PERC H730P Mini IO hang

​my raid controller is H730P Mini, I have some disk device in this controller. I use fio tool to test disk device performance, this disk device is HDD, and I set JBOD for disk. I want to test hot unplug disk during run fio. test result like below:​

​[root@node-2 ~]# fio --name=test --rw=randwrite --bs=4k --runtime=600 --ioengine=libaio --iodepth=1 --numjobs=1 --filename=/dev/sdl --direct=1 --group_reporting=1 --time_based=1 --eta-newline 1​
​test: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=1​
​fio-3.1​
​Starting 1 process​
​Jobs: 1 (f=1): [w(1)][0.5%][r=0KiB/s,w=328KiB/s][r=0,w=82 IOPS][eta 09m:57s]​
​Jobs: 1 (f=1): [w(1)][0.8%][r=0KiB/s,w=324KiB/s][r=0,w=81 IOPS][eta 09m:55s]​
​Jobs: 1 (f=1): [w(1)][1.2%][r=0KiB/s,w=348KiB/s][r=0,w=87 IOPS][eta 09m:53s]​
​Jobs: 1 (f=1): [w(1)][1.5%][r=0KiB/s,w=332KiB/s][r=0,w=83 IOPS][eta 09m:51s]​
​Jobs: 1 (f=1): [w(1)][1.8%][r=0KiB/s,w=328KiB/s][r=0,w=82 IOPS][eta 09m:49s]​
​Jobs: 1 (f=1): [w(1)][2.2%][r=0KiB/s,w=300KiB/s][r=0,w=75 IOPS][eta 09m:47s]​
​Jobs: 1 (f=1): [w(1)][2.5%][r=0KiB/s,w=316KiB/s][r=0,w=79 IOPS][eta 09m:45s]​
​Jobs: 1 (f=1): [w(1)][2.8%][r=0KiB/s,w=308KiB/s][r=0,w=77 IOPS][eta 09m:43s]​
​Jobs: 1 (f=1): [w(1)][3.2%][r=0KiB/s,w=0KiB/s][r=0,w=0 IOPS][eta 09m:41s]​
​Jobs: 1 (f=1): [w(1)][3.5%][r=0KiB/s,w=0KiB/s][r=0,w=0 IOPS][eta 09m:39s]​
​fio: io_u error on file /dev/sdl: Input/output error: write offset=208743833600, buflen=4096​
​fio: pid=60989, err=5/file:io_u.c:1756, func=io_u error, error=Input/output error​

​test: (groupid=0, jobs=1): err= 5 (file:io_u.c:1756, func=io_u error, error=Input/output error): pid=60989: Tue Aug 4 19:33:55 2020​
​write: IOPS=63, BW=252KiB/s (258kB/s)(5732KiB/22758msec)​
​slat (usec): min=6, max=104, avg=14.95, stdev= 7.13​
​clat (usec): min=2617, max=52364, avg=12421.42, stdev=4212.41​
​lat (usec): min=2645, max=52389, avg=12436.63, stdev=4212.92​
​clat percentiles (usec):​
​| 1.00th=[ 4228], 5.00th=[ 6390], 10.00th=[ 7504], 20.00th=[ 8717],​
​| 30.00th=[ 9896], 40.00th=[11207], 50.00th=[12256], 60.00th=[13435],​
​| 70.00th=[14484], 80.00th=[15533], 90.00th=[17433], 95.00th=[18744],​
​| 99.00th=[21627], 99.50th=[23725], 99.90th=[45351], 99.95th=[52167],​
​| 99.99th=[52167]​
​bw ( KiB/s): min= 216, max= 368, per=100.00%, avg=318.44, stdev=30.65, samples=36​
​iops : min= 54, max= 92, avg=79.61, stdev= 7.66, samples=36​
​lat (msec) : 4=0.70%, 10=29.85%, 20=66.32%, 50=3.00%, 100=0.07%​
​cpu : usr=0.05%, sys=0.10%, ctx=1439, majf=0, minf=18​
​IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%​
​submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%​
​complete : 0=0.1%, 4=99.9%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%​
​issued rwt: total=0,1434,0, short=0,0,0, dropped=0,0,0​
​latency : target=0, window=0, percentile=100.00%, depth=1​

​I hot unplug disk at [eta 09m:43s], have about 4s IO down to 0,and then IO error exit. It is strange, I think IO should error exit immediately, why delay about 4s? Is this about raid controller?​