On Fri, May 26, 2023 at 11:15:13PM +0200, jashax64@xxxxxxxxx wrote:
> Dear sir or madam,
>
> fio reports an error "io_u error on file /dev/nvme0n2: Value too large for
> defined data type: write offset=51539607552, buflen=65536" on my computer.
> I am using Western Digital ZN540 1TB. Do you have any idea on this? More
> importantly, **when I ran the same command a few days ago there were no
> errors**.
>
> Thank you!
> -Yijun Ma
>
> P.S. The command and original output are as follows:
>
> $ sudo fio --ioengine=psync --direct=1 --filename=/dev/nvme0n2 --rw=write
> --group_reporting --zonemode=zbd --bs=64k --offset_increment=8z
> --size=8z --numjobs=14 --job_max_open_zones=1 --name=seqwrite1
> seqwrite1: (g=0): rw=write, bs=(R) 64.0KiB-64.0KiB, (W) 64.0KiB-64.0KiB,
> (T) 64.0KiB-64.0KiB, ioengine=psync, iodepth=1
> ...
> fio-3.28
> Starting 14 processes
> fio: io_u error on file /dev/nvme0n2: Value too large for defined data
> type: write offset=51539607552, buflen=65536
> fio: pid=76338, err=75/file:io_u.c:1845, func=io_u error, error=Value too
> large for defined data type
Hello Yijun,
include/uapi/asm-generic/errno.h:#define EOVERFLOW 75 /* Value too large for defined data type */
block/blk-core.c: [BLK_STS_ZONE_ACTIVE_RESOURCE] = { -EOVERFLOW, "active zones exceeded" },
This suggests that you have too many active zones open.
Currently, fio does not handle active zones very well.
I know that someone in my team was working on a patch series which
adds support for the max active zones limit in zonemode=zbd in fio,
but I do not know the current status of the patch series.
A possible workaround could be to perform a:
# blkzone reset /dev/nvme0n2
to reset all the zones before starting a new benchmark.
Kind regards,
Niklas
