Very very slow transfer rates

Status
Not open for further replies.

Kory

Dabbler
Joined
Jul 21, 2017
Messages
10
I'm currently transferring large files, 100MB+ and my FreeNAS box is only receiving them at around 700 kB/s, anyone have any idea what is going on? I first mentioned this issue here: https://forums.freenas.org/index.ph...-im-having-really-slow-transfer-issues.56367/

But I'm reposting it here in the hopes that someone will have an idea what is going on. I have attached an additional debug log from today as well.

Old posts:
I'm having really really slow transfer speeds between Win 10 and my FreeNAS box.

Over SMB/CIFS I'm getting around 15 kB/s - 4 MB/s depending on file size. Via SCP I'm getting ~3 MB/s.

I ran the dd test and this system is very very fast (apparently on the order of 2.6 GB/s write and 9.1 GB/s read), see below:
Code:
[root@data] /mnt/data/data# dd if=/dev/zero of=tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 41.377843 secs (2594968100 bytes/sec)
[root@data] /mnt/data/data# dd of=/dev/zero if=tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 11.836208 secs (9071670711 bytes/sec)


Looking at the system panel stuff the CPU load is very small, the disks are barely being used. I have no idea why it is so so slow.

I have also tried to bypass the switch seeing if that was slowing it down and it is not. Everything in-between is 1 Gbps connection.

Any advice? I have attached my debug logs. Thanks in advance!

----------------------------------------------------------------
Yep something is really wrong on the networking side, weird... 21.8 Mbps = 2.7 MB/s

Any ideas where to start looking?
Code:
------------------------------------------------------------
Client connecting to 192.168.1.23, TCP port 5001
TCP window size: 512 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.1.29 port 59913 connected with 192.168.1.23 port 5001
[ ID] Interval	 Transfer	 Bandwidth
[ 3] 0.0-10.1 sec 26.1 MBytes 21.8 Mbits/sec

----------------------------------------------------------------
It pegs out the raspberry pi (limited to 100 Mbps):
Code:
------------------------------------------------------------
Client connecting to 192.168.1.23, TCP port 5001
TCP window size: 43.8 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.1.20 port 42068 connected with 192.168.1.23 port 5001
[ ID] Interval	 Transfer	 Bandwidth
[ 3] 0.0-10.0 sec 112 MBytes 93.8 Mbits/sec

----------------------------------------------------------------
Router is a Netgear Nighthawk X4S R7800 with latest firmware and then connected to an older 1 Gbps switch that is a Linksys branded Cisco SR2024 from 2006 and I have never had slow issues with it. From that switch it connects to another switch in another room that is a TP-LINK TL-SG108. The switch I tried to bypass is the Cisco SR2024 and it made no difference. I have attached the debug up above so I'm not sure what is going on, maybe a Win10 issue? I tried SCP on this same Win10 machine and had similar slow performance.

It is interesting that the RPi will max out, the only other thing I could try is to connect a machine into that same TL-SG108 switch and see if it is faster there with iperf.

Thanks for any hints, tips, or advice you or anyone else may offer.
----------------------------------------------------------------
Code:
CPU: Intel(R) Pentium(R) CPU G4400 @ 3.30GHz (3312.14-MHz K8-class CPU)
real memory = 18102616064 (17264 MB)
avail memory = 16482664448 (15719 MB)
em0: <Intel(R) PRO/1000 Network Connection 7.6.1-k> mem 0xf7000000-0xf701ffff irq 16 at device 31.6 on pci0
umass0: <SanDisk Ultra Fit, class 0/0, rev 2.10/1.00, addr 1> on usbus0
ada0: <WDC WD4000FYYZ-01UL1B3 01.01K04> ATA8-ACS SATA 3.x device
ada0: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada1: <WDC WD4000FYYZ-01UL1B3 01.01K04> ATA8-ACS SATA 3.x device
ada1: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada2: <WDC WD4000FYYZ-01UL1B3 01.01K04> ATA8-ACS SATA 3.x device
ada2: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada3: <WDC WD4000FYYZ-01UL1B3 01.01K04> ATA8-ACS SATA 3.x device
ada3: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
da0: <SanDisk Ultra Fit 1.00> Removable Direct Access SPC-4 SCSI device
da0: 40.000MB/s transfers


I was able to test it from a machine on that same TP-LINK switch and it too gets slow transfer speeds from Win10. Only about 10-20 MB/s read which is what I'm getting now on the first machine after rebooting router, cable modem, etc.
----------------------------------------------------------------
 

Attachments

  • debug-data-20170721164024.tgz
    382.2 KB · Views: 370
  • debug-data-20170804082733.tgz
    406.5 KB · Views: 389

m0nkey_

MVP
Joined
Oct 27, 2015
Messages
2,739
Please paste the output of ifconfig and paste it using the [ code ] tags.
 

Kory

Dabbler
Joined
Jul 21, 2017
Messages
10
Please paste the output of ifconfig and paste it using the [ code ] tags.
Code:
[x@x ~]$ ifconfig
em0: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
		options=40098<VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,VLAN_HWTSO>
		ether 38:d5:47:ad:83:78
		inet 192.168.1.23 netmask 0xffffff00 broadcast 192.168.1.255
		nd6 options=9<PERFORMNUD,IFDISABLED>
		media: Ethernet autoselect (1000baseT <full-duplex>)
		status: active
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384
		options=600003<RXCSUM,TXCSUM,RXCSUM_IPV6,TXCSUM_IPV6>
		inet6 ::1 prefixlen 128
		inet6 fe80::1%lo0 prefixlen 64 scopeid 0x2
		inet 127.0.0.1 netmask 0xff000000
		nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
bridge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
		ether 02:cb:c0:14:ce:00
		nd6 options=1<PERFORMNUD>
		id 00:00:00:00:00:00 priority 32768 hellotime 2 fwddelay 15
		maxage 20 holdcnt 6 proto rstp maxaddr 2000 timeout 1200
		root id 00:00:00:00:00:00 priority 32768 ifcost 0 port 0
		member: epair0a flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
				ifmaxaddr 0 port 4 priority 128 path cost 2000
		member: em0 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
				ifmaxaddr 0 port 1 priority 128 path cost 20000
epair0a: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
		options=8<VLAN_MTU>
		ether 02:ff:20:00:04:0a
		nd6 options=1<PERFORMNUD>
		media: Ethernet 10Gbase-T (10Gbase-T <full-duplex>)
		status: active
 

Kory

Dabbler
Joined
Jul 21, 2017
Messages
10
I have attached a copy of the read/write speed for CrystalDiskMark 5, 50MiB on the shared drive. As you can see the read speed is only 14-19 MB/s and the write speed came back as 0.2 MB/s.

The harddrives I'm reading from are attached in the second image but read is 117-143 MB/s and write is around 100 MB/s.

So it is definitely something between this Win10 box and the FreeNAS system.
 

Attachments

  • DiskMark64_2017-08-04_16-05-59.png
    DiskMark64_2017-08-04_16-05-59.png
    35.9 KB · Views: 472
  • DiskMark64_2017-08-04_16-13-18.png
    DiskMark64_2017-08-04_16-13-18.png
    37.1 KB · Views: 575

m0nkey_

MVP
Joined
Oct 27, 2015
Messages
2,739
Network troubleshooting 101 is required :)

I would start checking/replacing cables, make sure both your Windows machine and switch are negotiating GbE speeds.
 

Kory

Dabbler
Joined
Jul 21, 2017
Messages
10
Computer restart to fix this problem? Geez, Windows, really? really? See attached picture now, what a freaking joke windows! Anyone have a good Linux distro? It's been about 5 years since I seriously used one...
 

Attachments

  • DiskMark64_2017-08-04_22-03-46.png
    DiskMark64_2017-08-04_22-03-46.png
    36 KB · Views: 413

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
I can relate to Windoze issues. On Monday my work computer was very slow and it was taking forever to do IOPS with that hard drive. I rebooted and not effect, took just under 1 hour to reboot Windoze 10. I ended up powering off the unit for a few hours and then powering it back on, Shazam! it works again. I'm sure it was the fact of just powering it off was the solution but I have no idea why it happened. Well I have a new computer headed my way regardless. I can't afford to waste time troubleshooting a computer when the company pays me to manage the site.
 

Kory

Dabbler
Joined
Jul 21, 2017
Messages
10
I can relate to Windoze issues. On Monday my work computer was very slow and it was taking forever to do IOPS with that hard drive. I rebooted and not effect, took just under 1 hour to reboot Windoze 10. I ended up powering off the unit for a few hours and then powering it back on, Shazam! it works again. I'm sure it was the fact of just powering it off was the solution but I have no idea why it happened. Well I have a new computer headed my way regardless. I can't afford to waste time troubleshooting a computer when the company pays me to manage the site.

Yes, it is so frustrating at times. The funny thing is that this machine was off for a whole week and when I turned it back on I had the same problem. It seemed like windows kept defaulting to the Wifi connection even with the cable plugged in. I had to completely disable wifi and it still didn't work. After the reboot and disabling wifi that seems to be the ticket. Also, no idea why this wifi is so slow when I get 600 Mbps from speed test sites with the same wifi, but on the local network it just completely bogs down. But I'm glad to know I'm not the only one having weird windows issues.
 

gpsguy

Active Member
Joined
Jan 22, 2012
Messages
4,472
Maybe it was doing the Win 10 upgrade to the "Creators Update". ;-)

I hate MS's idea of WAAS. Just going from Win 10 1511 -> 1703 is a bit challenging.

On Monday my work computer was very slow
 

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
Maybe it was doing the Win 10 upgrade to the "Creators Update". ;-)
All updates are pushed by the company and they didn't mention to me of any problems like that. Of course I could have been the first complaint.
 
Status
Not open for further replies.
Top