Re: err=84/file:io_u.c:2033, func=io_u_sync_complete, error=Invalid or incomplete multibyte or wide character

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, 11 Dec 2018 at 18:01, Mohan Mohan <mohan@xxxxxxxxxx> wrote:
>
> Hi,
>
> Is the error below is expected one for the read operation? I am not getting any error for the write operation.
>
> root@ertino-ThinkPad-T431s:/home/ertino--verify crc32c

^^^ It looks like you asked for verifying reads (although this line
looks mangled - it's hard to tell without seeing the full job).

>
> xspp: (g=0): rw=randread, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1
> fio-3.12-17-g0fcbc0
> Starting 1 process
> xspp: Laying out IO file (1 file / 10MiB)
> verify: bad magic header 1950, wanted acca at file xspp.0.0 offset 630784, length 4096 (requested block: offset=630784, length=4096)

^^^ This is saying that the block read lacked the expected fio header.

>
> fio: pid=32055, err=84/file:io_u.c:2033, func=io_u_sync_complete, error=Invalid or incomplete multibyte or wide character
>
> xspp: (groupid=0, jobs=1): err=84 (file:io_u.c:2033, func=io_u_sync_complete, error=Invalid or incomplete multibyte or wide character): pid=32055: Tue Dec 11 23:09:30 2018

I'm guessing you asked fio to do verifying reads (e.g. fio
--name=verifread --rw=read --verify=crc32c --size=1M
--filename=fio.tmp) *before* you wrote data with a verify header.
Unless you have already written all of the file that is being read
back file with appropriate verification headers (i.e. you ran a write
job first) then it's expected fio will complain about not being able
to read the expected headers. A write job won't have this problem
because it writes data (sometimes before trying to read it back). See
https://fio.readthedocs.io/en/latest/fio_doc.html#cmdoption-arg-verify
for details.

-- 
Sitsofe | http://sucs.org/~sits/



[Index of Archives]     [Linux Kernel]     [Linux SCSI]     [Linux IDE]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux SCSI]

  Powered by Linux