Iperf3 error unable to connect to server connection timed out

iperf3: error — unable to connect to server: Connection timed out #979 Comments 5% of time) keep seeing Connection time out on my test bed. If I retest it usually is not seen. I am trying to understand what it means by this error message «iperf3: error — unable to connect to server: Connection […]

Содержание

  1. iperf3: error — unable to connect to server: Connection timed out #979
  2. Comments
  3. iperf3: error — unable to connect to server: Connection refused #1092
  4. Comments
  5. Context
  6. Bug Report
  7. iperf3: error — unable to receive control message: Connection timed out #320
  8. Comments
  9. The test bed looks like this:
  10. Expected behaviour
  11. What is Actually Happening
  12. Recreating the bug
  13. Sys info
  14. iperf server on debian squeeze #133
  15. Comments
  16. Linux SAM 2.6.33 #1 PREEMPT Wed Jun 30 11:35:16 EDT 2010 i586 GNU/Linux
  17. Server listening on 5201 [3]
  18. Iperf3 error client connection refused #758
  19. Comments
  20. Context
  21. Bug Report
  22. Enhancement Request

iperf3: error — unable to connect to server: Connection timed out #979

5% of time) keep seeing Connection time out on my test bed. If I retest it usually is not seen. I am trying to understand what it means by this error message «iperf3: error — unable to connect to server: Connection timed out» to help debug better. Below log snippet shows that the DUT is able to pass traffic on one port but failed on the other one. The connection time out message is persistent unless restart the whole test.

Version of iperf3:
iperf 3.1.3

Operating system (and distribution, if any):
Linux user-server-generic

18.04.1-Ubuntu SMP Thu Feb 7 14:06:04 UTC 2019 x86_64

The text was updated successfully, but these errors were encountered:

It looks like the log snippets show that you’re trying to run the same command twice. the first time it worked and the second time (same parameters) it didn’t. Correct?

Basically the error means that the client is trying to set up to the control connection (over TCP) to the server, but the client is not getting any response from the server (in other words it can’t set up that TCP connection). This can happen in cases where the server host is down or unreachable due to network routing or other issues like that. It’s usually not an intermittent thing.

I’m not sure how to interpret your comment of «able to pass traffic on one port but failed on the other one». Are there multiple network interfaces involved? If so, consider possible routing or host-based firewall issues. Is there a load-balancer involved (since you were apparently going to the same server IP address both times), and could that be an issue?

Finally, you’re using a version of iperf3 that’s almost four years old. Consider using a newer version (3.7 is current as of this writing), newer versions have many bug fixes and enhancements.

Источник

iperf3: error — unable to connect to server: Connection refused #1092

NOTE: The iperf3 issue tracker is for registering bugs, enhancement
requests, or submissions of code. It is not a means for asking
questions about building or using iperf3. Those are best directed
towards the iperf3 mailing list at iperf-dev@googlegroups.com or
question sites such as Stack Overflow
(http://www.stackoverflow.com/). A list of frequently-asked questions
regarding iperf3 can be found at http://software.es.net/iperf/faq.html.

Context

Version of iperf3: iperf 3.7 (cJSON 1.5.2)

Hardware: Intel CPU > i5-9300, Memory > 16G

Operating system (and distribution, if any): Windows 10 and running Ubuntu 20.04 LTS(WSL 2)

Please note: iperf3 is supported on Linux, FreeBSD, and macOS.
Support may be provided on a best-effort basis to other UNIX-like
platforms. We cannot provide support for building and/or running
iperf3 on Windows, iOS, or Android.

  • Other relevant information (for example, non-default compilers,
    libraries, cross-compiling, etc.):

Please fill out one of the «Bug Report» or «Enhancement Request»
sections, as appropriate.

Bug Report

Now I’m running iperf3 on two machine. and Two machines are running windows 10, WSL 2, Ubuntu 20.04.
iPerf3 version has same both server and client machine. iperf 3.7

Two machine is possible icmp request and reply without any problem.
But when I tried to start iperf3 from client I met this error.

  1. iPerf3 Server screen(Server IP is 192.168.200.253)
    root@SHINCH01-15U560:/usr/bin# iperf3 -s

Server listening on 5201

  1. iPerf Client Screen.(Client IP is 192.168.200.251)
    root@CHANGHEESHIN01-5501I5:/usr/bin# iperf3 -c 192.168.200.253 -i 2
    iperf3: error — unable to connect to server: Connection refused

I already shutdown windows firewall and ubuntu firewall with the following command

root@CHANGHEESHIN01-5501I5:/usr/bin# ufw disable
Firewall stopped and disabled on system startup
root@CHANGHEESHIN01-5501I5:/usr/bin# ufw status
Status: inactive

Can I know what is root cause?

The text was updated successfully, but these errors were encountered:

Источник

iperf3: error — unable to receive control message: Connection timed out #320

I am having an issue with iperf3, where at the end of a timed test session, the connection between the iperf3 server and client does not close correctly.

The test bed looks like this:

  • Command Server (128.0.0.2) — Connected to a dumb switch. This is used to
    issue commands to the clients
  • Switch
  • Client (128.0.1.9) — A Raspberry Pi with the latest version of iperf3. One
    interface is connected to the dumb switch, the other is connected to a gateway
  • Gateway (192.168.0.1) — Gateway for traffic to and from the server
  • Server (10.123.126.2) — A Debian machine running an iperf3 server

Expected behaviour

The iperf3 client should open a session with the iperf3 server, report some
data to stdout every ith second, close the session, and report some more data

What is Actually Happening

The iperf3 client opens a session with the iperf3 server, reports some data to
standard out every ith second up until the last few seconds of the run, then
hangs for about 15 minutes before reporting this message:

This bug occurs whenever the run lasts for more than 10 minutes or so (i.e.,
whenever the -t >= 600).

Recreating the bug

Sys info

No matter what the -t value is, as long as it is over 600 or so, the same behaviour is exhibited — iperf3 runs until just a few seconds before the run is scheduled to end, waits for about 15 minutes, then complains about not receiving a control message because of a Connection timeout.

SSH TCPKeepAlive is on. I can confirm using tcpdump that there is network activity over that port right up until the end of the run is scheduled, and then nothing during the timeout period. I have tried to recreate the problem using pv and netcat, and am unable to, which leads me to believe that this is a iperf3 problem.

Not sure how to move forward. Could you suggest another setting to look at? Perhaps this is a bug with iperf3?

The text was updated successfully, but these errors were encountered:

Источник

iperf server on debian squeeze #133

From wpmccormick on January 16, 2014 05:37:06

What steps will reproduce the problem? 1. fresh compile on debian squeeze
2. iperf3 -s or iperf3 -s -D
3. netstat -a; tcp/5201 is listening What is the expected output? What do you see instead? nmap 172.16.125.5 shows no port open
cannot connect clients except from the same machine What version of the product are you using? On what operating system? root@SAM.304-ms:/home/fiber# iperf3 -v
iperf version 3.0.1 (10 January 2014)
Linux SAM 2.6.33 #1 PREEMPT Wed Jun 30 11:35:16 EDT 2010 i586 GNU/Linux Please provide any additional information below. I thought that perhaps iperf might be binding to eth0, an unused dhcp port, but I think I ruled that out by swapping eth0 & eth1.

Client seems to work ok.

The text was updated successfully, but these errors were encountered:

From wpmccormick on January 16, 2014 08:55:13

In trying a different port (I used 11111), nmap did show the port as open but a non local client still couldn’t connect to it.

From bmah@es.net on January 16, 2014 09:24:04

If you can connect to the iperf server locally but not from a different host, this sounds like a firewall problem. By default the iperf server doesn’t bind to any particular port (it would if you use the -B option), so there’s no particular reason for iperf to take connections via one interface versus another.

When you say «cannot connect clients except from the same machine» what error message do you get from your client?

Any way you can tell if you have a firewall configured on this host and if so what its ruleset is? (I wish I could give more specific guidance but I’m more of a FreeBSD person, most of my Linux experience is CentOS.)

From wpmccormick on January 16, 2014 13:48:21

I agree it sounds like a firewall issue, but the fact that client and server are at the opposite ends of a VPN, and also that the the machine where the iperf server is running has NO iptables config (below), leaves me scratching my head. Even though I can connect locally, nmap still does not show the port as being open. Just to prove that this isn’t a fire wall issues, I can connect to a server that I wrote that is listening on tcp/11111. I didn’t do anything special to make this work.

/* testing client connection to another server app . it works */
root@fibuntu:

# telnet 172.16.125.5 11111
Trying 172.16.125.5.
Connected to 172.16.125.5.
Escape character is ‘^]’.
read|NO_FAN_FAULT_M1
!read|NO_FAN_FAULT_M1|1
disconect
ERROR : invalid command : disconect
quit
ERROR : invalid command : quit
disconnect

Connection closed by foreign host.

/* start iperf, the [3] is something I added so I could see what the sock fd was */
root@SAM.304-ms:/home/fiber# /usr/local/bin/iperf3 -s -4 -V
iperf version 3.0.1 (10 January 2014)

Linux SAM 2.6.33 #1 PREEMPT Wed Jun 30 11:35:16 EDT 2010 i586 GNU/Linux

Server listening on 5201 [3]

/* from another console, see what iperf is doing /
root@SAM.304-ms:/home/fiber# netstat -ap| grep 5201
tcp 0 0 *:5201 *:
LISTEN 1099/iperf3
root@SAM.304-ms:/home/fiber# ps aux|grep iperf
root 1099 0.0 0.0 1728 544 pts/0 S+ 16:41 0:00 /usr/local/bin/iperf3 -s -4 -V
root 1108 0.0 0.0 3284 724 pts/2 S+ 16:41 0:00 grep iperf

/* try to connect to the iperf server across and OpenVPN tunnel, server to client*/
root@fibuntu:

# /usr/local/bin/iperf3 -c 172.126.125.5 -p 5201 -V -t 60
iperf version 3.0.1 (10 January 2014)
Linux fibuntu 3.5.0-39-generic #60-Ubuntu SMP Tue Aug 13 18:33:05 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
iperf3: error — unable to connect to server: Connection timed out

/* see what ports are open on the remote side */
root@fibuntu:

# nmap -sT 172.16.125.5

Starting Nmap 6.00 ( http://nmap.org ) at 2014-01-16 15:33 CST
Nmap scan report for 172.16.125.5
Host is up (0.083s latency).
Not shown: 995 closed ports
PORT STATE SERVICE
21/tcp open ftp
22/tcp open ssh
53/tcp open domain
80/tcp open http
11111/tcp open vce

Nmap done: 1 IP address (1 host up) scanned in 11.82 seconds

/* see what ports are open on the local side */
root@SAM.304-ms:/home/fiber# nmap -sT localhost

Starting Nmap 5.00 ( http://nmap.org ) at 2014-01-16 16:34 EST
Interesting ports on localhost.localdomain (127.0.0.1):
Not shown: 994 closed ports
PORT STATE SERVICE
21/tcp open ftp
22/tcp open ssh
53/tcp open domain
80/tcp open http
111/tcp open rpcbind
11111/tcp open unknown

Nmap done: 1 IP address (1 host up) scanned in 0.83 seconds

/* check iptables */
root@SAM.304-ms:/home/fiber# iptables -S
-P INPUT ACCEPT
-P FORWARD ACCEPT
-P OUTPUT ACCEPT
root@SAM.304-ms:/home/fiber# iptables -L
Chain INPUT (policy ACCEPT)
target prot opt source destination

Chain FORWARD (policy ACCEPT)
target prot opt source destination

Chain OUTPUT (policy ACCEPT)
target prot opt source destination

Источник

Iperf3 error client connection refused #758

NOTE: The iperf3 issue tracker is for registering bugs, enhancement
requests, or submissions of code. It is not a means for asking
questions about building or using iperf3. Those are best directed
towards the iperf3 mailing list at iperf-dev@google-groups.com or
question sites such as Stack Overflow
(http://www.stackoverflow.com/). A list of frequently-asked questions
regarding iperf3 can be found at http://software.es.net/iperf/faq.html.

Context

Version of iperf3: 3.1.3

Hardware: CPU 2.6HZ

Operating system (and distribution, if any): windows 10

Please note: iperf3 is supported on Linux, FreeBSD, and macOS.
Support may be provided on a best-effort basis to other UNIX-like
platforms. We cannot provide support for building and/or running
iperf3 on Windows, iOS, or Android.

  • Other relevant information (for example, non-default compilers,
    libraries, cross-compiling, etc.):

Please fill out one of the «Bug Report» or «Enhancement Request»
sections, as appropriate.

Bug Report

Expected Behavior
it has to link a connectivity between Server machine and client machine to showcase transfer rate and bandwidth

Actual Behavior
its throwing an error called iperf3 cannot be connected connection refused

Steps to Reproduce

Please submit patches or code changes as a pull request.

Enhancement Request

If submitting a proposed implementation of an enhancement request,
please use the pull request mechanism.

The text was updated successfully, but these errors were encountered:

Источник

I want to get iperf3 to benchmark a simple wireguard setup on my LAN.  I have two machines, foo is 192.168.1.228 and bar is 192.168.1.112.

These IP addresses are fully functional on their respective eth0 interfaces.  Each one also has a wg0 interface configured as shown below.  I can get iperf3 to work over the non-wireguard interfaces, but not over the wireguard interfaces.  Am I using iperf3 wrong?  Did I misconfigure wg?

Details
Here are the /etc/wireguard/wg0.conf files for both machines (I don’t care if the public/private keys are present/this is only a test environment and it will allow others to simply copy/paste if they want to test as well):

On foo:

[Interface]
Address = 10.0.9.15/24
SaveConfig = true
ListenPort = 500
PrivateKey = kLac+M+JJJ+gpsKo1DeUyfuwfDwGeKMj81+M3Z3mUkw=

[Peer]
PublicKey = y6DrsFVIgqBMc8joSuEytCpQEywk5dPKRoeLZZC9H34=
AllowedIPs = 10.0.9.16/32

On bar:

[Interface]
Address = 10.0.9.16/24
SaveConfig = true
ListenPort = 500
FwMark = 0xca6c
PrivateKey = 4EwMqSyvLaHrv5JvAAyKnmdbw9UN4rsjydkPXb5ll28=

[Peer]
PublicKey = SkBs9t96znOr+d60tgPlXkp7z6YABvfxD9KNVzZIP3I=
AllowedIPs = 0.0.0.0/0
Endpoint = 192.168.1.228:500

Again, both wireguard interfaces are up and connected.  For example, on foo:

# wg
interface: wg0
  public key: SkBs9t96znOr+d60tgPlXkp7z6YABvfxD9KNVzZIP3I=
  private key: (hidden)
  listening port: 500

peer: y6DrsFVIgqBMc8joSuEytCpQEywk5dPKRoeLZZC9H34=
  endpoint: 192.168.1.112:500
  allowed ips: 10.0.9.16/32
  latest handshake: 10 minutes, 41 seconds ago
  transfer: 32.45 KiB received, 8.24 KiB sent

If I run iperf3 in server mode on foo and in client mode on bar, it works as expected using the non-wireguard interfaces:

On foo:

% iperf3 -s -B 192.168.1.228
-----------------------------------------------------------
Server listening on 5201
-----------------------------------------------------------
Accepted connection from 192.168.1.112, port 58575
[  5] local 192.168.1.228 port 5201 connected to 192.168.1.112 port 46261
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-1.00   sec   107 MBytes   900 Mbits/sec
[  5]   1.00-2.00   sec   112 MBytes   939 Mbits/sec
[  5]   2.00-3.00   sec   112 MBytes   940 Mbits/sec
[  5]   3.00-4.00   sec   112 MBytes   938 Mbits/sec
[  5]   4.00-5.00   sec   112 MBytes   940 Mbits/sec
[  5]   4.00-5.00   sec   112 MBytes   940 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-5.00   sec   596 MBytes  1.00 Gbits/sec                  receiver
iperf3: the client has terminated

On bar:

% iperf3 -c 192.168.1.228 -B 192.168.1.112
Connecting to host 192.168.1.228, port 5201
[  5] local 192.168.1.112 port 46261 connected to 192.168.1.228 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec   113 MBytes   950 Mbits/sec    0    281 KBytes
[  5]   1.00-2.00   sec   112 MBytes   939 Mbits/sec    0    296 KBytes
[  5]   2.00-3.00   sec   112 MBytes   938 Mbits/sec    0    296 KBytes
[  5]   3.00-4.00   sec   112 MBytes   938 Mbits/sec    0    296 KBytes
[  5]   4.00-5.00   sec   112 MBytes   944 Mbits/sec    0    308 KBytes
^C[  5]   5.00-5.33   sec  36.7 MBytes   939 Mbits/sec    0    308 KBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-5.33   sec   598 MBytes   942 Mbits/sec    0             sender
[  5]   0.00-5.33   sec  0.00 Bytes  0.00 bits/sec                  receiver
iperf3: interrupt - the client has terminated

When I try using the IP addresses of the respective wireguard interfaces, I get nothing:

On foo:

% iperf3 -s -B 10.0.9.15
-----------------------------------------------------------
Server listening on 5201
-----------------------------------------------------------

On bar:

% iperf3 -c 10.0.9.15 -B 10.0.9.16
iperf3: error - unable to connect to server: Connection timed out

Last edited by graysky (2018-12-18 22:49:53)

iperf -c 192.168.3.2 -i 2
connect failed: Connection refused

3.2 is another machine. I think the port might be blocked.

How can I make this work?

asked Jun 16, 2011 at 7:28

Hao's user avatar

4

You also get a «connection refused» error when you use an iperf v3 client to send to an iperf v2 server.

The message is slightly different though:

iperf3: error - unable to connect to server: Connection refused

As the iperf3 doc states, they are not compatible (at least up to version 3.1.2). You can check the version on both server and client with

iperf -v

answered Apr 4, 2016 at 10:46

J.P. Tosoni's user avatar

J.P. TosoniJ.P. Tosoni

5154 silver badges6 bronze badges

2

I doubt that this is a firewall issue. Usually, firewalls employ the DROP rule for packets that are not allowed. A «connection refused» means that there is no application listening on the port it tries to connect to. The destination host indicates this with a corresponding ICMP message to the source.

In order to run iperf you must have a server running on the machine which you iperf to. To do this run iperf -s on the server. After that you can then run iperf -c <server IP or hostname> -i 2

After running this you will get an ouput that looks something like this:

------------------------------------------------------------
Client connecting to <server IP>, TCP port 5000
TCP window size:   256 KByte (default)
------------------------------------------------------------
[  3] local 123.123.123.123 port 4000 connected with 123.123.123.123 port 5000
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-10.0 sec  1.1 GBytes  1.09 Gbits/sec

Community's user avatar

answered Jun 16, 2011 at 8:05

Dabu's user avatar

DabuDabu

3591 gold badge5 silver badges23 bronze badges

Try turning off any firewall on the machines. On many linux distros, you can do

service firewall off

to disable the firewall.

answered Jun 16, 2011 at 7:30

Shannon Nelson's user avatar

Unable to get local wired transfer speeds higher than 11 MB/s


  • Thread starter

    shadowbeh2


  • Start date

    Jul 26, 2017

shadowbeh2


  • #1

Hello, here is my issue:

I have two PCs, one is my everyday use gaming desktop, the other is a small Plex server that I have recently built. My issue is that I am unable to transfer files between these two PCs faster than 11MB/s. I find it odd because both PCs can achieve higher transfer rates over the internet than they can locally, seems backwards to me.

Here is some extra info that may be useful:

Gaming PC:

  • Operating System: Windows 10 Pro 64-bit
    CPU: Intel Core i7 4770K @ 3.50GHz
    RAM: 16.0GB Dual-Channel DDR3
    Motherboard: MSI Z87-GD65 GAMING

Plex Server:

  • Operating System: Windows Server 2012 R2 Datacenter 64-bit
    CPU: AMD FX-6300
    RAM: 8.00GB Dual-Channel DDR3
    Motherboard: ASUSTeK Computer INC. M5A78L-M PLUS/USB3

Both of these PCs are connected to a 5 port gigabit switch, which is connected to an Asus RT-ACRH13 router.

Thank you!



Aug 9, 2012



26,234



2,118



116,690

8,227


  • #9

You must have a port that is running at 100mbit rather than 1gbit. Best place to look is in your switch if it has the ability to let you log in. Many switches also have lights. Otherwise you can dig around, microsoft is very inconsistent between software versions where they display the negotiated speeds.

Unless the speed is set to 100m ina end device or your switch can only run 100m the problem is almost always a damaged cable. Only takes a slightly loose cable end on one wire to cause it to drop to 100m.



Aug 9, 2012



26,234



2,118



116,690

8,227


  • #2

Try running the old IPERF program on both machine. It is a really simplistic program that only tests the network it is not affect by memory or disk or even cpu on large machines. You should see about 850-900mbits/sec when they are connected like your are.

Normally I would say you have a 10mbit connection because 11MB is about what you get on a 10mbit ethernet with the overhead. Since it runs faster to the internet then your cables/ports are fine.

If iperf runs good I would supect something in the disk system. Also a large number of small files will transfer much slower than 1 large file. Many times the name of the file and directory take up more room than the file itself, this is especially true when you copy some games that have many thousands of extremely tiny files.

shadowbeh2


  • #3

I have found the IPERF program but I am unsure on how to run the tests you are looking for, any tips?

Also, most of the files I am transferring are single large MP4s or MKVs



Aug 9, 2012



26,234



2,118



116,690

8,227


  • #4

You just run iperf -s on one end and iperf -c xx.xx.xx.xx on the other where xx.xx.xx.xx is the ip of the machine you used the -s option on. There are many options but the most simplistic test generally will show if you have any issues.

shadowbeh2


  • #5

Okay, I seem to have run into an issue.

Running the server side on my Plex box, its sitting with the «Server listening on 5201» message

On the client end i enter «iperf.exe -c *serverIP*» and am met with «error — unable to connect to server: connection timed out»

I have verified that the IP for the server is correctly entered.

  • #6

Okay, I seem to have run into an issue.

Running the server side on my Plex box, its sitting with the «Server listening on 5201» message

On the client end i enter «iperf.exe -c *serverIP*» and am met with «error — unable to connect to server: connection timed out»

I have verified that the IP for the server is correctly entered.

Their system must be down.
https://www.youtube.com/watch?v=ILVfzx5Pe-A



Aug 9, 2012



26,234



2,118



116,690

8,227


  • #7

Okay, I seem to have run into an issue.

Running the server side on my Plex box, its sitting with the «Server listening on 5201» message

On the client end i enter «iperf.exe -c *serverIP*» and am met with «error — unable to connect to server: connection timed out»

I have verified that the IP for the server is correctly entered.

Pretty much the only way that can happen especially if you can share files to the same server address is some firewall is blocking iperf. iperf is really stupid it opens a simple tcp session between the 2 ip and transfers data. It is as simple as it gets.

shadowbeh2


  • #8

Got it through the firewall, here is what I got on the server end, I cannot currently get the output of the client side as I am at work and my remote connection is not working right now.

Server listening on 5201
————————————————————
Accepted connection from 192.168.50.29, port 50252
[ 5] local 192.168.50.83 port 5201 connected to 192.168.50.29 port 50253
[ ID] Interval Transfer Bandwidth
[ 5] 0.00-1.00 sec 11.1 MBytes 93.3 Mbits/sec
[ 5] 1.00-2.00 sec 11.3 MBytes 94.9 Mbits/sec
[ 5] 2.00-3.00 sec 11.3 MBytes 94.8 Mbits/sec
[ 5] 3.00-4.00 sec 11.3 MBytes 94.8 Mbits/sec
[ 5] 4.00-5.00 sec 11.3 MBytes 94.9 Mbits/sec
[ 5] 5.00-6.00 sec 11.3 MBytes 94.8 Mbits/sec
[ 5] 6.00-7.00 sec 11.3 MBytes 94.9 Mbits/sec
[ 5] 7.00-8.00 sec 11.3 MBytes 94.9 Mbits/sec
[ 5] 8.00-9.00 sec 11.3 MBytes 94.9 Mbits/sec
[ 5] 9.00-10.00 sec 11.3 MBytes 94.8 Mbits/sec
[ 5] 10.00-10.03 sec 360 KBytes 94.8 Mbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ ID] Interval Transfer Bandwidth
[ 5] 0.00-10.03 sec 0.00 Bytes 0.00 bits/sec sender
[ 5] 0.00-10.03 sec 113 MBytes 94.7 Mbits/sec receiver
————————————————————
Server listening on 5201
————————————————————



Aug 9, 2012



26,234



2,118



116,690

8,227


  • #9

You must have a port that is running at 100mbit rather than 1gbit. Best place to look is in your switch if it has the ability to let you log in. Many switches also have lights. Otherwise you can dig around, microsoft is very inconsistent between software versions where they display the negotiated speeds.

Unless the speed is set to 100m ina end device or your switch can only run 100m the problem is almost always a damaged cable. Only takes a slightly loose cable end on one wire to cause it to drop to 100m.

shadowbeh2


  • #10

Okay, I’ll swap some cables around and see what that does when I get home

shadowbeh2


  • #11

Issue has been resolved. Problem was a combination of wonky switch and some bad cat6 cables.

Thank you bill001g for your help =]

  • Advertising
  • Cookies Policies
  • Privacy
  • Term & Conditions
  • Topics

[H]ard|Forum

  • [H]ard|Ware

  • SSDs & Data Storage

You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.

10 Gb Connection


  • Thread starter

    N Bates


  • Start date

    May 26, 2018

  • #1

Joined
Jul 15, 2017
Messages
174


I have purchased two Intel X540 T2 and connected my windows 10 pc and the Omnios 101526 NAS, I have updated the drivers on the windows PC, I left the one in the NAS as is as I don’t know how to update these.

I have given both static addresses, enabled and Jumbo frames, I have connected both with an ethernet cable and both connections are showing green.

I have added a networ connection on the windows machine for both NICs, however, when I copy files between the two machines is still looking like the 1gb connection is being used, is there any way where I can force the connections to use 10 gbe for file transfer?

  • #2

Joined
Aug 24, 2005
Messages
1,884


You probably put them in the same network range your PC was already using, which is the default.

Either change the static IP’s on both boxes or add a static route to the NAS ip in windows to force it to go to the 10gb interface

  • #3

Joined
Jul 15, 2017
Messages
174


I have put both NIC’s outside the network range, the network is in 192.168.1.1 and the NIC’s are in 10.10.10.10 and 10.10.10.20.

The NAS has already got a static IP, I have added the NAS IP to the windows etchost file, is this what you mean by forcing it to go to the 10 gb interface?

  • #4

Joined
Oct 30, 2000
Messages
5,233


Go into the adapter properties, make sure the speed is set to 10G. Those support 1G as well. I’m assuming you’re using Cat-6 cables?? Cat-5 does not support 10G,

Last edited: May 26, 2018

  • #5

Joined
Jul 15, 2017
Messages
174


The adaptor properties in windows show as 10g and using cat 6A cable.

On the NAS side, I can see in Napp-it that the adaptor on there is at 10g also.

  • #6

Joined
Jul 24, 2008
Messages
5,155


You’ll also need to change the network connection priority in Windows to move the 10Gb connection to primary.

  • #7

Joined
Aug 24, 2005
Messages
1,884


The 10gb doesn’t need to be primary, if he’s hitting something on 10.x.x.x and he has a nic in that range it will go out that nic instead of the default one.

On the etc hosts file did you add the NAS as the 10.10.10.20 IP? Or was it in there before from the original IP on the 192 subnet?

From your windows PC can you ping/trace the NAS 10 net IP address?

  • #8

Joined
Jul 15, 2017
Messages
174


on the etchosts on the windows machine I have the below, which is the NAS IP’s

10.10.10.10 NAS
192.168.1.x NAS
10.10.10.15 NAS

The 10.10.10.15 is the second port on the intel NIC but not currently used, should I remove the 192.168.1.x IP of the NAS and leave only the ones referring to the Intel x540 NIC?

I have been reading a bit and I think I maybe using the windows machine etchosts file wrongly, the IP addresses on there should only be for the windows machine and not the NAS, not 100% sure though, I will keep trying and report back.

  • #9

Joined
Aug 24, 2005
Messages
1,884


The way windows works, it looks at hosts file before doing a DNS lookup. If you are mounting it as ‘NAS’, then when it tries to associate that name to an IP, it will look at hosts file, if it doesn’t have an entry, it does a DNS lookup. So the IP addresses in the hosts file are for other machines/ips, not the windows PC itself.

What IP gets resolved when you ping the NAS by name from the windows PC?

  • #10

Joined
Jul 15, 2017
Messages
174


This is really strange, I could access the NAS from the windows PC before when typing the NAS IP, however now when I ping the NAS, I get «request timed out» and when I ping the windows PC from the NAS I get «no answer», I might have to change the cable and see if this makes a difference.

EDIT:

Do I have to set new shares for the 10gb NIC’s to use instead of the existing share that was created?

Last edited: May 28, 2018

  • #11

_Gea

Supreme [H]ardness

Joined
Dec 5, 2010
Messages
4,100


I would never use two nics with an ip from the same subnet.
This can result in a traffic that goes in into one nic and back through the other — not finding its way back.

If a connect via ip works, set only this entry in the hosts file

For initial tests, remove Jumbo as this can be a good for performance or troubles.

On Windows I would disable interrupt throtteling on Windows in the driver settings for the nic.
Increasing tcp buffers on OmniOS can also increase 10G performance.

btw
You cannot bind a share to a nic as the SMB server is listening on all nics.

  • #12

Joined
Oct 30, 2000
Messages
5,233


  • #13

Joined
Oct 23, 2007
Messages
2,214


I have purchased two Intel X540 T2 and connected my windows 10 pc and the Omnios 101526 NAS, I have updated the drivers on the windows PC, I left the one in the NAS as is as I don’t know how to update these.

I have given both static addresses, enabled and Jumbo frames, I have connected both with an ethernet cable and both connections are showing green.

I have added a networ connection on the windows machine for both NICs, however, when I copy files between the two machines is still looking like the 1gb connection is being used, is there any way where I can force the connections to use 10 gbe for file transfer?

Seems everyone forgot that you can only go as fast as the HD can read and write. Is your NAS and PC have SSDs? And if so, are you copying between the SSDs? Regular spinning disks will never be able to go over what a 1gb connection can do. You need SSDs for that.

  • #14

Joined
Jul 15, 2017
Messages
174


So, should the static IP of the dual port NIC be like:
10.10.10.10
10.10.11.15

and the second dual port NIC be:
10.10.12.20
10.10.13.25

Will the above be better, is that what you mean by two NIC’s not in the same subnet?

This is what I have in the hosts file of the windows machine:
10.10.10.10 NAS

And on the hosts file in Omnios:
10.10.10.20 windows PC host name

Is the above incorrect, it is strange that I could access the NAS share from the windows PC for about a day, then it stopped working.

I will disable jumbo frame on both the windows PC and the Omnios NAS and try and see if it makes a difference.

Also, interrupt throtteling is disabled on the windows PC.

@ sniper, I can ssh into the NAS, however, I am not sure on how to update the driver on Omnios.

  • #15

Joined
Aug 24, 2005
Messages
1,884


You’re trying to connect BOTH 10g ports of each host to each other? You can’t really do that without some kind of link aggregation and even then it won’t be much use for file transfers.

Just use one connection from each nic, each side has to be in the same subnet (not your main one), so windows PC 10.10.10.10 (with mask of 255.255.255.0) and NAS ip of 10.10.10.20 (with mask of 255.255.255.0)

  • #16

Joined
Jul 15, 2017
Messages
174


Sorry, forgot to say, that just one port of each NIC is connected currently the 10.10.10.10 and the 10.10.10.20 the other two ports are spares ready for when I get the backup server built.

  • #17

_Gea

Supreme [H]ardness

Joined
Dec 5, 2010
Messages
4,100


Do what Eickst suggests (connect both nics with an ip from same subnet). If you want to connect the second port from each to a backup server later, add ips then for them from a different subnet.

about updates on OmniOS
OmniOS stable is a freeze of Illumos development with a dedicated repository per release where only critical bugs are fixed. As the ixgbe driver is maintained at Illumos (the common development project for Nexenta, OmniOS, OpenIndiana, SmartOS and others) only on critical fixes a «pkg update» gives a new driver release. On non critical updates, the next stable (every 6 months) or a switch to the bloody release (newest beta) gives new features.

For the current ixgbe driver I only know of one bug (SFP+ problems with some fiber tranceivers). For open issues, see https://www.illumos.org/issues

  • #18

Joined
Jul 15, 2017
Messages
174


Thank you guys for all your help, I will do what you suggest and report back.

EDIT:
Looks like disabeling the static IP’s on the none used second ports of both NIC’s made it work, now I can access the NAS from the windows PC, however the file transfere speed is still slower than 1g.

I can usually transere file between 60 and 70 mbps on the 1g and now it is between 50 or 60 mbps on the 10g nics, in order to confirm that I am on the 10g network, I have disabeled the 1g nic.

Hopefully, this will only be a setting that could make the X540’s work.

Last edited: May 29, 2018

  • #19

/dev/null

Joined
Mar 31, 2001
Messages
15,190


So, should the static IP of the dual port NIC be like:
10.10.10.10
10.10.11.15

and the second dual port NIC be:
10.10.12.20
10.10.13.25

Will the above be better, is that what you mean by two NIC’s not in the same subnet?

This is what I have in the hosts file of the windows machine:
10.10.10.10 NAS

And on the hosts file in Omnios:
10.10.10.20 windows PC host name

Is the above incorrect, it is strange that I could access the NAS share from the windows PC for about a day, then it stopped working.

I will disable jumbo frame on both the windows PC and the Omnios NAS and try and see if it makes a difference.

Also, interrupt throtteling is disabled on the windows PC.

@ sniper, I can ssh into the NAS, however, I am not sure on how to update the driver on Omnios.

If your netmask on both is a /23, then sure…

  • #20

Joined
Aug 24, 2005
Messages
1,884


Thank you guys for all your help, I will do what you suggest and report back.

EDIT:
Looks like disabeling the static IP’s on the none used second ports of both NIC’s made it work, now I can access the NAS from the windows PC, however the file transfere speed is still slower than 1g.

I can usually transere file between 60 and 70 mbps on the 1g and now it is between 50 or 60 mbps on the 10g nics, in order to confirm that I am on the 10g network, I have disabeled the 1g nic.

Hopefully, this will only be a setting that could make the X540’s work.

Can you run iperf across the interfaces to see what the actual network transfer rate is and take file transfers out of it?

Also, there are tons of guides out there for tuning adapters for performance

  • #21

Joined
Nov 16, 2009
Messages
2,224


I think I see 2 problems causing your issue.
1- You have multiple entries for your NAS in the host file, including 1 that matches your main adapters subnet. If windows grabs that IP during a lookup, it will ALWAYS use the wrong adapter. Only have a single entry pointing at the static IP of the NAS’s 10g LAN.
2- The 10g adapters are on different subnets. With that setup, windows will NEVER use your 10g adapter to route traffic to your NAS. If your 10g adapter is configured as 10.0.0.1/24, windows will ONLY use that adapter to route traffic to IPs in the range 10.0.0.1 — 10.0.0.255. If your NAS is configured as 10.0.1.5/24, that subnet will not match your 10g adapter in windows, so it will use the default route (1g adapter). This is most likely the cause of your issue.

Here is what I would do to try and diagnose.

— Disable the second port on the X2 that you are not using on both client/server, so only 1 is enabled
— Use SFP cable to connect client/server on the enabled 10g port
— Set static IP for both client/server in the same subnet (10.0.0.1 and 10.0.0.2 or something similar), but a different subnet than your main 1g adapter.
—- On that adapter, make sure GW/DNS are not configure, just IP and subnet (/24 or smaller). Can enable jumbo frames on both sides as well.
— Update your host file so the only entry for your NAS server is the static IP on the 10g lan

This will work to force all traffic to the NAS over the 10g LAN, and all other traffic will go out default gateway configured on the other adapter. Do a file transfer with Resource Monitor open to confirm the 10g adapter is being used. If you are still getting a 1g limit, it sounds like a config setting on the NAS side. I’m not familiar with that product, so can’t help you there.

Last edited: May 29, 2018

  • #22

Richneerd

Joined
Jan 17, 2010
Messages
641


Chiming in here, I recommend you do iperf test.

I have a 10GB lab at work.

10GBE Workstation
10GBE Switch
10GBE QNAP NAS

9.54GB/s

It’s lovely!

  • #23

Joined
Oct 23, 2007
Messages
2,214


Last edited: May 29, 2018

  • #24

_Gea

Supreme [H]ardness

Joined
Dec 5, 2010
Messages
4,100


That depends whether you look at

— sequential performance
(one user writes/reads a large file from an empty array)

In such a case a single harddisk gives you around 200 MB/s (twice of 1G).
An array of 4 disks in Raid-0 is capable to give full 10G performance

see chapter 2.3 in http://napp-it.org/doc/downloads/optane_slog_pool_performane.pdf

— random performance
(small files, many users or high fragmentation)
This is where disks are bad. Expect around 40-80 MB/s from a disk on average random load,
With many very small files and multi-user access this can go down to a few MB/s

This is where SSDs or NVMe are much better and Intel Optane is the best. A system like ZFS and enough RAM for read/write caching can help

See chapter 2.3 http://napp-it.org/doc/downloads/optane_slog_pool_performane.pdf
and compare allcache vs nochache for effects of RAM (randomrw 1.2 MB/s vs 271 MB/s on a 4 disk raid-0 array)

sync write performance
This is a very special workload when you need a crash save write behaviour.
With a disk based pool without Slog you get a few MB/s, an SSD pool can go up to a few hundred MB/s with the best SSDs.
But as sync write performance is mainly limited by an Slog (in case of ZFS), you can get 10G sync write performance if you use Intel Optane or add an Intel Optane 800P/900P as Slog to a larger SSD or even disk pool, you can get 10G
sync write performance sequentially (400MB/s-1000 MB/s).
https://forums.servethehome.com/ind…s-pools-and-slog-on-solaris-and-omnios.19810/

Last edited: May 30, 2018

  • #25

Joined
Dec 16, 2006
Messages
500


Seems everyone forgot that you can only go as fast as the HD can read and write. Is your NAS and PC have SSDs? And if so, are you copying between the SSDs? Regular spinning disks will never be able to go over what a 1gb connection can do. You need SSDs for that.

How is that? My WD Gold 10TB is hitting 230-260 MB/S during backups and its the same read/write. I am sure there are faster spinners than this but maybe not by much.

  • #26

Joined
Jul 15, 2017
Messages
174


I just need to figure out how to enable iperf via Napp-it, is this already included with omnios or Napp-it or do I need to download and install?

  • #27

_Gea

Supreme [H]ardness

Joined
Dec 5, 2010
Messages
4,100


Iperf3 is included in napp-it (server and client)

iperf3 server, see menu Services > iperf server
iperf3 client, see menu System > Network Eth > iperf client

If you use OmniOS as server, start iperf3 client on another machine
otherwise start server there and use client on OmniOS.

  • #28

Joined
Jul 15, 2017
Messages
174


How do I enable the server it says that it is disabeled?

Iperf3 Server

If you want to run a network bench, start Iperf3 server on one machine and then start client in
System > Network Eth on the other client or appliance to run a benchmark.

Current state Iperf3 server: disabled

EDIT:
With windows firewall off:
Running the NAS on omnios as a server and windows (powershell) as a client, I get the below:
./iperf3.exe -c 10.10.10.10
iperf3: error — unable to connect to server: Connection refused

and the other way around, the NAS as a client and the windows machine as the server it just keeps running without results.

Last edited: May 30, 2018

  • #29

Joined
Jul 15, 2017
Messages
174


I think I see 2 problems causing your issue.
1- You have multiple entries for your NAS in the host file, including 1 that matches your main adapters subnet. If windows grabs that IP during a lookup, it will ALWAYS use the wrong adapter. Only have a single entry pointing at the static IP of the NAS’s 10g LAN.
2- The 10g adapters are on different subnets. With that setup, windows will NEVER use your 10g adapter to route traffic to your NAS. If your 10g adapter is configured as 10.0.0.1/24, windows will ONLY use that adapter to route traffic to IPs in the range 10.0.0.1 — 10.0.0.255. If your NAS is configured as 10.0.1.5/24, that subnet will not match your 10g adapter in windows, so it will use the default route (1g adapter). This is most likely the cause of your issue.

Here is what I would do to try and diagnose.

— Disable the second port on the X2 that you are not using on both client/server, so only 1 is enabled
— Use SFP cable to connect client/server on the enabled 10g port
— Set static IP for both client/server in the same subnet (10.0.0.1 and 10.0.0.2 or something similar), but a different subnet than your main 1g adapter.
—- On that adapter, make sure GW/DNS are not configure, just IP and subnet (/24 or smaller). Can enable jumbo frames on both sides as well.
— Update your host file so the only entry for your NAS server is the static IP on the 10g lan

This will work to force all traffic to the NAS over the 10g LAN, and all other traffic will go out default gateway configured on the other adapter. Do a file transfer with Resource Monitor open to confirm the 10g adapter is being used. If you are still getting a 1g limit, it sounds like a config setting on the NAS side. I’m not familiar with that product, so can’t help you there.

I have go the settings as per above, the windows hosts file entry is the NAS 10g only entry, what should the NAS side hosts file on Omnios be, I have got one entry on there for the 10g windows IP and hostname, is this correct?

  • #30

_Gea

Supreme [H]ardness

Joined
Dec 5, 2010
Messages
4,100


Napp-it comes with iperf3 binaries. They are installed if /usr/bin/iperf3 is missing
Depending on OS and setup it may be that there is a different version installed or libs are missing.

execute then (console or napp-it cmd field) the following command
rm /usr/bin/iperf3

If you then try to start iperf server, it will re-install the included iperf

about hosts file
You do not need to add anything if you use ip to connect
The host file is there to use/translate hostnames to ip in case of a missing DNS server or DNS entry.

Wrong entries may result in problems so for first tests do not add anything and use ip adresses

  • #31

Joined
Oct 23, 2007
Messages
2,214


How is that? My WD Gold 10TB is hitting 230-260 MB/S during backups and its the same read/write. I am sure there are faster spinners than this but maybe not by much.

Umm, mb doesn’t equal GB. Read the link I posted.

  • #32

Joined
Jul 15, 2017
Messages
174


The below is what I’m getting after executing ‘rm /usr/bin/iperf3’, this is with the windows firewall off:

Windows PowerShell
Copyright (C) Microsoft Corporation. All rights reserved.

PS C:UsersN> ./iperf3.exe -c 10.10.10.10
Connecting to host 10.10.10.10, port 5201
[ 4] local 10.10.10.20 port 62655 connected to 10.10.10.10 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 360 MBytes 3.02 Gbits/sec
[ 4] 1.00-2.00 sec 324 MBytes 2.72 Gbits/sec
[ 4] 2.00-3.00 sec 384 MBytes 3.22 Gbits/sec
[ 4] 3.00-4.00 sec 364 MBytes 3.06 Gbits/sec
[ 4] 4.00-5.00 sec 386 MBytes 3.23 Gbits/sec
[ 4] 5.00-6.00 sec 388 MBytes 3.24 Gbits/sec
[ 4] 6.00-7.00 sec 207 MBytes 1.74 Gbits/sec
[ 4] 7.00-8.00 sec 226 MBytes 1.90 Gbits/sec
[ 4] 8.00-9.00 sec 268 MBytes 2.26 Gbits/sec
[ 4] 9.00-10.00 sec 374 MBytes 3.14 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-10.00 sec 3.21 GBytes 2.75 Gbits/sec sender
[ 4] 0.00-10.00 sec 3.21 GBytes 2.75 Gbits/sec receiver

iperf Done.

When using the NAS as the client and the windows machine as the server, the test runs for longer than 10 s, whitout results with a connection timed out message on Napp-it:

PS C:UsersN> ./iperf3.exe -s
————————————————————
Server listening on 5201
————————————————————

On Napp-it
For results, please wait 10s..
iperf3: error — unable to connect to server: Connection timed out

number of streams: 4

EDIT:

The below with tcp changes to the below:

ndd -set /dev/ip ip_lso_outbound 0
ipadm set-prop -p max_buf=4194304 tcp
ipadm set-prop -p recv_buf=1048576 tcp
ipadm set-prop -p send_buf=1048576 tcp

Windows PowerShell
Copyright (C) Microsoft Corporation. All rights reserved.

PS C:UsersN> ./iperf3.exe -c 10.10.10.10
Connecting to host 10.10.10.10, port 5201
[ 4] local 10.10.10.20 port 64653 connected to 10.10.10.10 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 355 MBytes 2.98 Gbits/sec
[ 4] 1.00-2.00 sec 404 MBytes 3.39 Gbits/sec
[ 4] 2.00-3.00 sec 400 MBytes 3.36 Gbits/sec
[ 4] 3.00-4.00 sec 360 MBytes 3.02 Gbits/sec
[ 4] 4.00-5.00 sec 198 MBytes 1.66 Gbits/sec
[ 4] 5.00-6.00 sec 349 MBytes 2.93 Gbits/sec
[ 4] 6.00-7.00 sec 401 MBytes 3.36 Gbits/sec
[ 4] 7.00-8.00 sec 314 MBytes 2.63 Gbits/sec
[ 4] 8.00-9.00 sec 383 MBytes 3.21 Gbits/sec
[ 4] 9.00-10.00 sec 398 MBytes 3.34 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-10.00 sec 3.48 GBytes 2.99 Gbits/sec sender
[ 4] 0.00-10.00 sec 3.48 GBytes 2.99 Gbits/sec receiver

iperf Done.
PS C:UsersN> ./iperf3.exe -s
————————————————————
Server listening on 5201
————————————————————

Still getting the below when using the NAS as the client:

For results, please wait 10s..
iperf3: error — unable to connect to server: Connection timed out

number of streams: 4

Last edited: May 31, 2018

  • #33

Joined
Jul 15, 2017
Messages
174


Anyone know how to disable intr_throttling in Omnbios or Napp-it, the below does not seem to work:

ndd -set /dev/ixgbe0 intr_throttling 1

  • #34

_Gea

Supreme [H]ardness

Joined
Dec 5, 2010
Messages
4,100


This is an entry in /kernel/drv/ ixgbe.conf
intr_throttling = 0;

see also napp-it menu System > Tuning

But unlike Windows that is pre-optimized for forground/Desktop performance I would not expect a relevant difference.
On Windows, this setting is important — together with newest drivers ex from Intel to go beyond 300 MB/s

  • #35

Joined
Nov 16, 2009
Messages
2,224


Those stats confirm it’s at least using the 10g adapter now. Are you sure you’re not hitting limitations from the disks in the NAS? Have you checked their disk using during the transfers?

  • #36

Joined
Jul 15, 2017
Messages
174


Thank you _Gea, I will disable intr_throttling and see whether I get any difference, I don’t ubderstand though why I am not able to use the NAS as the client when I use iperf3, it always times out, could this be a permission issue?

Those stats confirm it’s at least using the 10g adapter now. Are you sure you’re not hitting limitations from the disks in the NAS? Have you checked their disk using during the transfers?

How do I check whether the drives are limiting the speed?, I was hoping that I could see more thaty 60 to 70 mbps with the X540 though, I am sure there is something more than just a drive limitation, I will keep tuning and testing hopefully I will come to the bottom of the problem.

EDIT:

At last, I managed to run ipref3 with the NAS as the client, it was not running earlier because when I stopped windows defender, I had only switched the domain side of the firewall off, with it all off, I get the below:

Windows PowerShell
Copyright (C) Microsoft Corporation. All rights reserved.

PS C:UsersN> ./iperf3.exe -s
————————————————————
Server listening on 5201
————————————————————
Accepted connection from 10.10.10.10, port 33711
[ 5] local 10.10.10.20 port 5201 connected to 10.10.10.10 port 59895
[ 7] local 10.10.10.20 port 5201 connected to 10.10.10.10 port 65009
[ 9] local 10.10.10.20 port 5201 connected to 10.10.10.10 port 42019
[ 11] local 10.10.10.20 port 5201 connected to 10.10.10.10 port 51522
[ ID] Interval Transfer Bandwidth
[ 5] 0.00-1.00 sec 50.7 MBytes 425 Mbits/sec
[ 7] 0.00-1.00 sec 49.7 MBytes 417 Mbits/sec
[ 9] 0.00-1.00 sec 50.6 MBytes 424 Mbits/sec
[ 11] 0.00-1.00 sec 51.3 MBytes 430 Mbits/sec
[SUM] 0.00-1.00 sec 202 MBytes 1.70 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 1.00-2.00 sec 60.3 MBytes 506 Mbits/sec
[ 7] 1.00-2.00 sec 60.1 MBytes 504 Mbits/sec
[ 9] 1.00-2.00 sec 55.6 MBytes 466 Mbits/sec
[ 11] 1.00-2.00 sec 69.4 MBytes 582 Mbits/sec
[SUM] 1.00-2.00 sec 245 MBytes 2.06 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 2.00-3.00 sec 62.1 MBytes 521 Mbits/sec
[ 7] 2.00-3.00 sec 60.9 MBytes 511 Mbits/sec
[ 9] 2.00-3.00 sec 57.4 MBytes 481 Mbits/sec
[ 11] 2.00-3.00 sec 63.2 MBytes 530 Mbits/sec
[SUM] 2.00-3.00 sec 244 MBytes 2.04 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 3.00-4.00 sec 58.3 MBytes 489 Mbits/sec
[ 7] 3.00-4.00 sec 55.4 MBytes 465 Mbits/sec
[ 9] 3.00-4.00 sec 53.4 MBytes 448 Mbits/sec
[ 11] 3.00-4.00 sec 55.2 MBytes 463 Mbits/sec
[SUM] 3.00-4.00 sec 222 MBytes 1.86 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 4.00-5.00 sec 60.4 MBytes 507 Mbits/sec
[ 7] 4.00-5.00 sec 58.6 MBytes 491 Mbits/sec
[ 9] 4.00-5.00 sec 57.0 MBytes 478 Mbits/sec
[ 11] 4.00-5.00 sec 57.7 MBytes 484 Mbits/sec
[SUM] 4.00-5.00 sec 234 MBytes 1.96 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 5.00-6.00 sec 50.8 MBytes 426 Mbits/sec
[ 7] 5.00-6.00 sec 51.9 MBytes 436 Mbits/sec
[ 9] 5.00-6.00 sec 48.1 MBytes 404 Mbits/sec
[ 11] 5.00-6.00 sec 55.5 MBytes 466 Mbits/sec
[SUM] 5.00-6.00 sec 206 MBytes 1.73 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 6.00-7.00 sec 53.8 MBytes 451 Mbits/sec
[ 7] 6.00-7.00 sec 52.8 MBytes 442 Mbits/sec
[ 9] 6.00-7.00 sec 50.9 MBytes 427 Mbits/sec
[ 11] 6.00-7.00 sec 52.8 MBytes 442 Mbits/sec
[SUM] 6.00-7.00 sec 210 MBytes 1.76 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 7.00-8.01 sec 49.0 MBytes 408 Mbits/sec
[ 7] 7.00-8.01 sec 48.2 MBytes 401 Mbits/sec
[ 9] 7.00-8.01 sec 47.0 MBytes 391 Mbits/sec
[ 11] 7.00-8.01 sec 53.6 MBytes 446 Mbits/sec
[SUM] 7.00-8.01 sec 198 MBytes 1.65 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 8.01-9.00 sec 53.5 MBytes 452 Mbits/sec
[ 7] 8.01-9.00 sec 49.5 MBytes 419 Mbits/sec
[ 9] 8.01-9.00 sec 47.6 MBytes 403 Mbits/sec
[ 11] 8.01-9.00 sec 53.7 MBytes 455 Mbits/sec
[SUM] 8.01-9.00 sec 204 MBytes 1.73 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 9.00-10.00 sec 54.6 MBytes 458 Mbits/sec
[ 7] 9.00-10.00 sec 53.6 MBytes 449 Mbits/sec
[ 9] 9.00-10.00 sec 53.2 MBytes 447 Mbits/sec
[ 11] 9.00-10.00 sec 58.8 MBytes 494 Mbits/sec
[SUM] 9.00-10.00 sec 220 MBytes 1.85 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ 5] 10.00-10.01 sec 416 KBytes 517 Mbits/sec
[ 7] 10.00-10.01 sec 416 KBytes 518 Mbits/sec
[ 9] 10.00-10.01 sec 416 KBytes 518 Mbits/sec
[ 11] 10.00-10.01 sec 398 KBytes 496 Mbits/sec
[SUM] 10.00-10.01 sec 1.61 MBytes 2.05 Gbits/sec
— — — — — — — — — — — — — — — — — — — — — — — — —
[ ID] Interval Transfer Bandwidth
[ 5] 0.00-10.01 sec 0.00 Bytes 0.00 bits/sec sender
[ 5] 0.00-10.01 sec 554 MBytes 464 Mbits/sec receiver
[ 7] 0.00-10.01 sec 0.00 Bytes 0.00 bits/sec sender
[ 7] 0.00-10.01 sec 541 MBytes 454 Mbits/sec receiver
[ 9] 0.00-10.01 sec 0.00 Bytes 0.00 bits/sec sender
[ 9] 0.00-10.01 sec 521 MBytes 437 Mbits/sec receiver
[ 11] 0.00-10.01 sec 0.00 Bytes 0.00 bits/sec sender
[ 11] 0.00-10.01 sec 572 MBytes 479 Mbits/sec receiver
[SUM] 0.00-10.01 sec 0.00 Bytes 0.00 bits/sec sender
[SUM] 0.00-10.01 sec 2.14 GBytes 1.83 Gbits/sec receiver
————————————————————
Server listening on 5201
————————————————————

Last edited: May 31, 2018

  • #37

Joined
Aug 23, 2013
Messages
415


Umm, mb doesn’t equal GB. Read the link I posted.

Little b vs big B. Bits vs. Bytes. Defacto standard: network speed is measured in b, disk read/write speed (NOT INTERFACE SPEED) is measured in B.

  • #38

Joined
Nov 16, 2009
Messages
2,224


Little b vs big B. Bits vs. Bytes. Defacto standard: network speed is measured in b, disk read/write speed (NOT INTERFACE SPEED) is measured in B.

Forgot to add 1B = 8b

N Bates You will have to check your NAS for some kind of performance monitor and check the activity time % on each of the disks. If they are all at 100% when doing the transfers, assuming the disks are in a raid array, then you’re being limited by disk speed.

  • #39

Joined
Oct 23, 2007
Messages
2,214


Little b vs big B. Bits vs. Bytes. Defacto standard: network speed is measured in b, disk read/write speed (NOT INTERFACE SPEED) is measured in B.

Yawn, I am a Storage admin for a Enterprise company, you don’t need to tell me about disk limitations when I work on EMC storages all day long.

  • #40

Joined
Oct 23, 2007
Messages
2,214


Forgot to add 1B = 8b

N Bates You will have to check your NAS for some kind of performance monitor and check the activity time % on each of the disks. If they are all at 100% when doing the transfers, assuming the disks are in a raid array, then you’re being limited by disk speed.

Glad someone here has a brain! :p

  • [H]ard|Ware

  • SSDs & Data Storage

Понравилась статья? Поделить с друзьями:
  • Iperf3 error unable to connect to server connection refused
  • Iperf3 error received an unknown control message
  • Iperf3 error control socket has closed unexpectedly
  • Ipdl protocol error handler returned error code
  • Ipconfig renew произошла ошибка при обновлении интерфейса