Hot answers tagged

10

All routing protocols try to avoid a loop topology. If there is a routing loop, data is not getting to the destination. If the path from source to destination is A -> B -> C -> D, but there is a loop between B and C, data will never reach D.


6

I don't understand how a bridge decides that it is not connected to another bridge performing STP. If it doesn't hear BPDUs, then there's no bridge connected. How long will it until it decides it is alone and probably the root bridge? It starts out assuming it's "alone." That is, the bridge sends BPDUs with its bridge ID as the root until it hears ...


6

Assuming the firewall is a Layer 3 firewall (and not in layer 2 mode like some firewalls are) this won't generate a loop. If it will work, that's a totally different question, that depends on your actual configuration.


5

You didn't actually connect two router ports, you connected two switch ports. The router has a switch module. You created a layer-2 broadcast storm. Broadcast is an essential part of ethernet. When a broadcast comes into a switch port, a switch must send it to all the other ports. When you connected the two switch ports, it created a loop where the ...


5

I think this might be related to a bug. This issue is documented in Cisco bug ID CSCea46385. To resolve the issue, you need to add the no keepalive command to the interface. Hope this helps you on your quest. SleepyMan


5

You should turn loop guard on those ports... if rstp blocks a port and looses bpdus, it will unblock and form an unmanaged loop unless stp loopguard is on the port. If an rstp port looses bpdus on a loopguard port, the port will disable in a loop inconsistent state until it starts receiving bpdus again.


4

You say you've read "the official documentation", but does that include the actual RFC, RFC 3561? I believe your concerns are addressed in Section 6.1, Maintaining Sequence Numbers. It says, in part: The only other circumstance in which a node may change the destination sequence number in one of its route table entries is in response to a lost ...


4

In link state protocols each participating router knows the whole topology of a network, which is stored in the link state database. Using the link state database each router will calculate the best path to the destination. The easiest way to think about it is as if with a link state routing protocol is that you give all your friends a map (the LSDB) and ...


4

If I put two interfaces (one on each SRX) in a reth, is that sufficient to ensure that there won't be any loops? Yes* - reth interfaces are essentially a logical L3 interface - they will not loop traffic between node0 and node1 of the SRX chassis cluster Is this the correct/optimal way to ensure redundancy with the two upstream links? No. Ask your ...


4

If this was your setup: Host A ---- Router B ---- Router C ---- Router D ---- Host E Tell Router B that the network for Host E exists at Router C. Tell Router C that the network for Host E exists at Router B. A packet from Host A to Host E will then get caught in a routing loop between Routers B and C.


3

The TTL field in the IP header does not eliminate a routing loop. It just prevents the packets from being endlessly caught circling in the loop once it exists in that after 255 hops at most, they will be dropped. However, the routing loop is a state of routing tables on multiple routers where, for a destination, router A points to router B, router B points ...


3

If the network manager does something by mistake it's just pilot error; if I sneak into your building and do something to make trouble for you on purpose, it's an attack. If the attack is designed so that you can't use your own things, it's a denial of service attack. Advertising film stars arriving at your building, for example, or throwing marbles down ...


3

Without knowing a lot more about your infrastructure... Disable the "data" port on all your phones. You appear to have dedicated voice and data ports to everyone, so there's no need for daisy chaining through the phone. Enable what Cisco calls bpdu guard. If any spanning-tree BPDU is seen on an "edge" port, the port is immediately disabled. Running CDP / ...


3

AIUI this is an application for vrfs. You have a "clean" vrf that faces the inside and a "dirty" vrf that reaches the outside. So effectively you would have two virtual routers one behind the other. You would then arrange the filters on your routing protocol such that the "scrubbing" routes only get installed into the "dirty" vrf. The packets coming out of ...


3

Actually, STP figures out the least-cost (shortest), loop-free path to the root switch. This is the path used for all traffic not directly attached to the switch. STP uses BPDUs to figure this out. With your scenario, three switches connected in a triangle, both of the non-root switches will have directly connected links to the root switch, and each link ...


2

RSTP is completely interoperable with STP. as per standards RSTP enabled port will go STP when connected to an STP enabled network. I assume the older 1810 switches support STP and the 1810-v2 switches support RSTP, So in your setup the loop prevention should work fine. Do config check on all switches to ensure they have the STP or RSTP enabled at switch ...


2

Troubleshooting is required. First identify the culprit. Turn off VRRP (or simply power off one of those routers). Does that fix it? If so look at the VRRP config. Otherwise, power off a VSS switch. Does that fix it? Start by identifying the problem... Not sure if you are talking FEX's here but ideally those are NOT dual homed (as in your diagram), ...


2

To meet this requirement, you can configure switch ports as protected ports (also referred to as private VLAN edge ports). Protected ports do not forward any traffic to protected ports on the same switchso configure port attached to host and server as protected port . This means that all traffic passing between protected ports be forwarded through a Layer 3 ...


2

This is your configuration. You need to configure the host and one port of the switch that connect to the host one the same vlan(23). Then, on the other switch, the 2 ports need to be on another same vlan(24). Finally, do the same with the other port and the server port(vlan25). All port must be configured like ACCESS-PORTS. Remember that


2

Switch uses Layer 2 Spanning Tree Protocol to ensures a loop free topology for any bridged Ethernet network. So if yours IOT device will forward bridge protocol data unit (BPDU) accross, switch will maintain loop free topology. BPDU are transsmited accross local network to detect loops in network topologies. If BPDU won't be forwarded you can possibly have a ...


2

The VPC peer link has a few special roles - it carries the CFSoE traffic (mentioned below) to allow the peers to sync (enabled automatically when feature vpc is turned on) and carries traffic for orphan ports. It doesn't behave completely like a normal port channel and, as a result, it's best practice to keep non-VPC VLAN's on a different physical link. ...


2

I'm not quite sure CFS is the thing preventing loops. It is more about syncing and consistency checks. There's a simple rule though: do not send traffic out of member port if it was received over a peer link. If peer link breaks, nothing works. It's a split brain scenario, where each switch believes it's a master. In theory there can be loop (you can see ...


2

Since you are considering changing switches, you can use mac-based-vlan on the switches (or voice-vlan). This will put the phones on a specific Vlan based on the MAC of your phones. Now all data port and phone port will be technically identical and you can chain phones and computer, even using 2 different DHCP servers. Then you will just have to prevent ...


2

If you plan to create a ring of switches, all those switches need to support RSTP or MSTP. That is, unless all the unmanaged switches completely ignore IEEE 802.1D and actually forward the RSTP BPDUs - which is very ugly but quite common. (That way, the RSTP switch sees its own BPDUs from one ring port coming in through the other one, and blocks the one with ...


1

A connected to B connected to C connected back to A. You now have a loop. A new frame arriving on an edge port on A will be sent to the links to B and C. They won't hand it back to A but they will hand it to each other. As the frame didn't come from the port leading to A, they both pass the frame on to A. Switches don't keep a hash table of every frame ...


1

After upgrade to newer version 9.8.2 resolved all of the issue. Thanks for your support.


1

An IP address isn't assigned to a router. The router has interfaces and that's where the IP addresses are assigned to. In the diagram, let's assume router 1 has 195.1.1.1/24 assigned to its downfacing interface. Accordingly, routers 2-5 would have an IP address from the same subnet assigned to each upfacing interface. Let's assume these are 195.1.1.2, 195.1....


1

First, routers do not forward broadcasts, but assume it was traffic destined for a different, unicast address. Routers route traffic between networks, not from a network back to the same network. Assume Router A has four interfaces, one for each of the networks. Any traffic from one of the networks to any of the other three networks will enter Router A from ...


1

There is no theoretical limit, but memory/CPU may impose practical limits on particular hardware. Depending on the topology, increasing the number of devices increases the convergence time. When that becomes a problem depends on your particular situation. In my experience, loops tend to be small. I don't recall ever seeing a looped configuration larger ...


1

BPDUs are sent out all STP ports. This is because you need to detect a topology change. If the topology changes at a distant switch, the current designated or root ports may need to start blocking, and a blocking interface may need to change to a designated or root port. A switch cannot detect this on a port where it is not sending or receiving BPDUs. STP is ...


Only top voted, non community-wiki answers of a minimum length are eligible