Linux - ServerThis forum is for the discussion of Linux Software used in a server related context.
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.
Distribution: ubuntu/debian server and centos/fedora server
Posts: 41
Rep:
Debian server 8 nic bonding help
good evening
im still pretty new to the linux world and i have setup a debian 8 server in my home.
i have a cisco switch setup with 6 ports with LACP.
i have set up NIC bonding on the debian server using mode 4 to coincide with my cisco switch and for some weird reason when i perform this command
cat /proc/net/bonding/bond0
output
Bonding mode: Load Balancing (round-robin)
MII Status: up MII Polling Interval (ms):100
i know i set this up as mode 4 for lacp
just wondering if anyone else has come across this issue and how they resolved it as im not sure whether it actually using mode 4 or round robin
# eth0 is manually configured, and slave to the "bond0" bonded nic
auto eth0
iface eth0 inet manual
bond-master bond0
bond-primary = eth0
auto eth1
iface eth1 inet manual
bond-master bond 0
Change your setup from "bond 0" to "bond0"
Quote:
#The bonded inteface
auto bond 0
iface bond 0 inet static
address 192.168.0.25
netmask 255.255.255.0
gateway 192.168.0.1
dns-nameservers 8.8.8.8
bond-mode 4
bond-primary eth0
bond-miimon 100
bond-slaves none
Shouldn't bond-slave read as follows:
Code:
bond-slaves eth0 eth1 eth2 eth3 eth4 eth5
Quote:
also seems to be a bit slow for some reason i am connected to the server via ssh and its lagging when copying large dumps of data over to the server
Not sure if you are aware of the fact that any conversation will only use one (1) interface for communications. Doesn't matter how many interfaces you bond together.
Distribution: ubuntu/debian server and centos/fedora server
Posts: 41
Original Poster
Rep:
Quote:
Originally Posted by lazydog
Change your setup from "bond 0" to "bond0"
Shouldn't bond-slave read as follows:
Code:
bond-slaves eth0 eth1 eth2 eth3 eth4 eth5
Not sure if you are aware of the fact that any conversation will only use one (1) interface for communications. Doesn't matter how many interfaces you bond together.
thanks Robert i have made the changes that you have suggested. as for being aware of the information being sent and received from one card i didn't know that.
Not sure if you are aware of the fact that any conversation will only use one (1) interface for communications. Doesn't matter how many interfaces you bond together.
Not sure the above changes fixed OP's issue but just to add more info: "bond-slaves none" is fine because he has "bond-master bond 0" (which should be "bond-master bond0") under the slave interface definitions. I find the configuration more flexible when done like that.
On Debian (and debian based distros), for LACP, /proc/net/bonding/bond0 should display:
Code:
Bonding Mode: IEEE 802.3ad Dynamic link aggregation
...
For reference and if it may help, I generally setup lacp as follows:
On the switch end and for cisco switches in particular for LACP your "channel-group" config line under all your member interfaces should end in "mode active".
Distribution: ubuntu/debian server and centos/fedora server
Posts: 41
Original Poster
Rep:
Quote:
Originally Posted by dgorack
Not sure the above changes fixed OP's issue but just to add more info: "bond-slaves none" is fine because he has "bond-master bond 0" (which should be "bond-master bond0") under the slave interface definitions. I find the configuration more flexible when done like that.
On Debian (and debian based distros), for LACP, /proc/net/bonding/bond0 should display:
Code:
Bonding Mode: IEEE 802.3ad Dynamic link aggregation
...
For reference and if it may help, I generally setup lacp as follows:
On the switch end and for cisco switches in particular for LACP your "channel-group" config line under all your member interfaces should end in "mode active".
This is so weird i have made the changes that you suggested and at the present moment i am copying stuff over to the linux server.
i have logged into my cisco switch to check my configuration to make sure that my channel-group is set to active and it is. so i then did a show ip int br and it says that it is down but i have lights on flashing to indicate that it is communicating and my files are on the server. so strange
mati
This is so weird i have made the changes that you suggested and at the present moment i am copying stuff over to the linux server.
i have logged into my cisco switch to check my configuration to make sure that my channel-group is set to active and it is. so i then did a show ip int br and it says that it is down but i have lights on flashing to indicate that it is communicating and my files are on the server. so strange
mati
Provide us a little more info and maybe we can help. Let's make sure LACP is properly configured. I assume you're restarting your networking/interfaces after you've made config changes.
On your server
Post your updated configuration (/etc/network/interfaces and anywhere else you set your network/lacp options)
Post the new output of /proc/net/bonding/bond0
On your switch
Post the your relevant interface and channel-group config or if you can link to a pastebin of your sanitized "sh run"
Post the output of "sh lacp neighbor" and "sh lacp internal"
Post the output of "sh int port-channel n" where n is your port-channel number
Distribution: ubuntu/debian server and centos/fedora server
Posts: 41
Original Poster
Rep:
Quote:
Originally Posted by dgorack
Provide us a little more info and maybe we can help. Let's make sure LACP is properly configured. I assume you're restarting your networking/interfaces after you've made config changes.
On your server
Post your updated configuration (/etc/network/interfaces and anywhere else you set your network/lacp options)
Post the new output of /proc/net/bonding/bond0
On your switch
Post the your relevant interface and channel-group config or if you can link to a pastebin of your sanitized "sh run"
Post the output of "sh lacp neighbor" and "sh lacp internal"
Post the output of "sh int port-channel n" where n is your port-channel number
please see the images attached i grabbed as much information as i could.
please see the images attached i grabbed as much information as i could.
It would really be helpful if you'd post the output instead of attaching screen grabs. I think it's just good etiquette and it also my life easier in trying to help you (copy/paste).
Your output of /proc/net/bonding/bond0 shows link failures which is probably why it's falling back to round-robin. Like @lazydog said, there are a number of way to do this but I can show you the way I use. Your switch output also shows the each port in your channel-group in state "indep" which mean it failed to bond lacp.
First of all let's clean up your /etc/network/interfaces file.
Check for typos as I'm typing this manually, you know, because you weren't bothered to copy/paste here.
Code:
...
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet manual
bond-master bond0
auto eth1
iface eth1 inet manual
bond-master bond0
auto eth2
iface eth2 inet manual
bond-master bond0
auto eth3
iface eth3 inet manual
bond-master bond0
auto eth4
iface eth4 inet manual
bond-master bond0
auto eth5
iface eth5 inet manual
bond-master bond0
auto bond0
iface bond0 inet static
address 192.168.0.25
netmask 255.255.255.0
broadcast 192.168.0.255
gateway 192.168.0.1
dns-nameservers 8.8.8.8 8.8.4.4
bond-mode 802.3ad
bond-miimon 100
bond-lacp-rate 1
bond-slaves none
Also make sure the bonding module is loaded
Code:
lsmod | grep bond
I can't help you with your switch because you didn't post your running-config, post that in a pastebin or in code blocks, no screen grabs please.
Distribution: ubuntu/debian server and centos/fedora server
Posts: 41
Original Poster
Rep:
Quote:
Originally Posted by dgorack
It would really be helpful if you'd post the output instead of attaching screen grabs. I think it's just good etiquette and it also my life easier in trying to help you (copy/paste).
Your output of /proc/net/bonding/bond0 shows link failures which is probably why it's falling back to round-robin. Like @lazydog said, there are a number of way to do this but I can show you the way I use. Your switch output also shows the each port in your channel-group in state "indep" which mean it failed to bond lacp.
First of all let's clean up your /etc/network/interfaces file.
Check for typos as I'm typing this manually, you know, because you weren't bothered to copy/paste here.
Code:
...
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet manual
bond-master bond0
auto eth1
iface eth1 inet manual
bond-master bond0
auto eth2
iface eth2 inet manual
bond-master bond0
auto eth3
iface eth3 inet manual
bond-master bond0
auto eth4
iface eth4 inet manual
bond-master bond0
auto eth5
iface eth5 inet manual
bond-master bond0
auto bond0
iface bond0 inet static
address 192.168.0.25
netmask 255.255.255.0
broadcast 192.168.0.255
gateway 192.168.0.1
dns-nameservers 8.8.8.8 8.8.4.4
bond-mode 802.3ad
bond-miimon 100
bond-lacp-rate 1
bond-slaves none
Also make sure the bonding module is loaded
Code:
lsmod | grep bond
I can't help you with your switch because you didn't post your running-config, post that in a pastebin or in code blocks, no screen grabs please.
Sorry i wasnt sure how to copy from putty. I hand typed the previous post i thought it would of been easier. Sorry again i shall copy and paste everything in now.
Just had a look at this as i noticed this on startup
Code:
root@server:/home/mati# systemctl status systemd-modules-load
● systemd-modules-load.service - Load Kernel Modules
Loaded: loaded (/lib/systemd/system/systemd-modules-load.service; static)
Active: failed (Result: exit-code) since Thu 2016-05-12 09:51:14 BST; 13min ago
Docs: man:systemd-modules-load.service(8)
man:modules-load.d(5)
Process: 173 ExecStart=/lib/systemd/systemd-modules-load (code=exited, status=1/FAILURE)
Main PID: 173 (code=exited, status=1/FAILURE)
May 12 09:51:14 server systemd-modules-load[173]: Inserted module 'ppdev'
May 12 09:51:14 server systemd-modules-load[173]: Inserted module 'parport_pc'
May 12 09:51:14 server systemd-modules-load[173]: Inserted module 'fuse'
May 12 09:51:14 server systemd-modules-load[173]: Inserted module 'loop'
May 12 09:51:14 server systemd-modules-load[173]: Failed to find module 'rtc'
May 12 09:51:14 server systemd-modules-load[173]: Inserted module 'bonding'
May 12 09:51:14 server systemd-modules-load[173]: Inserted module 'mii'
May 12 09:51:14 server systemd[1]: systemd-modules-load.service: main proce...RE
May 12 09:51:14 server systemd[1]: Failed to start Load Kernel Modules.
May 12 09:51:14 server systemd[1]: Unit systemd-modules-load.service entere...e.
Hint: Some lines were ellipsized, use -l to show in full.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
