Yes EX3400 (along with EX2300, EX4300, EX4600, and EX9200) are all ELS based hardware platforms. EX3400 is missing IP Source Guard. This feature is supposed to be added into this product in either 17.3 or 17.4 I believe.
This why you can not config IPSG on EX3400 with any current code available. There have been other threads in this forum covering this.
I do not know the reason why you are seeing what you are seeing, but I might assume an aging time of 0 would be associated with a static MAC address. Any chance you have static MAC associated with Physical Interfaces?
How about a global setting?
set protocols layer2-control bpdu-block disable-timeout
Regards, Wojtek
I powered up the EX2300-C switch tonight. I noticed that the default Vlan on the EX2300-C is VLAN #1. On the EX3300 switch, the default VLAN is VLAN #0. I am using the default VLAN on the EX3300 for most of the ports. This vlan is my HOME LAN. It is also setup as my native VLAN as well.
Do I need to setup a VLAN 1 on the EX3300 or a VLAN 0 on the EX2300-C?
Anyone? I need a response as soon as possible.
Is there any more data with the log like the actual index number?
We can see all the ifl setup on the device this way
show interfaces extensive | match ifIndex
I also did a search in the PR database (Problem Reports - Junos bugs) and I don't see this listed but your version of Junos was not availalbe as a filter either. This is starting to feel like a bug.
prsearch.juniper.net/InfoCenter/index?page=prsearch
If you are not using tagged/trunk connections between the switches then VLAN numbers only have local signature so it does not matter what the VLAN numbers are. Now if you want to interconnect the 2 switches and have all interfaces in the same VLAN then I would suggest you change the VLAN ID on the EX3300 from 0 to 1.
In the original EX3300 code the default configuration use VLAN ID 0 for all the interfaces. In general VLAN ID of 0 should not have been used as I think VLAN ID of 0 should really be the bit bucket; not the case for Juniper EX. Anyway in later version of code the default value was changed to a VLAN ID of 1. In either case, using VLAN ID of 0 or 1, all the interfaces belong to same L2 [bridge] domain or broadcast domain.
If you change the VLAN ID on the EX3300 to 1, then you can interconnect the switches and all interfaces on both switches will be members of same VLAN.
Hopefully this helps you.
I have no mac addresses defined anywhere in the config, hopefully that's what you're asking?
By the way, these are EX4500 virtual chassis all with 2 members. We are running 12.3R12.4 on 3 out of 4. The other one has 15.1R5.5. This problem with all macs with age 0 happens on both versions. These EX switches are strictly layer 2 devices and all are configured pretty similarly. 2 out of the 4 (on 12.3R12.4) have ages on some mac addresses. I can't figure out why.
Thanks.
For this one, I think you'll need to work with Juniper TAC. 12.3R12 should be good. There is a 12.3R12-S4 release which you could ask JTAC about and try that, but I suspect you'll see same situation and then need TAC to assist you with debugging. Could maybe be just a 'display' issue???
Got into the ex3300 and get the output of show configuration vlans | display set and send that, or better yet contact Juniper TAC and work with them. If you do not have a service contract with Juniper, purchase one -
I'm trying to enable "forwarding- options dhcp-relay" but keep getting an error when I try to enable dhcp on my me0 interface. I would think since the management interface is out of band it whouldn't conflict with my in-band dhcp relays.
root@AMTSGY-JN2200c-DSwc# commit check
[edit interfaces me0 unit 0 family inet]
'dhcp'
Incompatible with 'forwarding-options dhcp-relay'
error: configuration check-out failed: (statements constraint check failed)
Is there a work around for this?
I'm running on a JN2200C version 12.3R12.4
Here is the results of the command show configuration vlans | display set
root> show configuration vlans | display set
set vlans Guest_Network vlan-id 30
set vlans Hypervisors_Servers vlan-id 10
set vlans Hypervisors_Servers l3-interface vlan.10
set vlans Management vlan-id 99
set vlans Management l3-interface vlan.99
set vlans NAS_SAN_Storage vlan-id 20
set vlans NAS_SAN_Storage l3-interface vlan.20
set vlans Transit_Network vlan-id 2000
set vlans Transit_Network l3-interface vlan.2000
set vlans default l3-interface vlan.0
{master:0}
root>
Hi.
I'm planning to setup a network in one of our locations based on a number of VCs that due to physical limitations have to be connected togetter in a ring topology. I have been searching for a while for protocol alternatives in this situation, or is the only supportet protocol xSTP in the situation? Do Juniper provide any aditional options here? I looked at ERP, but this is not supported by EX3400.
Quick schematic:
Best regards,
Johan Christensson
Not exactly sure what you are trying to do, but maybe send your whole config, in both display set and hierarchy of statements format. If you are trying to make me0 into a dhcp-relay agent, I believe this will not work, as me0 is basically the CPU which acts like a host, not a router. You also can not configure a Junos system as both a local server and a dhcp-relay or bootp helper - https://www.juniper.net/documentation/en_US/junos/topics/concept/dhcp-relay-overview.html.
Bootp helper was the older CLI format, which has now changed to [extended] dhcp-relay nomenclature, which I assume you are using. EX2200 with 12.1R'x' code supports both was, but extended is recommended. New products/SW versions only support extended.
I believe this should do the trick
delete vlans default
set vlans default vlan-id 1
set interfaces interface-name unit 0 family ethernet-switching vlan members default (interface-name is all interfaces you want as member of vlan 'default', vlan-id which is now 1; unit number must be 0 for family ethernet switching. This may already be in the config, so need to add.
set interface vlan unit 1 family inet <IP address/mask>
set vlans default l3-interface vlan.1
For last 2 set commands unit # must match, and best practice is to map this to vlan-id, although not required.
I used a combination of the command line and the J-Web GUI to change the Default vlan ID from 0 to 1.
Here is the results of othe show configuration vlans | display set now...
root@:RE:0% cli
{master:0}
root> show configuration vlans | display set
set vlans Guest_Network vlan-id 30
set vlans Hypervisors_Servers vlan-id 10
set vlans Hypervisors_Servers l3-interface vlan.10
set vlans Management vlan-id 99
set vlans Management l3-interface vlan.99
set vlans NAS_SAN_Storage vlan-id 20
set vlans NAS_SAN_Storage l3-interface vlan.20
set vlans Transit_Network vlan-id 2000
set vlans Transit_Network l3-interface vlan.2000
set vlans default vlan-id 1
set vlans default l3-interface vlan.1
{master:0}
root>
All working? If not, need whole config.
rccpgm wrote:All working? If not, need whole config.
Everything seems to be working.. I will test more over the next couple days.
rccpgm wrote:All working? If not, need whole config.
So, I did not run the "delete vlans default" command.
Based on the results below, it looks like there are 2 link aggregation interfaces tied to vlan 0. These are the uplinks to my pfSense router/firewall. Are the ports used in the link aggregration still tied tyo vlan 0 and not vlan 1? Please advise on what to do to correct this issue.
root> show vlans 0 extensive
VLAN: __juniper_vlan_internal__, Created at: Mon Aug 7 23:24:07 2017
Internal index: 7, Admin State: Enabled, Origin: Static
Protocol: Port Mode, Mac aging time: 0 seconds
Number of interfaces: Tagged 0 (Active = 0), Untagged 2 (Active = 0)
ae1.0*, untagged, trunk
ae2.0*, untagged, trunk
