An unmanaged switch doesn't use/care for/understand IP addresses at all.
A managed L2 switch uses IP addresses for management only. Some L2 switches also support limited L3/IP functionality like ACLs. L3 switches use IP addresses for L3 forwarding = routing as well.
When computer A on a local network wants to communicate with computer B for example through ...
That shouldn't happen. Each interface should have a unique MAC address.
Actually, each interface's MAC address should be globally unique but that isn't really the case any more. However, the vendor needs to take care that an address collision is extremely unlikely (re-use addresses only between different geographic regions).
Using the same MAC address in ...
Any device randomizing its MAC address is responsible for the side effects, unless each MAC address has already been pre-allocated for that device (which kinda defies the randomization). Therefore, it needs to take all possible precaution to not collide with any other device. However, there's no standard protocol and the device is mostly restricted to ...
layer 2 protocols use MAC addresses for communication, which doesn't really depend on what network you're in or if you're in a network at all.
Oh yes, it does.
A data link layer (L2) network is required for communication, with an underlying physical layer network (L1: interfaces, ports, cabling) as well. The L2 network doesn't necessarily need to use MAC ...
Unless I'm completely mistaken, TCP/IP traffic is still routed to the
actual endpoint via MAC addressing on the network layer, hence the
need for unique MAC addresses.
No, you are mistaken. MAC addressing is used by some data-link protocols to deliver layer-2 frames in the same network. MAC addressing is only ever used and seen on the same layer-2 network. ...
MAC address: DE:AD:BE:EF:00:00
That is a locally administered MAC address as indicated by the second-least significant bit in the first octet (0xE = 1110b). As such, it is the responsibility of the user/administrator to make sure it's unique in their network.
AFAIK there's no regulation prohibiting such a practice (there should be) but it may be considered ...
When a device tries to associate an IP address with a MAC address for a given device, it sends ARP requests to all devices connected to the same network.
For IPv4 that's correct. IPv6 uses the Neighbor Discovery Protocol (NDP).
how does the searching device find the MAC addresses of the other devices in the first place?
It doesn't. ARP uses the link-layer ...
When a device tries to associate an IP address with a MAC address for
a given device, it sends ARP requests to all devices connected to the
It is really a device trying to discover the MAC address for a given IPv4 address (IPv6 uses NDP).
My question is this: how does the searching device find the MAC
addresses of the other devices in the ...
Switches do forward broadcasts out all ports. They don't care what kind of broadcast it is, and they're unaware of layer 3 information. So yes, they'll forward it. L3 devices, like routers will drop broadcasts. Not sure what you consider "chaos."
can 2 device communicate with each other without data going to physical layer?
No. The physical layer is the one actually doing the bit lifting. Without physically transporting the data to another location there's no communication.
You're forgetting that the data link layer 'only' controls the flow of packets over a local network. Still, the packets need to ...
Do unicast frames get sent to all switches associated with the target VLAN?
No. Switches forward unicast frames out of the port associated with their destination address only (might be another switch or the destination itself). The port associations are learned and updated by inspecting each frame's source address. If you use VLANs, frames do not leave ...
How does the process of replacing the incoming message (say from stackexchange) to the router, and modifying the header with my PC IP happen?
Since you're referring to Stack Exchange, you seem to mean source NAT (NAPT) which is used for private-to-public translation, especially for Internet access.
For a new outgoing connection, the NAT router creates an ...
I've also encountered this problem two times allready. Unfortunaltely HPE support was not very helpfull. But my mac-learning problems always related to distributed trunking.
My workaround was to step away from dt.
I recommend to use VSF if possible (V3 Modules and zl2 Switch required among other things: https://higherlogicdownload.s3.amazonaws.com/HPE/...
The quote is a clumsy attempt to say that the NIC is the hardware that actually transmits and receives the frames. As such, it creates the frames when transmitting (using it's built-in address as the source). For receiving, it determines if the frames are addressed to the device (NIC), ignoring frames that aren't.
The MAC address is the "hardware" address of a NIC. It uses that address as source address for frames it sends, and only accepts frames that are destined for that address (plus broadcasts and possibly other MAC addresses it has been configured with).
The NIC is also the device that's forming the frames that run across the network.
I understand that a L2 device like WAP is needed to forward packets
from one client to another within the same network (based on MAC
mostly like in 802.11)
Only in Infrastructure Mode.
However my question is, if there is no WAP and just a wifi router
available(L3 device) to which clients are connected..then.. is it
really possible for clients to be able to ...
I want to understand why two devices can not talk to each other directly when they are in wireless range of each other?
All devices only communicate with the WAP in Infrastructure Mode. That actually increases the range of the Wi-Fi network, and it makes for a more stable network, especially if the Wi-Fi devices are mobile.
Is knowing MAC of each other ...
I notice two things:
On the router, GigabitEthernet5 where your end system is connected to does not have spanning-tree portfast (or spanning tree port type edge in some younger Cisco dialects).
(assuming that Windows still has that 4sec-something delay to report a "time out" in ping) there seems to be somewhat of a pattern of 6 timeouts, 1 success,...
Client's ARP requests affect whether Cisco C897VA receives packages or not
ping is not necessarily a meaningful instrument to determine if packets make it or not in general. ICMP generation is low-priority anyway and may additionally be limited by (default) configuration for DoS protection.
What you see is likely an artifact created by a) the router's ICMP ...
If two devices use the same MAC address they won't function correctly on a LAN. Similar issues could happen with Bluetooth and other technologies which depend on the uniqueness guarantee of MAC addresses.
IEEE previously allocated OUI addresses in a sequential manner. Some manufacturers simply stopped paying for allocations, and began squatting on blocks ...
short answer: the first packet is not sent out before ARP completes. In order to send a packet, the sender (usually operating system) needs to fill out its IP and Ethernet header. Thus it cannot send a packet before MAC of P3 is known, i.e. ARP completes.
The idea of what happens on the host is following.
application (ping) gets an address (...
Each host maintains an ARP cache, where it stores for a predefined amount of time (quite large) this information.
Thus if a communication occurred recently between the two hosts, the address is already in the cache.
Additionally some host (depending on their operating system) emit a gratuitous arp when their interface comes UP. I.E. they advertise to the ...
when I try to ping PC3 from PC1, I expected the first packet to timeout
The attempt of using PC3's address as destination triggers the ARP request. ping's ICMP echo request can only be sent after the ARP has succeeded. Before knowing the destination's MAC address, there is nothing to even send the echo request to.
How did the packet arrive without knowing ...
DHCP isn't of any help here, but a managed switch usually is.
You cannot really test a cable against the specs without the proper equipment ie. a certifying cable tester. These testers start around 800€, so you might not have one around. Cheaper cable testers test for shorts and continuity only (10-50€), or may also check correct pairings (200-400€). Neither ...
The command arp -a can be used to list the IP addresses and physical addresses of the devices that are on the same network.
No. arp -a lists the MAC-IP associations that are currently in the host cache (from being used previously for L3 traffic).
To find all IP addresses on your network check your DHCP server and/or your network documentation. Alternatively,...
frames addressed to a particular machine (ip 192.168.0.20, mac bbbb.bbbb.bbbb) are sent to certain devices in that VLAN.
Unless that MAC address is a multicast address, it needs to be unique in its VLAN - "certain devices" should be "a certain device".
 You seem to be referring to devices attached to the right-hand Nexus - sorry ...
The mac address-table entries, however, are missing bbbb.bbbb.bbbb.
That is how a self-learning bridge works: it learns the location of nodes by the frames' source addresses and then forwards frames by the destination address based on the learned table.
If the node in question has never send a frame, the bridge/switch cannot know its location and ...