earlephilhower
commented
3 years ago Thanks for the report.
Looking at the FIO bug reports, none have been confirmed and I've not seen this during my pretty extensive testing of many different enterprise-class SSDs from multiple vendors. W/O the FIO team acknowledging the issue, I'd lean towards an SSD firmware bug more than an FIO one, but in any case wouldn't disable one of the FIO built-in checks.
JosephNew
---Sustained Multi-Threaded Random Read Tests by Block Size---
Sustained Multi-Threaded Random Read Tests by Block Size, BS=512 4.36 8,920 85241.6 Sustained Multi-Threaded Random Read Tests by Block Size, BS=1024 8.74 8,947 85069.5 Sustained Multi-Threaded Random Read Tests by Block Size, BS=2048 17.65 9,035 82437.5 Sustained Multi-Threaded Random Read Tests by Block Size, BS=4096 34.71 8,886 85614.8 Sustained Multi-Threaded Random Read Tests by Block Size, BS=8192 68.37 8,751 86823.2 Sustained Multi-Threaded Random Read Tests by Block Size, BS=16384 101.25 6,480 112674.6 Sustained Multi-Threaded Random Read Tests by Block Size, BS=32768 149.72 4,791 155977.1 Sustained Multi-Threaded Random Read Tests by Block Size, BS=65536 213.44 3,415 218972.7 Sustained Multi-Threaded Random Read Tests by Block Size, BS=131072 235.98 1,888 387823.7
---Sequential Write Tests with Queue Depth=1 by Block Size---
Sequential Write Tests with Queue Depth=1 by Block Size, BS=512 00:00:02
FIO Error! fio /tmp/tmp3nliobop --output-format=json+ STDOUT: fio: pid=26829, err=84/file:io_u.c:2141, func=io_u_queued_complete, error=Invalid or incomplete multibyte or wide character { "fio version" : "fio-3.27", "timestamp" : 1624370289, "timestamp_ms" : 1624370289118,
.... STDERR: verify: bad header rand_seed 5513545861456270363, wanted 4503225908322601180 at file /dev/vdd offset 0, length 512 (requested block: offset=0, length=512)
Exception in thread Thread-21: Traceback (most recent call last): File "/usr/lib64/python3.6/threading.py", line 916, in _bootstrap_inner