Gigabit LAN connection with an average speed of 2,5 MB p/s

Got problems with your B2 or B3? Share and get helped!
PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Gigabit LAN connection with an average speed of 2,5 MB p/s

Post by PaulW » 22 Dec 2008, 08:29

My gigabit connection (LAN) does not work as it should be.
Hope someone can help me.

On the LAN connector on the Bubba there is just one led blinking i.s.o. two leds.
The average speed of the gigabit connection is approx 2,5 MB p/s. This is measured with scp.

If I transfer data from PC1 (with gigabit network card) to PC2 (with 10/100 Mb p/s network card) I will get a average speed of 10MB p/s. This is just as expected due to the 10/100 mb network card available in PC2.

When transferring data from PC1 to PC3 (with gigabit networkcard) then I will get a avarage speed of 55MB p/s. This looks a little more as a gigabit connection.

When connecting PC4 (Also a gigabit bit network card) directly to the bubba I will still get the average speed of 2,5 MB p/s.

Therefor the problem must be caused on the bubba itself.

I logged in on the bubba and enter the command

Code: Select all

sudo ifconfig
This gives the information:

Code: Select all

eth0      Link encap:Ethernet  HWaddr 00:22:02:00:01:2E  
          inet addr:213.51.224.50  Bcast:255.255.255.255  Mask:255.255.254.0
          UP BROADCAST RUNNING  MTU:1500  Metric:1
          RX packets:1939291 errors:0 dropped:0 overruns:0 frame:120
          TX packets:640010 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:1439497306 (1.3 GiB)  TX bytes:61564575 (58.7 MiB)
          Base address:0xe000 

eth1      Link encap:Ethernet  HWaddr 00:22:02:00:01:2F  
          inet addr:192.168.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
          inet6 addr: fe80::222:2ff:fe00:12f/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:805800 errors:0 dropped:0 overruns:0 frame:0
          TX packets:983939 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:463846586 (442.3 MiB)  TX bytes:1421172299 (1.3 GiB)

lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:9691 errors:0 dropped:0 overruns:0 frame:0
          TX packets:9691 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:1692029 (1.6 MiB)  TX bytes:1692029 (1.6 MiB)
The command (eth1 is my LAN connection)

Code: Select all

sudo ethtool eth1
Gives th information:

Code: Select all

Settings for eth1:
	Supported ports: [ MII ]
	Supported link modes:   10baseT/Half 10baseT/Full 
	                        100baseT/Half 100baseT/Full 
	                        1000baseT/Full 
	Supports auto-negotiation: Yes
	Advertised link modes:  10baseT/Half 10baseT/Full 
	                        100baseT/Half 100baseT/Full 
	                        1000baseT/Full 
	Advertised auto-negotiation: Yes
	Speed: 1000Mb/s
	Duplex: Full
	Port: MII
	PHYAD: 0
	Transceiver: external
	Auto-negotiation: on
	Current message level: 0x0000003f (63)
	Link detected: yes
So we can see that eth1 is a gigabit connection and that we do not get the expected speed from a gigabit LAN.
Enabling or disabling Jumbo frames does not make any differences.

Hope someone can help.............. :cry:

squadra
Posts: 96
Joined: 19 Sep 2008, 09:02

Post by squadra » 28 Dec 2008, 20:58

Did you also make the software updates (see anouncements and two fixes) ? If i transfer wired i get 80mbs.

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 29 Dec 2008, 18:39

Yes, the update and the fixes are installed but the current speed is still 2,4 MB/s

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 18 Jan 2009, 15:40

In the meantime I went back to the local distributor.
He reproduced the problem. With the old bubba he got 4 MB/s and with a new one 27 MB/s.
Coming back home I installed the brandnew bubba (which he had tested) in my network.
Guess what............

Still 2,5 MB/s :shock:

I took a PC with a gigabit network card and connect it directly to the bubba. (On the LAN)
Yes, it is a little bit faster. 3,3 MB/s (With MTU value 9000 on eth1)

I really don't know what to do to get it solved and where to look for the problem.

johannes
Posts: 1470
Joined: 31 Dec 2006, 07:12
Location: Sweden
Contact:

Post by johannes » 20 Jan 2009, 15:09

This does sound strange. Can it be bad cables that you used also at the distributor?

If you can't find it, please return it to us and we'll have a look. In that case, please email support@excito.com and we'll let you know the procedure.
/Johannes (Excito co-founder a long time ago, but now I'm just Johannes)

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 22 Jan 2009, 07:19

At the local distributor we've used different cables as I did at home.
Yesterday I took also a brand new cat6 patch cable and I have connected one PC with a Gb LAN directly to the Bubba2.
This is still 2,5 MB.

I have also disabled the auto-Negotiation because I know that everything is Gb which is attached to the Bubba2. However the speed was the same.

This evening I will do some experiments with different network cards to make sure that we are not facing a compatibility issue.

If someone else has some other suggestions please let me know.

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 22 Jan 2009, 17:26

I have connected a laptop directly to the bubba2 to see what kind of speed I got.
The network card in my laptop is 100 mb/s.

The speed using the command:

Code: Select all

scp <file> <username>@<ip bubba2>:<path + directory>/. 

shows again a speed of 2,5 MB/s.

Even for a 100 Mb/s much too slow. :(
Last edited by PaulW on 23 Jan 2009, 17:03, edited 1 time in total.

GotenXiao
Posts: 2
Joined: 07 Jan 2009, 06:47

Post by GotenXiao » 23 Jan 2009, 08:10

Confirming.

In the following, casull is my server, nightstalker is my desktop, 45acp is the Bubba Two. testfile is a 1GB file generated using the following command:

Code: Select all

dd if=/dev/zero of=testfile bs=1M count=1K

Code: Select all

goten@nightstalker:~$ scp testfile casull:/home/goten/
testfile 100% 1024MB  32.0MB/s   00:32
goten@casull:~$ scp testfile nightstalker:/home/goten
testfile 100% 1024MB  42.7MB/s   00:24
goten@nightstalker:~$ scp testfile 45acp:/home/goten/
testfile 13%  142MB   2.5MB/s   05:56 ETA
goten@casull:~$ scp testfile 45acp:/home/goten
testfile 6%   62MB   2.4MB/s   06:35 ETA
Running an scp from both nightstalker and casull to 45acp at the same time results in a halving of bandwidth. nc suffers similar problems, so it's not the CPU load from running SSH's crypto... *headscratches*

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 24 Jan 2009, 17:06

I found something which might have something to do with it.

When I execute the command (as root):

Code: Select all

apt-get update
I get the following errors in my terminal:

Code: Select all

Get:1 http://ftp.se.debian.org etch Release.gpg [386B]
Get:2 http://update.excito.org marielle Release.gpg [189B]
Get:3 http://update.excito.org upstream_etch Release.gpg [189B]
Hit http://update.excito.org marielle Release
Hit http://ftp.se.debian.org etch Release
Hit http://update.excito.org upstream_etch Release
Err http://update.excito.org marielle Release
  
Get:4 http://update.excito.org marielle Release [2970B]
Err http://update.excito.org upstream_etch Release
  
Ign http://ftp.se.debian.org etch/main Packages/DiffIndex            
Get:5 http://update.excito.org upstream_etch Release [2651B]         
Ign http://update.excito.org marielle Release  
Ign http://ftp.se.debian.org etch/contrib Packages/DiffIndex         
Ign http://ftp.se.debian.org etch/non-free Packages/DiffIndex        
Ign http://update.excito.org upstream_etch Release                   
Hit http://update.excito.org marielle/main Packages/DiffIndex        
Hit http://ftp.se.debian.org etch/main Packages
Hit http://ftp.se.debian.org etch/contrib Packages
Ign http://update.excito.org marielle/main Sources/DiffIndex
Hit http://update.excito.org upstream_etch/main Packages/DiffIndex
Hit http://ftp.se.debian.org etch/non-free Packages
Hit http://update.excito.org marielle/main Sources
Fetched 6000B in 1s (4193B/s)
Reading package lists... Done
W: GPG error: http://update.excito.org marielle Release: Unknown error executing gpgv
W: GPG error: http://update.excito.org upstream_etch Release: Unknown error executing gpgv
W: You may want to run apt-get update to correct these problems
Till now I didn't find a solution for this. (Google shows a lot of this...)
I can imagine that because of this some updates are not or not correctly installed which is causing my speed problem.

peter
Posts: 26
Joined: 07 Oct 2008, 09:35
Location: the Netherlands

Post by peter » 25 Jan 2009, 06:00

Paul, I had a problem with an incomplete Bubba too.
For me the solution was an apt-get dist-upgrade (as root).

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 25 Jan 2009, 07:31

Thx Peter,

I tried to do that but unfortunately this didn't solve the issue neither.

Code: Select all

cp128139-b:/home/paul# apt-get dist-upgrade (as root)
Reading package lists... Done
Building dependency tree... Done
Calculating upgrade... Done
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
I bought an audio and media streamer a few weeks ago.
Yesterday I've connected both streamers to the network.
When I was copying data from an external drive to the Bubba2 I saw that my audio streamer lost its connection with the Bubba2. Seems that the network load was too much.

jonte
Posts: 65
Joined: 05 Nov 2008, 11:52

Post by jonte » 25 Jan 2009, 15:44

PaulW, I am to geting the same speed problems with file transfers with my B2 over Gigabit LAN, with SCP as well as SMB-transfers. I know SMB is slower, but not that slow, and SCP shouldn't be that slow either.

No more than 2,5 megabyte / s. Still trying to figure out the problem.

Just wanted to show you that there are more with the same issues.

//Jonathan

PaulW
Posts: 15
Joined: 15 Oct 2008, 15:27

Post by PaulW » 25 Jan 2009, 17:18

I am glad to know that I am not the only one who is facing this problem.
Hopefully somebody from Excito can reproduce the problem in order to fix this.

@Johannes: Are you able to reproduce this? Else I can ask the local distributor to sent you my old Bubba2 which have been changed for a new one for this reason.

Or do you have any suggestions where we can look for....

Thanks in advance,

Paul

kees
Posts: 16
Joined: 23 Jan 2009, 17:59

slow network speed

Post by kees » 26 Jan 2009, 14:33

I have the same speed problem. FTP tranfers 4.0 Mb/s and SCP, SMB transfers ca. 2.5 Mb/s. This should be greater then 35Mb/s for a Gigabit network.
But also the disk writespeed is far below expectation:

Bubba with WD10EACS (WD greenpower 1Tb SATA):

Code: Select all

root@bubba:/home/kees#  dd if=/dev/zero of=bigfile count=5k bs=100k
5120+0 records in
5120+0 records out
524288000 bytes (524 MB) copied, 32.4735 seconds, 16.1 MB/s

Old celeron with WD3200JB (WD 320Gb PATA ):

Code: Select all

root@ganzen2:~# dd if=/dev/zero of=bigfile count=5k bs=100k
5120+0 records in
5120+0 records out
524288000 bytes transferred in 6.297364 seconds (79.4 MB/s)

load average on both machines less then 0.1


Maybe the slow network tranfer speed is caused by poor disk performance.
question:
what are the results of the "dd"-command on your bubba?

jonte
Posts: 65
Joined: 05 Nov 2008, 11:52

Re: slow network speed

Post by jonte » 26 Jan 2009, 15:36

kees wrote: what are the results of the "dd"-command on your bubba?
Bubba with WDGP500GB

Code: Select all

root@bubba:/home/jonte# dd if=/dev/zero of=bigfile count=5k bs=100k
5120+0 records in
5120+0 records out
524288000 bytes (524 MB) copied, 33.2246 seconds, 15.8 MB/s
Even slower...

Post Reply