Hot answers tagged

20

You might care to read RFC 1547 "Requirements for an Internet Standard Point-to-Point Protocol" which explains how the PPP was chosen. The thing I'd suggest you are missing is that interoperability is one of the principal driving forces in the internet protocols, and efficiency is much less important. You do the highly talented engineers who ...


17

Why does PPP need a wrapping protocol? PPP is not a layer-1 protocol, so it needs a layer-1 protocol to carry it. Protocols like ethernet are both layer-1 and layer-2 protocols, so PPP can use ethernet as its layer-1 protocol, but that comes with the ethernet layer-2 protocol that wraps PPP. Why can't I just use PPP over Ethernet instead of PPPoE? PPPoE ...


12

PPP is designed to ride on top of a byte-oriented, point-to-point physical-layer protocol like a simple modem-style serial link. Ethernet is no simple serial protocol but it requires frame-level addressing (L2 MAC address), so PPPoE expands standard PPP to take care of discovery and addressing between the link partners.


9

PPTP, PPPoE, and L2TP all provide OSI Layer 2 services. That is, the user of these protocols (usually, a network layer protocol suite) thinks it's running over a "normal" link layer. However, each of these protocols provides the link layer service by transporting packets over another service, rather than over the physical layer. PPTP provides ...


6

( As others have already said, PPPoE is literally PPP over Ethernet. And similarly PPPoA is PPP over ATM. ) Ethernet and ATM are oddities in the networking world as they define both a layer-1 and layer-2 component. In the case of ethernet, it's layer-1 has always used it's layer-2; no one ever built it any other way. (Ethernet's layer-2 protocol, however, ...


5

In the 1990s in a British regional ISP, we had a choice from the telephone company of getting two settings for multiple lines: In the ordinary hunt group, the next incoming call would arrive on the lowest unused line on the group In the rotary hunt group, the next incoming call would hunt from the last position for the next unused line of the group The ...


5

From the HOWTO you linked: rotary dial in set of telephone lines When I wore a younger man's clothes, back in the late 1990ies, I ran a bunch of Shiva LAN Rovers (in their IBM guise labelled as IBM 8235). In short, they were essentially a 1RU box with a LAN interface (our first ones had Token Ring, of course!) and 8 slots for modems or ISDN BRI adapter ...


5

Sure, there is a 2-byte checksum followed by a magic byte, but that is hardly collision-proof. Occurrences of the magic byte are escaped. There are two types of escaping used, called "character stuffing" and "bit stuffing". https://technet.microsoft.com/en-us/library/cc957975.aspx


5

Note that layer three and above protocols can be encapsulated inside the protocols you mention but not layer two (or below). One way to look at it is that you build a protocol stack from the bottom up. So first we put down layers 1, 2, and 3. Then for layer 4 we put PPTP (for example) and that starts a NEW layer 2 on top of (inside of) the "real" layers 1, 2,...


5

In this case, your router talks PPP to the provider directly. But first, the router puts this data into Ethernet frames which it transmits to the modem. (hence PPP-over-Ethernet) Since the modem is in bridge-mode, it won't interpret the frames, only encapsulate them itself in AAL (ATM Adaption Layer) or whatever in order to transmit over the public network ...


5

I'd remove the Ericsson box. It isn't a BRAS. From your diagram it looks like you'd be using it as a aggregation switch and the Linux PPP Server would be your BRAS. Personally, for lab purposes I'd also remove the Linux PPP Server and hook the DSLAM up to a Cisco 3825 or 7200 configured as a BRAS. If you really want the Linux element set up a FreeRADIUS ...


4

Compression technologies rely on the fact that the data has patterns that can be "coded." Here's a simple (simplistic?) example: Instead of repeating long strings of 1s or 0s, use the string "00" followed by the number of 0s. So a string of 20 0s could be represented by "0014" (hex 14 = 20). This would shorten the stream by 16 bytes. But if you encrypt ...


4

CHAP is defined in RFC 1994. You concatenate the identifier, the password (secret), and the challenge, in ASCII, in that order. The response is hashMD5(identifier.secret.challenge), sent in binary (16 bytes for MD5). For your example, that should be in hex 0x017465737412345 hashed. Note that MD5 is not cryptographically safe any more due to advances in ...


4

In real life, you can't stick to the OSI model entirely. It's helpful and necessary as a guide on how to structure the components in a complex network but you can't always say that's layer x. Tunneling is wrapping packets or frames from one layer and using another transport mechanism to transport them where you need them. At the tunnel end you unwrap the ...


4

The address and control fields were designed with future changes in mind, and other values "may be defined at a later time, or by prior agreement." (RFC 1662 "PPP in HDLC-like Framing", section 3.1) You're absolutely right of course, this is overhead. The immediately following section 3.2 explains that they are able to be compressed: ...


3

PPP is independent of the underlying transport, PPPoE requires to be run on Ethernet (or a like tunnel). But of course, you can also run L2TP with PPPoE (RFC 3817).


3

The PPP header starts with a 16-bit protocol id. You need to look at that first in order to know how to interpret the rest. If the protocol id is LCP/IPCP/IPv6CP then the header structure typically looks like this: struct PPP { uint16_t protocolId; uint8_t code; uint8_t identifier; uint16_t length; // <= network-encoded length field }; ...


3

Actually this is relatively easy. Yes, you need AAA, but you don't need an external server. aaa authentication ppp default local aaa authorization network default local aaa attribute list Static-1.2.3.4 attribute type ip-address "1.2.3.4" protocol ip username static privilege 0 password XXXX username static aaa attribute list Static-1.2.3.4 You can use ...


3

PPP is a protocol for transporting higher-layer protocols (like IP) over a simple serial interface. Since you'd already be running Ethernet, PPP doesn't work and is of no use anyway. You could use PPPoE but still without gain. Simply run Ethernet and IP on top and you're fine. All you need are routers or switches with SFP interfaces, suitable SFP ...


3

SLIP was an early protocol (mid-1980s) for running IPv4 over a serial line. It was very cumbersome due to both sides requiring identical configuration and compatible SLIP versions. It was succeeded by PPP in 1994 which enables the passing of parameters between partners and supports many more protocols and options in a downward-compatible way. When using a ...


2

Your question is really too broad, and home networking and consumer-grade devices are explicitly off-topic here, but I will try to give you some general answers. PPP is a layer-2 protocol, the same way that ethernet, HDLC, frame relay, ATM, etc. are. PPP is specifically a point-to-point protocol with only two endpoints. Other protocols, e.g. ethernet, are ...


2

Yes - the BRAS is both stripping the PPP[oE] frame (...and some L2/L3 underlay information, depending on the design in use) off of frames coming from the subscriber device and is adding both the PPP[oE] header and associated L2/L3 mapping information to get packets from the Internet back to the subscriber. Part of the process of the subscriber device ...


2

Note: By default, the router uses its hostname to identify itself to the peer. However, this CHAP username can be changed through the ppp chap hostname command. Refer to PPP Authentication Using the ppp chap hostname and ppp authentication chap callin Commands for more information The information above was taken from http://www.cisco.com/c/en/us/support/...


2

Yes. IP unnumbered allows for this type of configuration on Cisco devices. The ip unnumbered configuration command allows you to enable IP processing on a serial interface without assigning it an explicit IP address. The ip unnumbered interface can "borrow" the IP address of another interface already configured on the router, which conserves network and ...


2

...upon completion of the negotiation, different interface-identifier values are to be selected for the ends of the PPP link. In other words, by the time the negotiation is done, the two ends of the link must end up with two different values for the 64-bit interface identifier. IOW, it should be impossible that the "Client 1" end of the PPP link ends up ...


2

No. The only way to set the address per user is with AAA (radius/tacacs) There used to be ways to pin an address per line, but it's never been possible by user. – Ricky Beam I believe Ricky Beam's comment to be correct: there is no way to do this on IOS without using RADIUS.


2

1) Can I use PPP to connect to my router from PC? If no, then why? That depends. If you have network interfaces in your PC and router that support PPP, then yes. 2) Why I need PPPoE? Where it is used (in which situations)? PPPoE is needed for a link where the other end is running PPPoE. For example, many ISPs offering xDSL will use PPPoE, so you must ...


2

The main point of using PPPoE is to have an authenticated session, a concept Ethernet by itself doesn't have. That allows an ISP to concentrate packets from many users in their backhaul network and sort out the sessions later on. PPPoE is a variant of (serial) PPP, so you can use the exact same authentication scheme for xDSL and serial modem dial-in.


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