0
> > 3 32062 xpt_action_default:entry da5: CDB: 2a 00 c8 00 25 d0 00 00 08 0=
0
> > 3 32062 xpt_action_default:entry da5: CDB: 2a 00 c8 00 25 e0 00 00 10 0=
0
> > 3 32062 xpt_action_default:entry da5: CDB: 8a 00 00 00 00 01 00 00 2a d=
0 00 00 00 08 00 00
> > 3 32062 xpt_action_default:entry da5: CDB: 8a 00 00 00 00 01 00 00 2c 5=
0 00 00 00 20 00 00
> > 3 32062 xpt_action_default:entry da5: CDB: 8a 00 00 00 00 01 00 00 2d 6=
0 00 00 00 08 00 00
> > 3 32062 xpt_action_default:entry da5: CDB: 8a 00 00 00 00 01 3e 36 29 b=
0 00 00 00 30 00 00
> > 2 32062 xpt_action_default:entry da5: CDB: 2a 00 c8 00 25 f0 00 00 08 0=
0
> > 3 32062 xpt_action_default:entry da5: CDB: 2a 00 c8 00 25 f8 00 00 08 0=
0
> > 3 32062 xpt_action_default:entry da5: CDB: 2a 00 c8 00 26 00 00 00 08 0=
0
>=20
> etc
>=20
> So the first byte of the CDB is the opcode. 2a is WRITE(10) and 8a is WRI=
TE(16). So FreeBSD is deciding to write rather a lot to this drive. That's =
why it's waking up. I'm not sure why it's deciding to write to this zpool (=
if I recall your original email correctly). Is it mounted and getting traff=
ic somehow? That's the next thing to look into. I didn't see any read opcod=
es, so maybe this is a resilver?

No, it's not mounted and it's a single drive, not in a mirror or anything. =
No resilvering.

I have just upgraded ZFS on my Artix Linux installation from ZFS 2.3.5 to 2=
.4.0 and
now the same is happening there as well, so that's something.