Re: NetBSD Raid5, slow write speeds, using big disks?!

[Taylor R Campbell <riastradh%NetBSD.org@localhost>]

> From: smurfd <smurf.daemon%mail.smurfd.me@localhost>
> Date: Sat, 20 Jun 2026 18:55:25 +0200
> 
> Then on NetBSD, installed on the SSD i did this:
> 
> # gpt destroy wd0 ****
> # gpt create -Af wd0 ****
> # gpt add -t raid -l raid5@wd0 -b $(( 2048 )) -s 15628051053 wd0 ****

What was the alignment of this partition?

Does it make a difference if you use `-a 4096' or `-a 1m' with `gpt
add'?

zfs will warn if you try to `zpool add' a disk that natively uses
4096-byte sectors but advertises 512-byte sectors, implemented under
the hood by a read/modify/write cycle, if the starting sector is not
aligned to avoid the r/m/w cycles.  The message will look something
like:

ZFS WARNING: vdev ...: sectors are misaligned (alignment=..., firstaligned=...)

(In a future zfs update, if I recall correctly, `zpool add' will be
able to handle this misalignment on its own instead of telling you to
realign the partition.)

However, I don't think raidframe detects and warns about this case.
And perhaps gpt(8) should also use the disk's native alignment as the
default alignment instead of 512 bytes.

You should watch `iostat -x wd0 1' while you're doing writes to see if
the underlying disk's throughput and latency look reasonable.