Bonding of ethernet cards or second adapter eth1 not up

[jpcozar]

Hello,

We are using last version of fvdw firmware (16.0 version) in our Lacie 5gib Network vs2 and I am missing in the web interface load balance of ethernet adapters (or at least being detected both adapters) that we can find in Lacie firmware.
Is there any way to do it either web interface or command line ?
I've connected via ssh to NAS and this is output of dmesg related to eth:

Code: Select all

root@Sauron:/etc # dmesg |grep eth
[    8.192089] mv643xx_eth: MV-643xx 10/100/1000 ethernet driver version 1.4
[    8.271428] mv643xx_eth_port mv643xx_eth_port.0 eth0: port 0 with MAC address 00:d0:4b:8d:d1:02
[    8.361398] mv643xx_eth_port mv643xx_eth_port.1 eth1: port 0 with MAC address 00:d0:4b:8d:b5:ec
[    8.450079] usbcore: registered new interface driver cdc_ether
[    8.473801] usbcore: registered new interface driver ipheth
[   35.980440] mv643xx_eth_port mv643xx_eth_port.0 eth0: link down
[   35.986546] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[   39.085614] mv643xx_eth_port mv643xx_eth_port.0 eth0: link up, 1000 Mb/s, full duplex, flow control disabled
[   39.095440] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[  323.401316] mv643xx_eth_port mv643xx_eth_port.0 eth0: link down
[  336.355063] mv643xx_eth_port mv643xx_eth_port.0 eth0: link up, 1000 Mb/s, full duplex, flow control disabled


Both devices are detected but just eth0 is up. Output of ifconfig command is:

Code: Select all

eth0      Link encap:Ethernet  HWaddr 00:D0:4B:8D:D1:02 
          inet addr:10.70.8.80  Bcast:10.70.9.255  Mask:255.255.254.0
          inet6 addr: fe80::2d0:4bff:fe8d:d102/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:23609 errors:0 dropped:404 overruns:0 frame:0
          TX packets:7204 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:2491636 (2.3 MiB)  TX bytes:1469578 (1.4 MiB)
          Interrupt:11

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:65536  Metric:1
          RX packets:316 errors:0 dropped:0 overruns:0 frame:0
          TX packets:316 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:27303 (26.6 KiB)  TX bytes:27303 (26.6 KiB)


So just eth0 adapter and loopback are detected.
if I try:

Code: Select all

ifup eth1
ifup: can't open '/etc/network/interfaces': No such file or directory


So there's no an interfaces file where we can add a new adapter. And if I create that file myself
it says

Code: Select all

ifup eth1

/bin/sh: run-parts: not found


it is missing run-parts command.

So first question is... is there any way to get enable and up second ethernet adapter (eth1)? and once it's up... is there any way to do a bonding with both adapters so we can transfer files to 2 Gb instead of 1 Gb ? Right now second adapter is being useless with this firmware.

Thank you in advance!

[Jocko]

Hi jcozar,

Indeed it is a known limitation with our firmware and it is in the todo list since Oct 2013! (But up to now, nobody asked to deepen this point).
Currently, there is no feature available for bonding two Ethernet ports.

Do to do it we need:
- to add the bonding module in our kernel version (but maybe it is already done with the last version, fvdw can confirm this point).
- at different level in our firmware, we have to allow to use another NIC device that eth0 (by default the firmware always uses eth0).
- to change the LAN setup menu, to set eth1 and/or to enable a bond between them.

So do you have a real need to bond these 2 ports ?

Note: to get all NIC (bring up/down) the command is

Code: Select all

ifconfig -a

and then you will see eth1.

[jpcozar]

Thank you for answering,

Well, the only two limitations that I see of this firmware against "official's one" is the bonding/balancing network and iSCSI support.
We used iSCSI to get backups from a Vmware ESXi because old versions of official Lacie firmware wasnt able to use NFS. Nowdays, fwdv firmware and latest Lacie has NFS support so we can use NFS instead of ISCSI.

But yes, it's really necessary for us to use bonding because we can increase bandwith from 1 GB to 2GB when we do daily backups of vmware images (they are about 100 gb).

And it isnt necessary to change dashboard, we can do it via ssh but yes, we need bonding module loaded and permissions to modifiy network interfaces.

if someone else need to configure second network adapter, it can be done via ssh with ifconfig <ethx> instead of ifup as I was trying:

Code: Select all

ifconfig eth1 <ip> netmask <netmask>


Now I only need to get it at startup :-)

I'll be waiting about bonding help

[Jocko]

Hi jpcozar,

Ok, I go to look at this point in next weeks to see if we can add it in version 16.1.

And it isnt necessary to change dashboard, we can do it via ssh but yes, we need bonding module loaded and permissions to modifiy network interfaces.

No jpcozar,

Currently, the firmware allows to set all requires settings via its web-interface, we need also to save this choice and the bonding mode to restore it at next booting. A shell command is also not enough : in some servers conf file the NIC set what NIC having to be used (currently with eth0). So we need also to change that.

So wait few weeks

[Jocko]

Hi jpcozar,

We have implemented the Bonding feature with 5big 2 platform for the next version.
But as there are too many changed files, I can not do easily a patch (or do again the implement job on version 16.0 :disapprove ).

So you need to wait the next release. I think that we will do that at the end of this month. (If we delay the release, I will build a temporary firmware image)

[jpcozar]

Ok, thank you!

Here you have a beta-tester to try bonding feature in next release :-)

[Jocko]

Hi jpcozar,

Yes, we have built a beta firmware image version and currently we test it.

So may be next week, I will send you the image if there is no significant bug.

[jpcozar]

I am glad that you did bonding available in last version of your firmware.
Thank you very much!