I just installed EVE-NG on my DELL server and I can't start any node. I followed all the instructions on the website(checked the naming of the qcow2 files and the folders as well ..actually I did copy paste from the website to be sure I am not making any typo) and couldn't find any similar problems (sorry if there are and overlooked) .
My setup is the following:
Server: Dell Poweredge C1100 with 2 X Intel(R) Xeon(R) CPU X5550 @ 2.67GHz and 48 Gb RAM / VTx enabled
OS : Ubuntu 16.0.4.2 LTS 64bit
Nodes I was trying to start : Cisco XRv 5.2.2 and Juniper junos-vsrx-12.1X47-D20.7-domestic.ova
Now I looked into the logs and found the following after I try to start the nodes:
root@eve-ng:~# cat /opt/unetlab/data/Logs/unl_wrapper.txt
May 21 23:03:32 ERROR: echo 65535 > /sys/class/net/vnet0_1/bridge/group_fwd_mask 2>&1 --- Failed to set group_fwd_mask on bridge (80028).
May 21 23:03:32
May 21 23:03:32 May 21 23:03:32 ERROR: Failed to set group_fwd_mask on bridge (80028).
May 21 23:03:32 May 21 23:03:32 ERROR: Failed to create network (11).
May 21 23:03:33 INFO: tunctl -u unl0 -g root -t vunl0_1_0 2>&1
RTNETLINK answers: Invalid argument
May 21 23:03:33 ERROR: Failed to set MTU 9000 on interface (80085).
May 21 23:03:33 Set 'vunl0_1_0' persistent and owned by uid 32768 gid 0
May 21 23:03:33 May 21 23:03:33 ERROR: Failed to set MTU 9000 on interface (80085).
May 21 23:03:33 May 21 23:03:33 ERROR: Failed to start node (12).
root@eve-ng:~#
The thing that caught my attention was "RTNETLINK answers: Invalid argument" which lead me to the following bug : https://bugs.launchpad.net/ubuntu/+sour ... ug/1399064 which say the following :
A linux bridge always adopts the smallest MTU of the enslaved devices. When no device are enslaved, it defaults to a MTU of 1500 and refuses to use a larger one. This is problematic when using bridges enslaving only virtual NICs (vnetX) like it's common with KVM guests.
Now because I am not very experienced in Linux I am not sure which device is enslaved and which is the master (even though I am thinking that eth0 should be the enslaved one as it is the physical and vunl0_1_0 is the master ) which could give me at least a direction (maybe change the OS version)
Now if you can enlighten me here and offer any solution I will be glad to try it so I can go and continue with my labs.
PS: If there is an EVE-NG issue that needs to be patched and there will be a couple of months to wait . Is there any possibility to install UNL meanwhile ( it worked perfectly but I had some HDD failure this week and didn't find the resource on how to install it again). ? It doesn't matter if there will be no updates on it
Thank you !