LinuxQuestions.org
Visit Jeremy's Blog.
Home Forums Tutorials Articles Register
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Networking
User Name
Password
Linux - Networking This forum is for any issue related to networks or networking.
Routing, network cards, OSI, etc. Anything is fair game.

Notices


Reply
  Search this Thread
Old 10-17-2004, 02:17 PM   #1
synaptical
Senior Member
 
Registered: Jun 2003
Distribution: Mint 13/15, CentOS 6.4
Posts: 2,020

Rep: Reputation: 48
network performance question


i *should* be getting ~100M/s transfer rates (minus overhead) between computers on a LAN with 10/100 nics and router, correct? because since installing NFS, i only seem to be getting <10 (4 M/s going to windows), and i'm wondering what's limiting me. how can i diagnose this? thx.
 
Old 10-17-2004, 03:38 PM   #2
Mara
Moderator
 
Registered: Feb 2002
Location: Grenoble
Distribution: Debian
Posts: 9,696

Rep: Reputation: 232Reputation: 232Reputation: 232
1. ifconfig - are there errors?
2. any sniffer - you can see the connections etc and see what's taking the bandwith
 
Old 10-17-2004, 04:45 PM   #3
synaptical
Senior Member
 
Registered: Jun 2003
Distribution: Mint 13/15, CentOS 6.4
Posts: 2,020

Original Poster
Rep: Reputation: 48
no errors, no collisions, everything there seems to look okay:

Code:
[4] root:~ # /sbin/ifconfig
eth0      Link encap:Ethernet  HWaddr CE:00:6E:96:2B:9F  
          inet addr:192.168.xxx.xxx  Bcast:192.168.0.255  Mask:255.255.255.0
          UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:3246314 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2753904 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:3501427493 (3339.2 Mb)  TX bytes:3711941911 (3539.9 Mb)
          Interrupt:11 Base address:0xb000 
 
lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:287 errors:0 dropped:0 overruns:0 frame:0
          TX packets:287 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:25384 (24.7 Kb)  TX bytes:25384 (24.7 Kb)

~~~~~~~~~~~~~~~~~~~~~
root@ripper:~ # ifconfig
eth0      Link encap:Ethernet  HWaddr 00:09:5B:09:DD:05  
          inet addr:192.168.xxx.xxx  Bcast:192.168.0.255  Mask:255.255.255.0
          UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2054023 errors:0 dropped:0 overruns:0 frame:0
          TX packets:5653831 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:1703330043 (1624.4 Mb)  TX bytes:3714711912 (3542.6 Mb)
          Interrupt:10 Base address:0x9000 

lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:425459 errors:0 dropped:0 overruns:0 frame:0
          TX packets:425459 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:383857884 (366.0 Mb)  TX bytes:383857884 (366.0 Mb)

~~~~~~~~~~~~~~~~~~~~
jeff@moe:~ $ /sbin/ifconfig
eth0      Link encap:Ethernet  HWaddr 00:10:4B:33:7F:EB  
          inet addr:192.168.xxx.xxx  Bcast:192.168.0.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2871222 errors:0 dropped:0 overruns:0 frame:0
          TX packets:806439 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:2552570246 (2.3 GiB)  TX bytes:677167800 (645.7 MiB)
          Interrupt:10 Base address:0x1000 

lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:2095 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2095 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:105146 (102.6 KiB)  TX bytes:105146 (102.6 KiB)
my sense is it's some misconfiguration rather than something taking the bandwidth, though i could be wrong.

i just installed ethereal, so i'll check that out. i've never used it before, so any tips appreciated. thx.

~~~~~~~~~~~~~~~
>edit: okay, i think i mostly got it. running ethereal showed UDP protocol hitting 100%, and i found an LQ thread recommending putting tcp in the NFS mounting options in fstab. now i'm getting 11.5 MB/sec transfers between my two NFS computers, and about 6MB/s over ftp to my other linux machine. it starts out around 9MB/s for about 5-10 seconds and then starts degrading, so i think there's some disk overhead there or something (5400rpm drives). windows is still a little slow, but i think i have a cat5 cable to that machine. (or maybe windows just sucks. )

Last edited by synaptical; 10-17-2004 at 08:30 PM.
 
Old 10-18-2004, 12:32 AM   #4
mago
Member
 
Registered: Apr 2004
Location: Costa Rica
Distribution: slack current with 2.6.16.18 (still off the hook)
Posts: 284

Rep: Reputation: 33
You said:
Quote:
because since installing NFS, i only seem to be getting <10 (4 M/s going to windows), and i'm wondering what's limiting me. how can i diagnose this? thx.
Ok how exactly does NFS and windows has to do on the same network?, to comunicate with a windows machine use samba.

My personal experience shows that you will never get the 100Mbps, the most I've seen is about 64Mbps and that is from a straight NFS to NFS connection.

Besides have in mind that windows mesures the speed in MB/s and linux does it in Mb/s, so 4MB/s is actuali 32Mb/s on linux standard wich is 32% of the theoretical speed. One more thing are the NICs set up to FULL DUPLEX?


I hope this will help you a bit.
 
Old 10-18-2004, 10:03 AM   #5
synaptical
Senior Member
 
Registered: Jun 2003
Distribution: Mint 13/15, CentOS 6.4
Posts: 2,020

Original Poster
Rep: Reputation: 48
Quote:
Originally posted by mago
Ok how exactly does NFS and windows has to do on the same network?, to comunicate with a windows machine use samba.
i don't know much about nfs, i thought some aspect of the protocol (e.g., UDP) might be using up bandwidth, and if windows used that more than linux to do a transfer, that might account for it. (?) (i.e., even though windows is not using the NFS, is what i mean.)

Quote:
My personal experience shows that you will never get the 100Mbps, the most I've seen is about 64Mbps and that is from a straight NFS to NFS connection.
11.5 MB/s is pretty close to the theoretical max, isn't it? (11.5MB * 8 = 92Mbits http://www.matisse.net/mcgi-bin/bits.cgi).

Quote:
Besides have in mind that windows mesures the speed in MB/s and linux does it in Mb/s, so 4MB/s is actuali 32Mb/s on linux standard wich is 32% of the theoretical speed.


i was measuring from linux, though, with "traffic," an simpler iptraf-like program.

Quote:
One more thing are the NICs set up to FULL DUPLEX?
i can't seem to check the one on my main computer, i guess b/c of the forcedeth driver. i'm assuming with 11MB it is. the other computer on NFS is definitely full duplex, and so is the one with the 9800KB/s (before it degrades).

Code:
[1] root:~ # ethtool eth0
Settings for eth0:
        Supports Wake-on: g
        Wake-on: d
        Link detected: yes

[3] root:~ # ethtool -s eth0 speed 100 duplex full
Cannot get current device settings: Operation not supported
  not setting speed
  not setting duplex

~~~~~~~~~~~~~~~~~~~~~
root@ripper:~ # /usr/local/sbin/ethtool eth0
Settings for eth0:
        Supported ports: [ TP MII FIBRE ]
        Supported link modes:   10baseT/Half 10baseT/Full 
                                100baseT/Half 100baseT/Full 
        Supports auto-negotiation: Yes
        Advertised link modes:  10baseT/Half 10baseT/Full 
                                100baseT/Half 100baseT/Full 
        Advertised auto-negotiation: Yes
        Speed: 100Mb/s
        Duplex: Full
        Port: Twisted Pair
        PHYAD: 1
        Transceiver: internal
        Auto-negotiation: on
        Supports Wake-on: pumbags
        Wake-on: ubg
        SecureOn password: 00:00:00:00:00:00
        Current message level: 0x000040c5 (16581)
        Link detected: yes

~~~~~~~~~~~~~~
root@moe:~ # mii-tool eth0
eth0: negotiated 100baseTx-FD, link ok
Quote:
I hope this will help you a bit.
thanks! no pun intended, i hope!
 
  


Reply



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
horiffic network performance carboncopy Slackware 32 09-22-2005 04:58 AM
Pathetic network performance costyn Linux - Hardware 1 08-22-2004 10:36 AM
network performance tuning Skunk_Face Linux - Networking 1 04-16-2004 05:50 AM
Poor Network Performance ferrino Linux - Networking 3 11-02-2003 04:22 PM
Poor network performance RedH13 Linux - Networking 0 10-07-2003 10:11 PM

LinuxQuestions.org > Forums > Linux Forums > Linux - Networking

All times are GMT -5. The time now is 05:22 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Open Source Consulting | Domain Registration