Everything run fine about 15 hours. Then we got error "No space left" (dmesg output below) and /data remounted into read-only.

So we unmounted /data and mounted again, everything looks OK in read-write. Than we run full balance again:

btrfs filesystem balance /data

After ~15.5 hours finished successfully. Unfortunetally, I have no exact free space report before first balance, but it looked roughly like:

Label: 'DATA' uuid: 5b285a46-e55d-4191-924f-0884fa06edd8

        Total devices 16 FS bytes used 3.49TiB
        devid    1 size 558.41GiB used 448.66GiB path /dev/sda
        devid    2 size 558.41GiB used 448.66GiB path /dev/sdb
        devid    4 size 558.41GiB used 448.66GiB path /dev/sdd
        devid    5 size 558.41GiB used 448.66GiB path /dev/sde
        devid    7 size 558.41GiB used 448.66GiB path /dev/sdg
        devid    8 size 558.41GiB used 448.66GiB path /dev/sdh
        devid    9 size 558.41GiB used 448.66GiB path /dev/sdf
        devid   10 size 558.41GiB used 448.66GiB path /dev/sdi
        devid   11 size 558.41GiB used 448.66GiB path /dev/sdj
        devid   13 size 558.41GiB used 448.66GiB path /dev/sdk
        devid   14 size 558.41GiB used 448.66GiB path /dev/sdc
        devid   15 size 558.41GiB used 448.66GiB path /dev/sdl
        devid   16 size 558.41GiB used 448.66GiB path /dev/sdm
        devid   17 size 558.41GiB used 448.66GiB path /dev/sdn
        devid   18 size 837.84GiB used 448.66GiB path /dev/sdr
        devid   19 size 837.84GiB used 448.66GiB path /dev/sdq
        devid   20 size 837.84GiB used   0.00GiB path /dev/sds
        devid   21 size 837.84GiB used   0.00GiB path /dev/sdt

Are we doing something wrong? I found posts, where problems with balace of full filesystem are described. And as a recommendation is "add empty device, run balance, remove device".

Are there some requirements on free space for balance even if you add new device?

Maybe "fixed" in newer kernel?

Thank you. With kind regards
Milo

# uname -a
Linux imap 4.9.0-12-amd64 #1 SMP Debian 4.9.210-1 (2020-01-20) x86_64 GNU/Linux
# btrfs --version
btrfs-progs v4.7.3
# btrfs fi show (after successfull balance)
Label: 'DATA'  uuid: 5b285a46-e55d-4191-924f-0884fa06edd8
        Total devices 18 FS bytes used 3.49TiB
        devid    1 size 558.41GiB used 398.27GiB path /dev/sda
        devid    2 size 558.41GiB used 398.27GiB path /dev/sdb
        devid    4 size 558.41GiB used 398.27GiB path /dev/sdd
        devid    5 size 558.41GiB used 398.27GiB path /dev/sde
        devid    7 size 558.41GiB used 398.27GiB path /dev/sdg
        devid    8 size 558.41GiB used 398.27GiB path /dev/sdh
        devid    9 size 558.41GiB used 398.27GiB path /dev/sdf
        devid   10 size 558.41GiB used 398.27GiB path /dev/sdi
        devid   11 size 558.41GiB used 398.27GiB path /dev/sdj
        devid   13 size 558.41GiB used 398.27GiB path /dev/sdk
        devid   14 size 558.41GiB used 398.27GiB path /dev/sdc
        devid   15 size 558.41GiB used 398.27GiB path /dev/sdl
        devid   16 size 558.41GiB used 398.27GiB path /dev/sdm
        devid   17 size 558.41GiB used 398.27GiB path /dev/sdn
        devid   18 size 837.84GiB used 398.27GiB path /dev/sdr
        devid   19 size 837.84GiB used 398.27GiB path /dev/sdq
        devid   20 size 837.84GiB used 398.27GiB path /dev/sds
        devid   21 size 837.84GiB used 398.27GiB path /dev/sdt
# btrfs fi df /data/ (after successfull balance)
ata, RAID10: total=3.48TiB, used=3.48TiB
System, RAID10: total=144.00MiB, used=304.00KiB
Metadata, RAID10: total=20.25GiB, used=17.96GiB
GlobalReserve, single: total=512.00MiB, used=0.00B
# dmesg
[Tue Sep  8 06:45:36 2020] BTRFS info (device sda): relocating block group 33991231012864 flags 68
[Tue Sep  8 06:47:56 2020] BTRFS info (device sda): found 63581 extents
[Tue Sep  8 06:47:58 2020] BTRFS info (device sda): relocating block group 33990157271040 flags 68
[Tue Sep  8 06:48:34 2020] ------------[ cut here ]------------
[Tue Sep  8 06:48:34 2020] WARNING: CPU: 21 PID: 1342 at /build/linux-xZ5nrU/linux-4.9.210/fs/btrfs/extent-tree.c:2967 btrfs_run_delayed_refs+0x27e/0x2b0 [btrfs]
[Tue Sep  8 06:48:34 2020] BTRFS: Transaction aborted (error -28)
[Tue Sep  8 06:48:34 2020] Modules linked in: fuse ufs qnx4 hfsplus hfs minix ntfs vfat msdos fat jfs xfs dm_mod ipt_REJECT nf_reject_ipv4 nfsv3 nfs_acl nfs lockd grace fscache xt_multiport iptable_filter intel_rapl sb_edac edac_core x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel intel_cstate iTCO_wdt iTCO_vendor_support mxm_wmi intel_uncore ast intel_rapl_perf pcspkr ttm drm_kms_helper drm mei_me lpc_ich joydev sg mfd_core mei evdev ioatdma shpchp ipmi_si ipmi_msghandler wmi acpi_power_meter acpi_pad button sunrpc ip_tables x_tables autofs4 ext4 crc16 jbd2 fscrypto ecb mbcache btrfs raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c crc32c_generic raid0 multipath linear raid1 md_mod ses enclosure
[Tue Sep  8 06:48:34 2020]  scsi_transport_sas sd_mod hid_generic usbhid hid crc32c_intel aesni_intel aes_x86_64 glue_helper lrw gf128mul ablk_helper cryptd ahci xhci_pci libahci ehci_pci xhci_hcd ehci_hcd igb i2c_i801 libata megaraid_sas i2c_smbus i2c_algo_bit usbcore dca ptp usb_common pps_core scsi_mod
[Tue Sep  8 06:48:34 2020] CPU: 21 PID: 1342 Comm: btrfs-transacti Not tainted 4.9.0-12-amd64 #1 Debian 4.9.210-1
[Tue Sep  8 06:48:34 2020] Hardware name: Supermicro X10DRi/X10DRi, BIOS 2.1 09/13/2016
[Tue Sep  8 06:48:34 2020]  0000000000000000 ffffffff97936bfe ffffb0d54d8a7d68 0000000000000000
[Tue Sep  8 06:48:34 2020]  ffffffff9767b98b ffffffffc04cbc28 ffffb0d54d8a7dc0 ffff8f09538faa48
[Tue Sep  8 06:48:34 2020]  ffff8f1c38030000 ffff8f1c36f830c0 000000000023d11a ffffffff9767ba0f
[Tue Sep  8 06:48:34 2020] Call Trace:
[Tue Sep  8 06:48:34 2020]  [<ffffffff97936bfe>] ? dump_stack+0x66/0x88
[Tue Sep  8 06:48:34 2020]  [<ffffffff9767b98b>] ? __warn+0xcb/0xf0
[Tue Sep  8 06:48:34 2020]  [<ffffffff9767ba0f>] ? warn_slowpath_fmt+0x5f/0x80
[Tue Sep  8 06:48:34 2020]  [<ffffffffc04295fe>] ? btrfs_run_delayed_refs+0x27e/0x2b0 [btrfs]
[Tue Sep  8 06:48:34 2020]  [<ffffffffc043f823>] ? btrfs_commit_transaction+0x53/0xa10 [btrfs]
[Tue Sep  8 06:48:34 2020]  [<ffffffffc0440276>] ? start_transaction+0x96/0x480 [btrfs]
[Tue Sep  8 06:48:34 2020]  [<ffffffffc043a67c>] ? transaction_kthread+0x1dc/0x200 [btrfs]
[Tue Sep  8 06:48:34 2020]  [<ffffffffc043a4a0>] ? btrfs_cleanup_transaction+0x590/0x590 [btrfs]
[Tue Sep  8 06:48:34 2020]  [<ffffffff9769bdc9>] ? kthread+0xd9/0xf0
[Tue Sep  8 06:48:34 2020]  [<ffffffff97c1e2f1>] ? __switch_to_asm+0x41/0x70
[Tue Sep  8 06:48:34 2020]  [<ffffffff9769bcf0>] ? kthread_park+0x60/0x60
[Tue Sep  8 06:48:34 2020]  [<ffffffff97c1e377>] ? ret_from_fork+0x57/0x70
[Tue Sep  8 06:48:34 2020] ---[ end trace 7be8e0e64ac370b3 ]---
[Tue Sep  8 06:48:34 2020] BTRFS: error (device sda) in btrfs_run_delayed_refs:2967: errno=-28 No space left
[Tue Sep  8 06:48:34 2020] BTRFS info (device sda): forced readonly
[Tue Sep  8 07:31:16 2020] BTRFS info (device sda): disk space caching is enabled
[Tue Sep  8 07:31:16 2020] BTRFS error (device sda): Remounting read-write after error is not allowed
[Tue Sep  8 07:32:50 2020] BTRFS error (device sda): cleaner transaction attach returned -30
[Tue Sep  8 07:33:01 2020] BTRFS info (device sdt): disk space caching is enabled
[Tue Sep  8 07:33:01 2020] BTRFS info (device sdt): has skinny extents
[Tue Sep  8 07:35:38 2020] BTRFS info (device sdt): checking UUID tree
[Tue Sep  8 07:35:38 2020] BTRFS info (device sdt): continuing balance
[Tue Sep  8 07:35:38 2020] BTRFS info (device sdt): relocating block group 40004587880448 flags 68
[Tue Sep  8 07:38:49 2020] BTRFS info (device sdt): found 32320 extents
[Tue Sep  8 07:38:50 2020] BTRFS info (device sdt): relocating block group 39945397862400 flags 68
[Tue Sep  8 07:42:07 2020] BTRFS info (device sdt): found 53227 extents
[Tue Sep  8 07:42:08 2020] BTRFS info (device sdt): relocating block group 39944189902848 flags 68
[Tue Sep  8 07:44:29 2020] BTRFS info (device sdt): found 57961 extents
[Tue Sep  8 07:44:30 2020] BTRFS info (device sdt): relocating block group 33985728086016 flags 66
[Tue Sep  8 07:44:30 2020] BTRFS info (device sdt): found 20 extents
#1 lore raw
Quoted 25 lines · Miloslav Hůla

After ~15.5 hours finished successfully. Unfortunetally, I have no exact free space report before first balance, but it looked roughly like:

Label: 'DATA' uuid: 5b285a46-e55d-4191-924f-0884fa06edd8

         Total devices 16 FS bytes used 3.49TiB
         devid    1 size 558.41GiB used 448.66GiB path /dev/sda
         devid    2 size 558.41GiB used 448.66GiB path /dev/sdb
         devid    4 size 558.41GiB used 448.66GiB path /dev/sdd
         devid    5 size 558.41GiB used 448.66GiB path /dev/sde
         devid    7 size 558.41GiB used 448.66GiB path /dev/sdg
         devid    8 size 558.41GiB used 448.66GiB path /dev/sdh
         devid    9 size 558.41GiB used 448.66GiB path /dev/sdf
         devid   10 size 558.41GiB used 448.66GiB path /dev/sdi
         devid   11 size 558.41GiB used 448.66GiB path /dev/sdj
         devid   13 size 558.41GiB used 448.66GiB path /dev/sdk
         devid   14 size 558.41GiB used 448.66GiB path /dev/sdc
         devid   15 size 558.41GiB used 448.66GiB path /dev/sdl
         devid   16 size 558.41GiB used 448.66GiB path /dev/sdm
         devid   17 size 558.41GiB used 448.66GiB path /dev/sdn
         devid   18 size 837.84GiB used 448.66GiB path /dev/sdr
         devid   19 size 837.84GiB used 448.66GiB path /dev/sdq
         devid   20 size 837.84GiB used   0.00GiB path /dev/sds
         devid   21 size 837.84GiB used   0.00GiB path /dev/sdt
Quoted 3 lines

Are we doing something wrong? I found posts, where problems with balace of full filesystem are described. And as a recommendation is "add empty device, run balance, remove device".

It's raid10, so in this case, you probably need to add 4 devices. It's not required they be equal sizes but it's ideal.

Quoted 2 lines

Are there some requirements on free space for balance even if you add new device?

The free space is reported upon adding the two devices; but the raid10 profile requires more than just free space, it needs free space on four devices.

Quoted 5 lines
[Tue Sep  8 06:48:34 2020] BTRFS info (device sda): forced readonly
[Tue Sep  8 07:31:16 2020] BTRFS info (device sda): disk space caching
is enabled
[Tue Sep  8 07:31:16 2020] BTRFS error (device sda): Remounting

read-write after error is not allowed

Yeah it's confused. You need to reboot.

Signature · 1 line
Chris Murphy
#2 lore raw

Dne 10.09.2020 v 21:18 Chris Murphy napsal(a):

Quoted 34 lines · Miloslav Hůla
Quoted 26 lines

After ~15.5 hours finished successfully. Unfortunetally, I have no exact free space report before first balance, but it looked roughly like:

Label: 'DATA' uuid: 5b285a46-e55d-4191-924f-0884fa06edd8

          Total devices 16 FS bytes used 3.49TiB
          devid    1 size 558.41GiB used 448.66GiB path /dev/sda
          devid    2 size 558.41GiB used 448.66GiB path /dev/sdb
          devid    4 size 558.41GiB used 448.66GiB path /dev/sdd
          devid    5 size 558.41GiB used 448.66GiB path /dev/sde
          devid    7 size 558.41GiB used 448.66GiB path /dev/sdg
          devid    8 size 558.41GiB used 448.66GiB path /dev/sdh
          devid    9 size 558.41GiB used 448.66GiB path /dev/sdf
          devid   10 size 558.41GiB used 448.66GiB path /dev/sdi
          devid   11 size 558.41GiB used 448.66GiB path /dev/sdj
          devid   13 size 558.41GiB used 448.66GiB path /dev/sdk
          devid   14 size 558.41GiB used 448.66GiB path /dev/sdc
          devid   15 size 558.41GiB used 448.66GiB path /dev/sdl
          devid   16 size 558.41GiB used 448.66GiB path /dev/sdm
          devid   17 size 558.41GiB used 448.66GiB path /dev/sdn
          devid   18 size 837.84GiB used 448.66GiB path /dev/sdr
          devid   19 size 837.84GiB used 448.66GiB path /dev/sdq
          devid   20 size 837.84GiB used   0.00GiB path /dev/sds
          devid   21 size 837.84GiB used   0.00GiB path /dev/sdt
Quoted 3 lines

Are we doing something wrong? I found posts, where problems with balace of full filesystem are described. And as a recommendation is "add empty device, run balance, remove device".

It's raid10, so in this case, you probably need to add 4 devices. It's not required they be equal sizes but it's ideal.

Quoted 6 lines
Quoted 2 lines

Are there some requirements on free space for balance even if you add new device?

The free space is reported upon adding the two devices; but the raid10 profile requires more than just free space, it needs free space on four devices.

I didn't realize that. It makes me sense now. So we are probably "wrong" with 18 devices. Multiple of 4 would be better I guess.

Thank you!

#3 lore raw
Quoted 34 lines · Miloslav Hůla

Dne 10.09.2020 v 21:18 Chris Murphy napsal(a):

Quoted 33 lines · Miloslav Hůla
Quoted 26 lines

After ~15.5 hours finished successfully. Unfortunetally, I have no exact free space report before first balance, but it looked roughly like:

Label: 'DATA' uuid: 5b285a46-e55d-4191-924f-0884fa06edd8

          Total devices 16 FS bytes used 3.49TiB
          devid    1 size 558.41GiB used 448.66GiB path /dev/sda
          devid    2 size 558.41GiB used 448.66GiB path /dev/sdb
          devid    4 size 558.41GiB used 448.66GiB path /dev/sdd
          devid    5 size 558.41GiB used 448.66GiB path /dev/sde
          devid    7 size 558.41GiB used 448.66GiB path /dev/sdg
          devid    8 size 558.41GiB used 448.66GiB path /dev/sdh
          devid    9 size 558.41GiB used 448.66GiB path /dev/sdf
          devid   10 size 558.41GiB used 448.66GiB path /dev/sdi
          devid   11 size 558.41GiB used 448.66GiB path /dev/sdj
          devid   13 size 558.41GiB used 448.66GiB path /dev/sdk
          devid   14 size 558.41GiB used 448.66GiB path /dev/sdc
          devid   15 size 558.41GiB used 448.66GiB path /dev/sdl
          devid   16 size 558.41GiB used 448.66GiB path /dev/sdm
          devid   17 size 558.41GiB used 448.66GiB path /dev/sdn
          devid   18 size 837.84GiB used 448.66GiB path /dev/sdr
          devid   19 size 837.84GiB used 448.66GiB path /dev/sdq
          devid   20 size 837.84GiB used   0.00GiB path /dev/sds
          devid   21 size 837.84GiB used   0.00GiB path /dev/sdt
Quoted 3 lines

Are we doing something wrong? I found posts, where problems with balace of full filesystem are described. And as a recommendation is "add empty device, run balance, remove device".

It's raid10, so in this case, you probably need to add 4 devices. It's not required they be equal sizes but it's ideal.

Something is wrong there. Each new balanced chunk will free space on the first 16 drives (each new chunk is 9GB, each old one is 8GB, so the number of chunks on each of the old disks required to hold the same data decreases by 1/8th each time a chunk is relocated in balance). Every drive had at least 100GB of unallocated space at the start, so the first 9GB chunk should have been allocated without issue. Assuming nobody was aggressively writing to the disk during the balance to consume all available space, it should not have run out of space in a full balance.

You might have hit a metadata space reservation bug, especially on an older kernel. It's hard to know what happened without a log of the 'btrfs fi usage' data over time and a stack trace of the ENOSPC event, but whatever it was, it was probably fixed some time in the last 4 years.

There's a bug (up to at least 5.4) where scrub locks chunks and triggers aggressive metadata overallocations, which can lead to this result. It forms a feedback loop where scrub keeps locking the metadata chunk that balance is using, so balance allocates another metadata chunk, then scrub moves on and locks that new metadata chunk, repeat until out of space, abort transaction, all the allocations get rolled back and disappear. Only relevant if you were running a scrub at the same time as balance.

Quoted 9 lines
Quoted 6 lines
Quoted 2 lines

Are there some requirements on free space for balance even if you add new device?

The free space is reported upon adding the two devices; but the raid10 profile requires more than just free space, it needs free space on four devices.

I didn't realize that. It makes me sense now. So we are probably "wrong" with 18 devices. Multiple of 4 would be better I guess.

Not really. btrfs profile "raid10" distribute stripes over pairs of mirrored drives. It will allocate raid10 chunks on 4 + N * 2 drives at a time, but each chunk can use different drives so all the space is filled. Any number of disks above 4 is OK (including odd numbers), but sequential read performance will only increase when the number of drives increases to the next even number so the next highest stripe width can be used.

i.e. 5 drives will provide more space than 4, but 5 drives will not be significantly faster than 4. 6 is faster and larger than 4 or 5. 7 will be larger, 8 will be larger and faster, 9 will be larger, etc. (assuming all the drives are identical)

At some point there's a chunk size limit that kicks in and limits the number of drives in a chunk, but I'm not sure what the limit is (the kernel code does math on struct and block sizes and the result isn't obvious to me). I think it ends up being 10 of something, but not sure if the unit is drives (in which case your 16 drives were already over the limit) or gigabytes of logical chunk size (for raid10 that means 20 drives). Once that limit is reached, adding more drives only increases space and does not improve sequential read performance any further.

Quoted 1 line

Thank you!

#4 lore raw

Dne 11.09.2020 v 17:33 Zygo Blaxell napsal(a):

Quoted 45 lines · Miloslav Hůla
Quoted 34 lines

Dne 10.09.2020 v 21:18 Chris Murphy napsal(a):

Quoted 33 lines · Miloslav Hůla
Quoted 26 lines

After ~15.5 hours finished successfully. Unfortunetally, I have no exact free space report before first balance, but it looked roughly like:

Label: 'DATA' uuid: 5b285a46-e55d-4191-924f-0884fa06edd8

           Total devices 16 FS bytes used 3.49TiB
           devid    1 size 558.41GiB used 448.66GiB path /dev/sda
           devid    2 size 558.41GiB used 448.66GiB path /dev/sdb
           devid    4 size 558.41GiB used 448.66GiB path /dev/sdd
           devid    5 size 558.41GiB used 448.66GiB path /dev/sde
           devid    7 size 558.41GiB used 448.66GiB path /dev/sdg
           devid    8 size 558.41GiB used 448.66GiB path /dev/sdh
           devid    9 size 558.41GiB used 448.66GiB path /dev/sdf
           devid   10 size 558.41GiB used 448.66GiB path /dev/sdi
           devid   11 size 558.41GiB used 448.66GiB path /dev/sdj
           devid   13 size 558.41GiB used 448.66GiB path /dev/sdk
           devid   14 size 558.41GiB used 448.66GiB path /dev/sdc
           devid   15 size 558.41GiB used 448.66GiB path /dev/sdl
           devid   16 size 558.41GiB used 448.66GiB path /dev/sdm
           devid   17 size 558.41GiB used 448.66GiB path /dev/sdn
           devid   18 size 837.84GiB used 448.66GiB path /dev/sdr
           devid   19 size 837.84GiB used 448.66GiB path /dev/sdq
           devid   20 size 837.84GiB used   0.00GiB path /dev/sds
           devid   21 size 837.84GiB used   0.00GiB path /dev/sdt
Quoted 3 lines

Are we doing something wrong? I found posts, where problems with balace of full filesystem are described. And as a recommendation is "add empty device, run balance, remove device".

It's raid10, so in this case, you probably need to add 4 devices. It's not required they be equal sizes but it's ideal.

Something is wrong there. Each new balanced chunk will free space on the first 16 drives (each new chunk is 9GB, each old one is 8GB, so the number of chunks on each of the old disks required to hold the same data decreases by 1/8th each time a chunk is relocated in balance). Every drive had at least 100GB of unallocated space at the start, so the first 9GB chunk should have been allocated without issue. Assuming nobody was aggressively writing to the disk during the balance to consume all available space, it should not have run out of space in a full balance.

Quoted 5 lines

You might have hit a metadata space reservation bug, especially on an older kernel. It's hard to know what happened without a log of the 'btrfs fi usage' data over time and a stack trace of the ENOSPC event, but whatever it was, it was probably fixed some time in the last 4 years.

It happend with 4.9.210-1 (it's ~4 years old), now running 4.19.132-1 after Debian Buster upgrade.

Quoted 7 lines

There's a bug (up to at least 5.4) where scrub locks chunks and triggers aggressive metadata overallocations, which can lead to this result. It forms a feedback loop where scrub keeps locking the metadata chunk that balance is using, so balance allocates another metadata chunk, then scrub moves on and locks that new metadata chunk, repeat until out of space, abort transaction, all the allocations get rolled back and disappear. Only relevant if you were running a scrub at the same time as balance.

We were not running scrub in time of crash. But we were running rsync of whole btrfs to NetApp via NFS at the same sime.

Quoted 17 lines
Quoted 9 lines
Quoted 6 lines
Quoted 2 lines

Are there some requirements on free space for balance even if you add new device?

The free space is reported upon adding the two devices; but the raid10 profile requires more than just free space, it needs free space on four devices.

I didn't realize that. It makes me sense now. So we are probably "wrong" with 18 devices. Multiple of 4 would be better I guess.

Not really. btrfs profile "raid10" distribute stripes over pairs of mirrored drives. It will allocate raid10 chunks on 4 + N * 2 drives at a time, but each chunk can use different drives so all the space is filled. Any number of disks above 4 is OK (including odd numbers), but sequential read performance will only increase when the number of drives increases to the next even number so the next highest stripe width can be used.

Quoted 4 lines

i.e. 5 drives will provide more space than 4, but 5 drives will not be significantly faster than 4. 6 is faster and larger than 4 or 5. 7 will be larger, 8 will be larger and faster, 9 will be larger, etc. (assuming all the drives are identical)

Thanks for the clarification. I used to think that. It was one of pros to make a decision for btrfs RAID-10 few years before. I'm glad that's right.