GNS3 / gns3-server

GNS3 server
GNU General Public License v3.0
768 stars 258 forks source link

GNS3 Network Adapter Config Not Applying #2323

Closed nepgituser closed 7 months ago

nepgituser commented 7 months ago

Hello,

Been trying to set up a test environment with a Dell OS9 instance but ran into an issue with network adapters. What's configured in "Configure custom adapters" on the Network tab is not properly loaded onto the switch. Even more puzzling, every time the switch is rebooted the MAC addresses on the fortyGigE ports are different than what they were before, but all of the ports have the same one.

fortyGigE 0/0 is up, line protocol is up (keepalive not set) Hardware is DellEth, address is 00:01:e8:1e:b9:13

fortyGigE 0/4 is up, line protocol is up (keepalive not set) Hardware is DellEth, address is 00:01:e8:1e:b9:13

The management network does work. However, it too has a different MAC that what's displayed in the screenshot below. At first it didn't until "no keepalive" was run on it. The same was done for the fortyGigE ports but it didn't make a difference.

ManagementEthernet 0/0 is up, line protocol is up (keepalive not set) Hardware is DellEth, address is dc:d4:ea:cc:00:02

NetworkAdapters

FILES USED

This is running on a remote server (Hyper-V) and we access it via a client-side GNS3 install. We are using 2.2.43 across the board.

I'm hoping this is a simple configuration issue, but when we ran into the issue, we set up an OS10 instance and it is working correctly. Not sure if the OS9 issue is to the image, it's from Dell (seemingly legit), but maybe that is the problem.

Anyway, looking for a little help if someone is familiar. Not sure what other information to provide, so if you need something please ask. Thanks.

Edit: We have also tried the second 9.13 version on that Dell page. It simply won't boot. The process starts and it runs through the memory check but then it simply stalls. After a long while, it appears to restart too.

Edit: I noticed that the remote server VM has two QEMU versions available: 7 and 4.2.1 (using this). Is it possible that this could cause an issue? I'd think unlikely due to OS10 working fine.

grossmj commented 7 months ago

Correct the interfaces are completely messed up, looks like Dell OS9 is expecting quad port NICs, I could see this message when it booted up with a pcnet adapter:

stack-unit 0 quad-port-profile % Error: Incomplete command.

Also, booting with e1000 interfaces, I could see a bunch of "Unable to get the mac-addr property!".

wm0 at pci0: Intel i82540EM 1000BASE-T Ethernet, rev. 3
wm0: interrupting at irq 11
Unable to get the mac-addr property!
wm0: Ethernet address 0c:12:17:56:00:00
makphy0 at wm0 phy 1: Marvell 88E1011 Gigabit PHY, rev. 0
makphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
wm1 at pci0: Intel i82540EM 1000BASE-T Ethernet, rev. 3
wm1: interrupting at irq 11
Unable to get the mac-addr property!
wm1: Ethernet address 0c:12:17:56:00:01
makphy1 at wm1 phy 1: Marvell 88E1011 Gigabit PHY, rev. 0
makphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
wm2 at pci0: Intel i82540EM 1000BASE-T Ethernet, rev. 3
wm2: interrupting at irq 10
Unable to get the mac-addr property!
wm2: Ethernet address 0c:12:17:56:00:02
makphy2 at wm2 phy 1: Marvell 88E1011 Gigabit PHY, rev. 0
makphy2: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
wm3 at pci0: Intel i82540EM 1000BASE-T Ethernet, rev. 3
wm3: interrupting at irq 10
Unable to get the mac-addr property!
wm3: Ethernet address 0c:12:17:56:00:03
makphy3 at wm3 phy 1: Marvell 88E1011 Gigabit PHY, rev. 0
makphy3: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
wm4 at pci0: Intel i82540EM 1000BASE-T Ethernet, rev. 3
wm4: interrupting at irq 11
Unable to get the mac-addr property!
wm4: Ethernet address 0c:12:17:56:00:04
makphy4 at wm4 phy 1: Marvell 88E1011 Gigabit PHY, rev. 0
makphy4: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
wm5 at pci0: Intel i82540EM 1000BASE-T Ethernet, rev. 3
wm5: interrupting at irq 11
Unable to get the mac-addr property!
wm5: Ethernet address 0c:12:17:56:00:05
makphy5 at wm5 phy 1: Marvell 88E1011 Gigabit PHY, rev. 0

Screenshot from 2023-12-07 15-38-45

I'm hoping this is a simple configuration issue, but when we ran into the issue, we set up an OS10 instance and it is working correctly. Not sure if the OS9 issue is to the image, it's from Dell (seemingly legit), but maybe that is the problem.

I think the problem comes from the appliance itself. Ideally, Dell should provide instructions to run it with Qemu/KVM.

Anyway, I don't want to spend too much time on this since it is marked as "End of Support" and that OS10 works as expected. Maybe you could try to but with different network adapters to see if there are any change?

Edit: We have also tried the second 9.13 version on that Dell page. It simply won't boot. The process starts and it runs through the memory check but then it simply stalls. After a long while, it appears to restart too.

I have updated our appliance file to include version 9.13. We had to increase the RAM (we picked 1024 MB) and used "telnet" for the console.

Good luck :)