You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Simplify your job file to the minimum options required to trigger the issue: does it still happen for a fixed block size? Without compression? Without verify_interval? With a single loop?
Please acknowledge the following before creating a ticket
Description of the bug:
bad header rand_seed verify fail when do fio randrw test
command:
sudo fio -filename=/dev/nvme0n1 -iodepth=64 -rw=randrw -bssplit=4k/30:16K/30:64K/40 -numjobs=1 -name=fiotest -direct=1 -ioengine=libaio -size=100% -rwmixread=50 -group_reporting -loops=3 -do_verify=1 -verify=crc32c-intel -verify_interval=4096 -random_generator=tausworthe64 -bssplit=4k/30:16K/30:64K/40 -buffer_compress_chunk=4k -buffer_compress_percentage=60
Starting 1 process
verify: bad header rand_seed 8588354323626423928, wanted 17337998516205655122 at file /dev/nvme0n1 offset 1348557807616, length 4096 (requested block: offset=1348557807616, length=65536)
fio: pid=344153, err=84/file:io_u.c:2074, func=io_u_queued_complete, error=Invalid or incomplete multibyte or wide character
dec: 8588354323626423928 hex: 772FF6C118F17678
dec:17337998516205655122 hex: F09CF1CA32B01452
Environment:
kernel:4.18.0-305.3.1.el8.x86_64
fio version:
fio-3.13
Reproduction steps
sudo fio -filename=/dev/nvme0n1 -iodepth=64 -rw=randrw -bssplit=4k/30:16K/30:64K/40 -numjobs=1 -name=fiotest -direct=1 -ioengine=libaio -size=100% -rwmixread=50 -group_reporting -loops=3 -do_verify=1 -verify=crc32c-intel -verify_interval=4096 -random_generator=tausworthe64 -bssplit=4k/30:16K/30:64K/40 -buffer_compress_chunk=4k -buffer_compress_percentage=60
Please help check if it is a known issue, thank you!
The text was updated successfully, but these errors were encountered: