Multiple CRS317 switches - no more than one management port works on the same VLAN

We have two CRS317-1G-16S switches and just experienced something which was so unexpected, took me over a day to figure out.

For some reason, when we connect both switches management ports to the same VLAN to manage (we keep all management ports on a separate VLAN) one switch is forced off the network and refuses to respond.

It does not matter if we use DHCP or static IP address, same result.

We have to manage via one of the other main SFP ports or the trunk uplinks and appropriate vlan.

Maybe there is something I am missing from the setup on this but whatever it is, it is not obvious and haven’t yet seen anything anywhere which describes this behaviour.

maybe someone can tell me if there is a solution for this as we would like to use the dedicated management ports.

Graham

Can you share their config (I assume you are using RouterOS)?

/export file=anynameyoulike

Remove serial and any other private info, post as Preformatted text by using the </> button.

First thing I would check is if they have their own unique MAC addresses.

I can;t seem to upload the swb file but the configuration on both switches remain unchanged form the default configuration after reset apart form the specification of the ipaddress. currently using static IP as shown.

Not RouterOS, Using SWOS 2.18.

Wanted to keep it simple as do not need any more functionality than this.

Yes, Mac addresses are unique

Switch1

        model: CRS317-1G-16S+

serial-number: XXXXXXXXXX61

current-firmwware: 2.18

  mac-address: 08:55:31:01:57:62

   ip-address: 10.1.4.189

eth1 MAC address: 08:55:31:01:57:62

eth2 MAC address: 08:55:31:01:57:63

eth3 MAC address: 08:55:31:01:57:64

eth4 MAC address: 08:55:31:01:57:65

eth5 MAC address: 08:55:31:01:57:66

eth6 MAC address: 08:55:31:01:57:67

eth7 MAC address: 08:55:31:01:57:68

eth8 MAC address: 08:55:31:01:57:69

eth9 MAC address: 08:55:31:01:57:6A

eth10 MAC address: 08:55:31:01:57:6B

eth11 MAC address: 08:55:31:01:57:6C

eth12 MAC address: 08:55:31:01:57:6D

eth13 MAC address: 08:55:31:01:57:6E

eth14 MAC address: 08:55:31:01:57:6F

eth15 MAC address: 08:55:31:01:57:70

eth16 MAC address: 08:55:31:01:57:71

eth17 MAC address: 08:55:31:01:57:72

Switch2

        model: CRS317-1G-16S+

serial-number: XXXXXXXXXX98

current-firmwware: 2.18

  mac-address: 08:55:31:01:4F:E1

   ip-address: 10.1.4.188

eth1 MAC address: 08:55:31:01:4F:E1

eth2 MAC address: 08:55:31:01:4F:E2

eth3 MAC address: 08:55:31:01:4F:E3

eth4 MAC address: 08:55:31:01:4F:E4

eth5 MAC address: 08:55:31:01:4F:E5

eth6 MAC address: 08:55:31:01:4F:E6

eth7 MAC address: 08:55:31:01:4F:E7

eth8 MAC address: 08:55:31:01:4F:E8

eth9 MAC address: 08:55:31:01:4F:E9

eth10 MAC address: 08:55:31:01:4F:EA

eth11 MAC address: 08:55:31:01:4F:EB

eth12 MAC address: 08:55:31:01:4F:EC

eth13 MAC address: 08:55:31:01:4F:ED

eth14 MAC address: 08:55:31:01:4F:EE

eth15 MAC address: 08:55:31:01:4F:EF

eth16 MAC address: 08:55:31:01:4F:F0

eth17 MAC address: 08:55:31:01:4F:F1

Don’t know the answer, but if I were in your situation I would verify each switch independently by setting up the vlan on my laptop and connecting directly to each switch one at a time. If that failed I’d wireshark the situation. If it succeeded then I’d move to the next step by connecting both switches and my laptop to a basic dumb switch I have laying around and see how that went. I’d just keep working up from there trying to add as few variables as possible on each new step until I isolated the problem.

Are there loops in your network?

Maybe STP (SwOS supports only RSTP, not MSTP) is disabling the port that has your management vlan.

Just a guess.