Re-2: verify seems not to take place on iSCSI-Target writing

["Stefan Hübner" <shu@xtivate.de>]

Dear Jens,

this is the output.  I added some supplementary details in the hope it's useful ...

UltraBitch ~ # ISCSI_DEVICE=/dev/sdb SECONDS=$((20*3600)) fio iscsi-stress.fio 
iSCSI: (g=0): rw=randwrite, bs=4K-4M/4K-4M, ioengine=libaio, iodepth=16
fio 2.0.3
Starting 1 process
Jobs: 1 (f=1): [w] [17.3% done] [0K/113.2M /s] [0 /71  iops] [eta 03d:23h:28m:21s]
iSCSI: (groupid=0, jobs=1): err= 0: pid=18184
  write: io=7740.2GB, bw=112723KB/s, iops=63 , runt=72000222msec
    bw (KB/s)  : min= 3669, max=593424, per=100.00%, avg=113561.25, stdev=26868.27
  cpu          : usr=4.18%, sys=1.28%, ctx=4760480, majf=0, minf=106912
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 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=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
     issued    : total=r=0/w=4560578/d=0, short=r=0/w=0/d=0

Run status group 0 (all jobs):
  WRITE: io=7740.2GB, aggrb=112723KB/s, minb=115428KB/s, maxb=115428KB/s, mint=72000222msec, maxt=72000222msec

Disk stats (read/write):
  sdb: ios=80/18153218, merge=0/0, ticks=60/156412264, in_queue=156415080, util=100.00%
UltraBitch ~ # lsscsi 
[0:0:0:0]    disk    ATA      GB0160EAPRR      HPG2  /dev/sda 
[3:0:0:0]    cd/dvd  TSSTcorp CDDVDW SN-S083F  SB01  /dev/sr0 
[19:0:0:0]   disk    SCST_BIO WgAdboGFHRVydqoQ  200  /dev/sdb
UltraBitch ~ # iscsiadm -m session -P 3
iSCSI Transport Class version 2.0-870
version 2.0-871
Target: iqn.2012-02:dss-hh-pri.target0
        Current Portal: 192.168.4.231:3260,1
        Persistent Portal: 192.168.4.231:3260,1
                **********
                Interface:
                **********
                Iface Name: default
                Iface Transport: tcp
                Iface Initiatorname: iqn.2011-12.de.wwwgroup.intern.UltraBitch:openiscsi-6f9714c1547a126c14b1721b310daf11
                Iface IPaddress: 192.168.4.1
                Iface HWaddress: (null)
                Iface Netdev: (null)
                SID: 13
                iSCSI Connection State: LOGGED IN
                iSCSI Session State: LOGGED_IN
                Internal iscsid Session State: NO CHANGE
                ************************
                Negotiated iSCSI params:
                ************************
                HeaderDigest: None
                DataDigest: None
                MaxRecvDataSegmentLength: 262144
                MaxXmitDataSegmentLength: 8192
                FirstBurstLength: 65536
                MaxBurstLength: 262144
                ImmediateData: No
                InitialR2T: Yes
                MaxOutstandingR2T: 1
                ************************
                Attached SCSI devices:
                ************************
                Host Number: 19 State: running
                scsi19 Channel 00 Id 0 Lun: 0
                        Attached scsi disk sdb          State: running

Cheers,
Stefan

-------- Original Message --------
Subject: Re: verify seems not to take place on iSCSI-Target writing (01-Mrz-2012 14:02)
From:    Jens Axboe <axboe@kernel.dk>
To:      shu@wwwgroup.de

> On 03/01/2012 11:09 AM, Stefan Hübner wrote:
> > Hi List,
> > 
> > I'm using this FIO-job to stress-test iSCSI-Targets:
> > 
> > [global]
> > disable_lat=1
> > disable_clat=1
> > disable_slat=1
> > clat_percentiles=0
> > time_based
> > runtime=${SECONDS}
> > nice=10
> > prio=5
> > ioengine=libaio
> > iodepth=16
> > direct=1
> > iomem_align=4M
> > verify=crc32c-intel
> > verify_async=2
> > do_verify=1
> > verify_dump=1
> > [iSCSI]
> > rw=randwrite
> > bsrange=4k-4M
> > filename=${ISCSI_DEVICE}
> > 
> > ISCSI_DEVICE=/dev/sdX
> > SECONDS=some value how long it should take
> > 
> > When running this job I cannot see the verification taking place.  I.e. 
> > network thruput to the target (writing) is happening at around 115MB/s, but 
> > thruput from the target (reading/verifying) doesn't exceed 2,7MB/s.
> > 
> > Am I doing something wrong, here?
> 
> Does the fio job transition to the verify phase? You can see that in the
> [] listing, a V would be the verify. When the job exits, it also shows
> you both how many bytes were read and written. The read could should be
> the verified part.
> 
> So please post the output of the job so we can take a closer look.
> 
> -- 
> Jens Axboe
> 


To: axboe@kernel.dk
Cc: fio@vger.kernel.org