Slow transfer rate from WIN10 to OMV with WinSCP

raulfg3 · 1. Dezember 2021

Zitat von shecky66

Win 10 or OMV setup?

RE: Slow transfer rate from WIN10 to OMV with WinSCP

shecky66 · 16. Dezember 2021

Two days ago I bought some cat. 6 ethernet cables on Amazon (these: High Quality Network Cable) and I replaced all my home network cables with which the various peripherals were connected.

Result? The transfer speed has reached 1000 Mbps!

Thanks to everyone for the support, I have however increased my knowledge of network connections.

Jirri · 26. Dezember 2021

Zitat von auanasgheps

NO WAY, DO NOT USE SMB 1.x!!!!
You should not trust random and unreliable tutorials on Youtube.

Hello

When I wrote my first comment, I was an enthusiastic new user of Rpi4 and I felt I would help. The opposite was true, he has progressed a bit since then, and I can safely say that I have a lot to learn. This advice of mine was very bad.

So thanks for kicking in the right direction.

In order not to be completely out of the thread, at a lower bit rate, the change to a newer Linux kernel for Rpi is probably possible, and it is probably related to overclocking (as determined by the procedure of many different installation configurations, and a lot of information read)

Raspberry PI OS Buster with OMV 5 higher read and write speeds around 100MB / s and more.

Raspberry PI OS Bullseye OMV 6 reads 50MB / s write 100MB / s.

I've gone through things like UAS, network element speed and various settings and everything is OK.

But I will stay despite the lower speeds of the OMV 6 due to the nice user interface, and I will hope that the problems will be solved.

The only thing I miss is the new inability to install OMV on the desktop version of Raspberry PI OS, sd card copier with easy system disk backup (the option just suited me).

I understand why this is so, but I liked the opportunity. Maybe I read somewhere that there is a possibility to skip the dektop check during installation ... we'll see I'm looking for information

PC and everything around me has always been a big hobby, so building a home store on raspbian with OMV is endless fun for me for long winter evenings

Thanks for this site and everyone who contributes here.

Although I find that I would probably build it cheaper on conventional hardware than Rpi, but the fun around it

Thanks for this site and everyone who contributes here.

Google translator so please excuse it

Jirka

mi-hol · 27. Dezember 2021

Zitat von Jirri

new inability to install OMV on the desktop version of Raspberry PI OS

it was implemented to prevent issues after OS updates, the forum is full of them.

I was myself affected some time ago.
Therefore if you bypass the safeguard, don't be surprised if OMV wont work as expected after some time!

Zitat von Jirri

sd card copier with easy system disk backup

is available via several approaches.
1. OMV plugin for "backup", just use "dd" as backup method and select a sd-card contained in a USB-based carrier as target

2. Docker container with your favorite file manager tool

edubois · 8. Januar 2022

Zitat von Jirri

Hello
When I wrote my first comment, I was an enthusiastic new user of Rpi4 and I felt I would help. The opposite was true, he has progressed a bit since then, and I can safely say that I have a lot to learn. This advice of mine was very bad.
So thanks for kicking in the right direction.

In order not to be completely out of the thread, at a lower bit rate, the change to a newer Linux kernel for Rpi is probably possible, and it is probably related to overclocking (as determined by the procedure of many different installation configurations, and a lot of information read)
Raspberry PI OS Buster with OMV 5 higher read and write speeds around 100MB / s and more.
Raspberry PI OS Bullseye OMV 6 reads 50MB / s write 100MB / s.
I've gone through things like UAS, network element speed and various settings and everything is OK.

But I will stay despite the lower speeds of the OMV 6 due to the nice user interface, and I will hope that the problems will be solved.

The only thing I miss is the new inability to install OMV on the desktop version of Raspberry PI OS, sd card copier with easy system disk backup (the option just suited me).
I understand why this is so, but I liked the opportunity. Maybe I read somewhere that there is a possibility to skip the dektop check during installation ... we'll see I'm looking for information

PC and everything around me has always been a big hobby, so building a home store on raspbian with OMV is endless fun for me for long winter evenings
Thanks for this site and everyone who contributes here.
Although I find that I would probably build it cheaper on conventional hardware than Rpi, but the fun around it
Thanks for this site and everyone who contributes here.

Google translator so please excuse it
Jirka

Alles anzeigen

I ran into the same issue. Loved the new UX, but had terrible transfers speeds. Finally, I went back to OMV5. Was it fixed or are you still waiting?

Regards, ED.

Jirri · 10. Januar 2022

Zitat von edubois

I ran into the same issue. Loved the new UX, but had terrible transfers speeds. Finally, I went back to OMV5. Was it fixed or are you still waiting?

Regards, ED.

Since I use Pi (network HDD with occasional access, network movie playback and security camera recording), I'm actually very happy with everything.

If I copied hundreds of GB a day, I would mind. But because Pi with OMV does everything I want him to do, and he does it perfectly. So finding solutions to problems is really just the icing on the cake for me

It will take a long time, and that is good. At least I'll learn something.

Actually, I have one new finding - the writing speeds on Pi are higher if I turn off the device from the source and after turning on for some time (about 2 hours) the speeds are 70- + 100MB / s. Of course not always

I don't even count how many times I reinstalled Pi or uploaded the backup back, but it's routine

mi-hol · 10. Januar 2022

Jirri the signature mentions "RPi4 4GB--128GB SSD systém--2x2TB data SMB" but not how the drives are connected to the RPi and what file system is used. I suspect the root cause of degrading performance lies there or the home networking hardware.

Jirri · 11. Januar 2022

@ Mi-Hol

disks are connected using Ugreen USB 3.0 to SATA III Adapter Cable via USB 3.0 Charging HUB,Archer AX23, formatted ext4, network cat6 cables, transfers from another PC with windows with the same disk and adapter are high.

I've tried many different scenarios, Rpi just sends 50MB / s

But it writes 100 MB / s

Extra smb combination combination of all different options tested.

iperf3 and hdparm high speed

Is this thread probably not the most suitable place?

I would move somewhere, but where?

mi-hol · 12. Januar 2022

Well the common pattern for causing low I/O performance could be related to 2.5" HDDs using USB3 to SATA adapters.

I'd suggest to create a new thread with title "Please report I/O performance of 2.5" SSDs and HDDs using USB3 to SATA adapter"

I'll start it but need to prepare instructions first on "how to collect the required data" because i.e. "chipset" is not so easy to retrieve and the fio benchmark needs setup requiring technical LINUX skills too. As example please see https://www.jeffgeerling.com/b…-raid-on-raspberry-pi-cm4

What came to mind (first quick draft) are below fields:

Read IO Performance (fio for 1 MB random reads)	Write IO Performance (fio)	SATA Adapter Brand & Name	SATA adapter chipset	Drive Brand & Name	Drive Cache size
50MB / s	100 MB / s	Ugreen USB 3.0 to SATA III Adapter	something like JMicro 123?	something like Western Digital WD Green Mobile 2TB (WD20NPVX) ?	8Mb

This data would allow us to establish a baseline.

What is your view? Any comments or suggestions?

Note: I don't use any 2.5" drives because of performance limits experienced already.

mi-hol · 13. Januar 2022

Jirri could you please post output of command lsblk --list --nodeps --json --scsi ?

Jirri · 13. Januar 2022

mi-hol Hi, thanks for the interest you are

lsblk --list --nodeps --json --scsi

Code

{
   "blockdevices": [
      {"name":"sda", "hctl":"0:0:0:0", "type":"disk", "vendor":"Gigabyte", "model":"GIGABYTE_GP-GSTFS31120GNTD", "rev":"0   ", "serial":"SN212008944125", "tran":"usb"},
      {"name":"sdb", "hctl":"1:0:0:0", "type":"disk", "vendor":"WD      ", "model":"WDC_WD20SPZX-08UA7", "rev":"0   ", "serial":"WD-WXF2EA04ZRX9", "tran":"usb"},
      {"name":"sdc", "hctl":"2:0:0:0", "type":"disk", "vendor":"Ugreen  ", "model":"TOSHIBA_HDWL120", "rev":"0   ", "serial":"71A2PGP2T", "tran":"usb"},
      {"name":"sdd", "hctl":"3:0:0:0", "type":"disk", "vendor":"Seagate ", "model":"ST1000LM048-2E7172", "rev":"0   ", "serial":"WKPSHEP0", "tran":"usb"}
   ]
}

The chipset of all usb adapters is 1153E

I use Google Translate, so communication with me may not always be easy

Thank you

mi-hol · 14. Januar 2022

Jirri below is a draft for benchmark instructions. Are you familiar with bash?

PS: please translate your posts to English

Installing fio on Debian

Code

#NOTE: only draft, script will not fully work yet!!!
# due to error:
# Can not open temp file: DEVICE_MOUNT_PATH/iozone


# copy each line to bash (line by line) and execute it
#install benchmark components
sudo apt-get -y install wget
sudo apt-get -y install fio
wget -q https://raw.githubusercontent.com/geerlingguy/raspberry-pi-dramble/master/setup/benchmarks/disk-benchmark.sh
chmod +x disk-benchmark.sh

export BENCHMARK_SCRIPT_NAME=disk-benchmark-custom-tmp.sh
echo $BENCHMARK_SCRIPT_NAME
export DEVICE_MOUNT_PATH=/mnt/bmark 


#list connected storage devices
lsblk --list --nodeps --scsi
#note: interpretation of lsblk output is required to fill the next command parameter correctly!!
# sd? has to be replaced with the real drive to be tested examples: sda or sdb ...
export DEVICE_UNDER_TEST=/dev/sd?

sudo mount $DEVICE_UNDER_TEST $DEVICE_MOUNT_PATH 

sudo DEVICE_UNDER_TEST=$DEVICE_UNDER_TEST DEVICE_MOUNT_PATH=DEVICE_MOUNT_PATH ./disk-benchmark.sh >${BENCHMARK_SCRIPT_NAME}.log

#attach content of log file
cat ${BENCHMARK_SCRIPT_NAME}.log

Alles anzeigen

my 3.5'' drive has performance for READ: bw=205MiB/s (215MB/s)

Code

READ: bw=205MiB/s (215MB/s)

Code

fio-3.25
Starting 4 processes
Jobs: 4 (f=4): [R(4)][30.0%][r=213MiB/s][r=213 IOPS][eta 00m:07s]
Jobs: 4 (f=4): [R(4)][50.0%][r=207MiB/s][r=207 IOPS][eta 00m:05s]
Jobs: 4 (f=4): [R(4)][70.0%][r=183MiB/s][r=183 IOPS][eta 00m:03s]
Jobs: 4 (f=4): [R(4)][90.0%][r=180MiB/s][r=180 IOPS][eta 00m:01s]
Jobs: 4 (f=4): [R(4)][100.0%][r=179MiB/s][r=179 IOPS][eta 00m:00s]
fio-rand-read-sequential: (groupid=0, jobs=4): err= 0: pid=15923: Fri Jan 14 22:51:35 2022
  read: IOPS=204, BW=205MiB/s (215MB/s)(2112MiB/10319msec)
    slat (usec): min=59, max=237064, avg=19087.60, stdev=51282.32
    clat (msec): min=174, max=1850, avg=1186.90, stdev=204.81
     lat (msec): min=373, max=1850, avg=1205.99, stdev=203.54
    clat percentiles (msec):
     |  1.00th=[  451],  5.00th=[  835], 10.00th=[  995], 20.00th=[ 1045],
     | 30.00th=[ 1116], 40.00th=[ 1167], 50.00th=[ 1183], 60.00th=[ 1200],
     | 70.00th=[ 1301], 80.00th=[ 1385], 90.00th=[ 1435], 95.00th=[ 1452],
     | 99.00th=[ 1620], 99.50th=[ 1670], 99.90th=[ 1821], 99.95th=[ 1821],
     | 99.99th=[ 1854]
   bw (  KiB/s): min=67584, max=265970, per=95.30%, avg=199734.79, stdev=12479.73, samples=76
   iops        : min=   66, max=  258, avg=192.21, stdev=12.16, samples=76
  lat (msec)   : 250=0.19%, 500=1.47%, 750=1.52%, 1000=7.24%, 2000=89.58%
  cpu          : usr=0.09%, sys=1.61%, ctx=379, majf=0, minf=65635
  IO depths    : 1=0.2%, 2=0.4%, 4=0.8%, 8=1.5%, 16=3.0%, 32=6.1%, >=64=88.1%
     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=99.8%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.2%, >=64=0.0%
     issued rwts: total=2112,0,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=64

Run status group 0 (all jobs):
   READ: bw=205MiB/s (215MB/s), 205MiB/s-205MiB/s (215MB/s-215MB/s), io=2112MiB (2215MB), run=10319-10319msec

Disk stats (read/write):

Alles anzeigen

Jirri · 15. Januar 2022

mi-hol

I don't know Bash, but I'll look into it.

I should have all day tomorrow

Jirri · 15. Januar 2022

mi-hol This is the only thing I can get out of it. The script from you says ...

#NOTE: only draft, script will not fully work yet!!!

# due to error:

# Can not open temp file: DEVICE_MOUNT_PATH/iozone

Code

fio-3.25
Starting 4 processes

fio-rand-read-sequential: (groupid=0, jobs=4): err= 0: pid=5388: Sat Jan 15 20:57:50 2022
  read: IOPS=150, BW=151MiB/s (158MB/s)(1543MiB/10246msec)
    slat (usec): min=82, max=153148, avg=26041.30, stdev=42610.22
    clat (msec): min=133, max=2452, avg=1569.21, stdev=340.59
     lat (msec): min=221, max=2452, avg=1595.26, stdev=342.05
    clat percentiles (msec):
     |  1.00th=[  288],  5.00th=[  701], 10.00th=[ 1183], 20.00th=[ 1452],
     | 30.00th=[ 1536], 40.00th=[ 1620], 50.00th=[ 1687], 60.00th=[ 1737],
     | 70.00th=[ 1770], 80.00th=[ 1787], 90.00th=[ 1787], 95.00th=[ 1804],
     | 99.00th=[ 1921], 99.50th=[ 2366], 99.90th=[ 2467], 99.95th=[ 2467],
     | 99.99th=[ 2467]
   bw (  KiB/s): min=63488, max=192512, per=94.96%, avg=146432.00, stdev=7267.01, samples=72
   iops        : min=   62, max=  188, avg=143.00, stdev= 7.10, samples=72
  lat (msec)   : 250=0.52%, 500=2.46%, 750=2.66%, 1000=2.07%, 2000=91.51%
  lat (msec)   : >=2000=0.78%
  cpu          : usr=0.04%, sys=2.91%, ctx=767, majf=0, minf=65640
  IO depths    : 1=0.3%, 2=0.5%, 4=1.0%, 8=2.1%, 16=4.1%, 32=8.3%, >=64=83.7%
     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=99.7%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.3%, >=64=0.0%
     issued rwts: total=1543,0,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=64

Run status group 0 (all jobs):
   READ: bw=151MiB/s (158MB/s), 151MiB/s-151MiB/s (158MB/s-158MB/s), io=1543MiB (1618MB), run=10246-10246msec

Disk stats (read/write):
  sdc: ios=3053/0, merge=0/0, ticks=548400/0, in_queue=548400, util=99.12%

Running iozone 1024K random read and write tests...
        Iozone: Performance Test of File I/O
                Version $Revision: 3.492 $
                Compiled for 32 bit mode.
                Build: linux-arm

        Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
                     Al Slater, Scott Rhine, Mike Wisner, Ken Goss
                     Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
                     Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
                     Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
                     Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
                     Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer,
                     Vangel Bojaxhi, Ben England, Vikentsi Lapa,
                     Alexey Skidanov, Sudhir Kumar.

        Run began: Sat Jan 15 20:57:50 2022

        Include fsync in write timing
        O_DIRECT feature enabled
        Auto Mode
        File size set to 102400 kB
        Record Size 1024 kB
        Command line used: ./iozone -e -I -a -s 100M -r 1024k -i 0 -i 2 -f DEVICE_MOUNT_PATH/iozone
        Output is in kBytes/sec
        Time Resolution = 0.000001 seconds.
        Processor cache size set to 1024 kBytes.
        Processor cache line size set to 32 bytes.
        File stride size set to 17 * record size.
                                                              random    random     bkwd    record    stride

              kB  reclen    write  rewrite    read    reread    read     write     read   rewrite      read   fwrite frewrite    fread  freread
          102400    1024
Can not open temp file: DEVICE_MOUNT_PATH/iozone

Alles anzeigen

mi-hol · 16. Januar 2022

Zitat von Jirri

sdc READ: bw=151MiB/s (158MB/s)

sdc has a reasonable performance

Zitat von mi-hol

#list connected storage devices
lsblk --list --nodeps --scsi

what is the output of this command? I assume it will provide the reason of bad performance and why only /dev/sdc provides a performance result

Jirri · 16. Januar 2022

mi-hol

Code

NAME HCTL       TYPE VENDOR   MODEL                       REV SERIAL          TRAN
sda  0:0:0:0    disk Gigabyte GIGABYTE_GP-GSTFS31120GNTD 0    SN212008944125  usb
sdb  1:0:0:0    disk WD       WDC_WD20SPZX-08UA7         0    WD-WXF2EA04ZRX9 usb
sdc  2:0:0:0    disk Ugreen   TOSHIBA_HDWL120            0    71A2PGP2T       usb
sdd  3:0:0:0    disk Seagate  ST1000LM048-2E7172         0    WKPSHEP0        usb

sdc as the only one providing performance because I place

export DEVICE_UNDER_TEST = / dev / sd?

used

exportDEVICE_UNDER_TEST = / dev / sdc

I have to say that due to testing I also performed a new installation (so as not to get bored and rule out my mistakes) and the speeds are still the same.

SMB and NFS

Rpi >>> pc 53MB / s

pc >>> Rpi 100-110MB / s

Iperf3 930 Mbps

I do not get it

mi-hol · 16. Januar 2022

Zitat von Jirri

export DEVICE_UNDER_TEST = / dev / sd?
used
export DEVICE_UNDER_TEST = / dev / sdc

Just to be sure, the read performance is 50MB/s for all 4 disc?

sda is a SSD, right?

can you run the commands for sda?
export DEVICE_UNDER_TEST=/dev/sda

Jirri · 16. Januar 2022

mi-hol sdb-c-d for all hdd same 50-55MB / s

export DEVICE_UNDER_TEST=/dev/sda SSD

Code

Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
wget is already the newest version (1.21-1+deb11u1).
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
root@raspberrypi:/home/pi# sudo apt-get -y install fio
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
fio is already the newest version (3.25-2).
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
root@raspberrypi:/home/pi# wget -q https://raw.githubusercontent.com/geerlingguy/raspberry-pi-dramble/master/setup/benchmarks/disk-benchmark.sh
root@raspberrypi:/home/pi# chmod +x disk-benchmark.sh
root@raspberrypi:/home/pi# export BENCHMARK_SCRIPT_NAME=disk-benchmark-custom-tmp.sh
root@raspberrypi:/home/pi# echo $BENCHMARK_SCRIPT_NAME
disk-benchmark-custom-tmp.sh
root@raspberrypi:/home/pi# export DEVICE_MOUNT_PATH=/mnt/bmark
root@raspberrypi:/home/pi# lsblk --list --nodeps --scsi
NAME HCTL       TYPE VENDOR   MODEL                       REV SERIAL          TRAN
sda  0:0:0:0    disk Gigabyte GIGABYTE_GP-GSTFS31120GNTD 0    SN212008944125  usb
sdb  1:0:0:0    disk WD       WDC_WD20SPZX-08UA7         0    WD-WXF2EA04ZRX9 usb
sdc  2:0:0:0    disk Ugreen   TOSHIBA_HDWL120            0    71A2PGP2T       usb
sdd  3:0:0:0    disk Seagate  ST1000LM048-2E7172         0    WKPSHEP0        usb
root@raspberrypi:/home/pi# export DEVICE_UNDER_TEST=/dev/sda
root@raspberrypi:/home/pi# sudo mount $DEVICE_UNDER_TEST $DEVICE_MOUNT_PATH
mount: /mnt/bmark: mount point does not exist.
root@raspberrypi:/home/pi# sudo DEVICE_UNDER_TEST=$DEVICE_UNDER_TEST DEVICE_MOUNT_PATH=DEVICE_MOUNT_PATH ./disk-benchmark.sh >${BENCHMARK_SCRIPT_NAME}.log
open: No such file or directory
root@raspberrypi:/home/pi# cat ${BENCHMARK_SCRIPT_NAME}.log

Raspberry Pi disk benchmarks
Running fio sequential read test...
fio-rand-read-sequential: (g=0): rw=read, bs=(R) 1024KiB-1024KiB, (W) 1024KiB-1024KiB, (T) 1024KiB-1024KiB, ioengine=libaio, iodepth=64
...
fio-3.25
Starting 4 processes

fio-rand-read-sequential: (groupid=0, jobs=4): err= 0: pid=5861: Sun Jan 16 22:15:30 2022
  read: IOPS=342, BW=342MiB/s (359MB/s)(3458MiB/10097msec)
    slat (usec): min=77, max=98301, avg=11560.00, stdev=18203.37
    clat (msec): min=48, max=893, avg=711.99, stdev=109.27
     lat (msec): min=93, max=939, avg=723.56, stdev=109.85
    clat percentiles (msec):
     |  1.00th=[  180],  5.00th=[  558], 10.00th=[  651], 20.00th=[  693],
     | 30.00th=[  701], 40.00th=[  701], 50.00th=[  718], 60.00th=[  735],
     | 70.00th=[  760], 80.00th=[  785], 90.00th=[  810], 95.00th=[  827],
     | 99.00th=[  852], 99.50th=[  869], 99.90th=[  877], 99.95th=[  894],
     | 99.99th=[  894]
   bw (  KiB/s): min=167936, max=405504, per=97.17%, avg=340780.66, stdev=13771.37, samples=77
   iops        : min=  164, max=  396, avg=332.62, stdev=13.47, samples=77
  lat (msec)   : 50=0.03%, 100=0.29%, 250=1.33%, 500=2.54%, 750=61.25%
  lat (msec)   : 1000=34.56%
  cpu          : usr=0.13%, sys=2.97%, ctx=1302, majf=0, minf=65646
  IO depths    : 1=0.1%, 2=0.2%, 4=0.5%, 8=0.9%, 16=1.9%, 32=3.7%, >=64=92.7%
     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=99.9%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0%
     issued rwts: total=3458,0,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=64

Run status group 0 (all jobs):
   READ: bw=342MiB/s (359MB/s), 342MiB/s-342MiB/s (359MB/s-359MB/s), io=3458MiB (3626MB), run=10097-10097msec

Disk stats (read/write):
  sda: ios=6755/2, merge=0/8, ticks=535244/233, in_queue=535556, util=99.14%

Running iozone 1024K random read and write tests...
        Iozone: Performance Test of File I/O
                Version $Revision: 3.492 $
                Compiled for 32 bit mode.
                Build: linux-arm

        Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
                     Al Slater, Scott Rhine, Mike Wisner, Ken Goss
                     Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
                     Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
                     Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
                     Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
                     Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer,
                     Vangel Bojaxhi, Ben England, Vikentsi Lapa,
                     Alexey Skidanov, Sudhir Kumar.

        Run began: Sun Jan 16 22:15:30 2022

        Include fsync in write timing
        O_DIRECT feature enabled
        Auto Mode
        File size set to 102400 kB
        Record Size 1024 kB
        Command line used: ./iozone -e -I -a -s 100M -r 1024k -i 0 -i 2 -f DEVICE_MOUNT_PATH/iozone
        Output is in kBytes/sec
        Time Resolution = 0.000001 seconds.
        Processor cache size set to 1024 kBytes.
        Processor cache line size set to 32 bytes.
        File stride size set to 17 * record size.
                                                              random    random     bkwd    record    stride

              kB  reclen    write  rewrite    read    reread    read     write     read   rewrite      read   fwrite frewrite    fread  freread
          102400    1024
Can not open temp file: DEVICE_MOUNT_PATH/iozone

Alles anzeigen

mi-hol · 3. Februar 2022

Jirri using Ugreen USB 3.0 to SATA III Adapter seems to be the common fact for another case with low read performance see https://github.com/raspberrypi…33#issuecomment-776642481

Jirri · 3. Februar 2022

Zitat von mi-hol

Jirri using Ugreen USB 3.0 to SATA III Adapter seems to be the common fact for another case with low read performance see https://github.com/raspberrypi…33#issuecomment-776642481

Well, yes, but ASM1051E is mentioned in that post.

For me, it is ASM1153E before the firmware update, after the update it also reports other types of chips.

I have three different adapters, two ASM1153E and one Jmicron, all behaving the same on Rpi, four of each. I have ordered another type.

If I try them on Windows, the speeds are maximum and they behave correctly ...

Debian problem?

I ordered a CM4 and a board, so I'm waiting for it to get to me.

I will test and keep looking.

I started with Raspberry sometime in the fall, and the first installation was OMV 5 and Buster, everything was OK (full speed). Then only OMV 6

I've already read the thread you are referring to, there is probably nothing on the Internet that I would not read about this topic.

Just expectations from Rpi was probably too high, but I'm basically satisfied. I just still can't do it, and I'm still looking for a solution.

Thank you very much

EDIT:

I found one more thing in the advanced SMB settings.

When I add write cache size = 524288, this is shown

So I removed it from the settings.

Slow transfer rate from WIN10 to OMV with WinSCP

shecky66 16. Dezember 2021

Installing fio on Debian

Jetzt mitmachen!

Tags