Hi, Thank you for this. The job command I'm using is: sudo fio --ioengine=libaio --size=100% --ramp_time=10 --runtime=40 --time_based=1 --verify_backlog=1 --verify_dump=1 --verify_fatal=1 --norandommap --numjobs=1 --direct=1 --rw=write --bs=1K --iodepth=8 --exitall_on_error --buffer_pattern=0x10000000000000000L --minimal --random_generator=tausworthe64 --name=/dev/sda --name=/dev/sdcd --name=/dev/sdce --name=/dev/sdcb On Sat, Aug 26, 2017 at 12:29 PM, Sitsofe Wheeler <sitsofe@xxxxxxxxx> wrote: > Hi, > > On 21 August 2017 at 03:32, siha lawrence <sihaj33@xxxxxxxxx> wrote: >> I'm using fio version 2.20 with the option exitall_on_error. On error, >> it exits and reports all stats as zero. >> How can this be modified so that we can get partial stats and not zeros? > > I'm afraid I can't reproduce the problem with a post 3.0 fio build: > > sudo -s > echo -e "0 511 zero\n511 1 error" | dmsetup create errorend > fio --exitall_on_error --direct=1 --bs=512 --rate=64k --name=null > --ioengine=null --size=512k --name=errorjob > --filename=/dev/mapper/errorend > null: (g=0): rw=read, bs=(R) 512B-512B, (W) 512B-512B, (T) 512B-512B, > ioengine=null, iodepth=1 > errorjob: (g=0): rw=read, bs=(R) 512B-512B, (W) 512B-512B, (T) > 512B-512B, ioengine=psync, iodepth=1 > fio-3.0-5-g168b > Starting 2 processes > fio: io_u error on file /dev/mapper/errorend: Input/output error: read > offset=261632, buflen=512 > fio: pid=26078, err=5/file:io_u.c:1756, func=io_u error, > error=Input/output error > > null: (groupid=0, jobs=1): err= 0: pid=26077: Sat Aug 26 05:24:23 2017 > read: IOPS=128, BW=64.1KiB/s (65.6kB/s)(257KiB/4001msec) > clat (nsec): min=800, max=1500, avg=1066.08, stdev=57.11 > lat (nsec): min=1100, max=12600, avg=3247.56, stdev=443.88 > clat percentiles (nsec): > | 1.00th=[ 1004], 5.00th=[ 1004], 10.00th=[ 1004], 20.00th=[ 1004], > | 30.00th=[ 1004], 40.00th=[ 1096], 50.00th=[ 1096], 60.00th=[ 1096], > | 70.00th=[ 1096], 80.00th=[ 1096], 90.00th=[ 1096], 95.00th=[ 1096], > | 99.00th=[ 1208], 99.50th=[ 1304], 99.90th=[ 1496], 99.95th=[ 1496], > | 99.99th=[ 1496] > bw ( KiB/s): min= 63, max= 65, per=50.39%, avg=64.00, stdev= > 0.58, samples=7 > iops : min= 127, max= 130, avg=128.14, stdev= 0.90, samples=7 > lat (nsec) : 1000=0.39% > lat (usec) : 2=99.61% > cpu : usr=0.40%, sys=0.00%, ctx=511, majf=0, minf=7 > 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=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% > complete : 0=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% > issued rwt: total=513,0,0, short=0,0,0, dropped=0,0,0 > latency : target=0, window=0, percentile=100.00%, depth=1 > errorjob: (groupid=0, jobs=1): err= 5 (file:io_u.c:1756, func=io_u > error, error=Input/output error): pid=26078: Sat Aug 26 05:24:23 2017 > read: IOPS=128, BW=63.0KiB/s (65.5kB/s)(256KiB/3993msec) > clat (nsec): min=4300, max=39500, avg=19138.94, stdev=1729.73 > lat (nsec): min=4600, max=41000, avg=20628.57, stdev=1758.02 > clat percentiles (nsec): > | 1.00th=[18304], 5.00th=[18560], 10.00th=[18560], 20.00th=[18560], > | 30.00th=[18816], 40.00th=[18816], 50.00th=[18816], 60.00th=[19072], > | 70.00th=[19328], 80.00th=[19328], 90.00th=[19584], 95.00th=[20096], > | 99.00th=[22912], 99.50th=[36096], 99.90th=[39680], 99.95th=[39680], > | 99.99th=[39680] > bw ( KiB/s): min= 63, max= 65, per=50.39%, avg=64.00, stdev= > 0.58, samples=7 > iops : min= 127, max= 130, avg=128.14, stdev= 0.90, samples=7 > lat (usec) : 10=0.20%, 20=93.75%, 50=5.86% > cpu : usr=0.40%, sys=0.20%, ctx=510, majf=0, minf=15 > 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=0.2%, 4=99.8%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% > issued rwt: total=512,0,0, short=0,0,0, dropped=0,0,0 > latency : target=0, window=0, percentile=100.00%, depth=1 > > Run status group 0 (all jobs): > READ: bw=128KiB/s (131kB/s), 63.0KiB/s-64.1KiB/s > (65.5kB/s-65.6kB/s), io=512KiB (524kB), run=3993-4001msec > > But I can't tell if there's more to your issue because you didn't > include the exact job file/command line you were running with. Can you > update your fio to at least 3.0 (or git master), reproduce the problem > and include at least the information in > https://github.com/axboe/fio/blob/master/REPORTING-BUGS ? Thanks! > > -- > Sitsofe | http://sucs.org/~sits/ -- To unsubscribe from this list: send the line "unsubscribe fio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
