Linux - NetworkingThis forum is for any issue related to networks or networking.
Routing, network cards, OSI, etc. Anything is fair game.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I have two computers, identical, with 10GB nics in them. I have them -directly- connected and addressed. Driver is installed. Ethtool says it is in 10000Mb/s mode full duplex.
But doing any transfers, I only get 129MB per second.
How would you do further troubleshooting on this?
Ethtool:
Code:
Settings for eth2:
Supported ports: [ TP ]
Supported link modes: 100baseT/Full
1000baseT/Full
10000baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 100baseT/Full
1000baseT/Full
10000baseT/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Speed: 10000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 0
Transceiver: external
Auto-negotiation: on
MDI-X: Unknown
Supports Wake-on: d
Wake-on: d
Current message level: 0x00000007 (7)
Link detected: yes
rsync:
Code:
sending incremental file list
testing.file
5536.35M 75% 128.77MB/s 0:02:36
dd write test:
Code:
dd if=/dev/zero of=/home/caldera/testing.file bs=50k count=50000
50000+0 records in
50000+0 records out
2560000000 bytes (2.6 GB) copied, 20.5084 s, 125 MB/s
hdparm:
Code:
hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads: 13676 MB in 2.00 seconds = 6842.34 MB/sec
Timing buffered disk reads: 520 MB in 3.00 seconds = 173.33 MB/sec
If they're directly connected, are you using a crossover cable? What's the distance between them? Is the cable rated for 10G? I'd like to be in optic fibre at that speed.
It shouldn't make any difference patch or crossover. All modern nic's correct for cable. If ethtool reports it is in 10G then it is. It is possible that a bad cable or nic or connector could cause a bad quality of service issue.
Cat 5 won't do usually.
It would have to be a friggin monster of a system with a raid ssd to offer 10G speeds I'd think. You'd have to have an equal system on other end also.
One thing that may help is top. If you find cpu usage is way high then see if checksum offload is enabled on both nic's.
We also may have a bit and byte issue here.
I'd create a contiguous binary file and transfer that for a test.
You could possibly make a loop back cable and test it on a single system but again you'd have to have a double huge system.
I can't get the drive to write faster than 140MB/s, and others have suggested that 10G is best used between a computer and a SAN or other array capable of writing as fast as the card.
I don't get much more than about half or less of a 1G nic in general systems.
In some weak systems using fast nic's also eat up cpu time. There is always a lot of chatter on nics and the cpu may need to work more on the now fast nic.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.