Wireshark-bugs: [Wireshark-bugs] [Bug 9427] Dissector for T1-channels-over-raw-Ethernet protocol
Comment # 34
on bug 9427
from Guy Harris
OK, so what I very strongly suspect is happening here in the process of going
from Frame Relay frames to the T1 is:
the Frame Relay frames have HDLC framing applied to them, so that
1) the octets (bytes) of the Frame Relay frames (address bytes, control
byte, payload bytes, FCS bytes) are turned into a bit stream;
2) if there are 5 1's in a row in that bit stream, a 0 is inserted
after the 5 1's;
3) between Frame Relay frames, a bit patterns of 01111110 is inserted
(sufficient to distinguish frames; it might only insert one of them between
immediately adjacent frames, but, if there's an idle time between frames, there
might be one at the end of a frame and then another at the beginning of the
next frame);
the resulting bit stream is handed to a CSU/DSU, and the DSU does whatever
magic is necessary to send that bit stream over an "unchannelized" T1:
http://www.infocellar.com/networks/t1-t3.htm/channelized.htm
I've already described the first of those processes. The infocellar.com page
says of the "unchannelized" part:
The concept of "unchannelized" simply means the sender and receiver have
configured their equipment to use all 24 DS0's as one serial stream of data.
The sender places a 1.536 Mbps stream of data onto the T1, breaking it up into
24 time-division multiplexed DS0's and adding 8 kbps of framing bits for
synchronization. The receiver takes the 1.544 Mpbs stream, removes the 8 kbps
framing bits, and concatenates the 24 DS0's into a single 1.536. So the T1 is
channelized but the way the data is used is not - it is unchannelized after
reception.
...
Therefore, all T1's are channelized, but when the sender & receiver combine
all 24 channels into one 1.536 data stream, then it is called unchannelized -
even though the T1 itself is still channelized. They could have avoided all
the confusion by simply calling it T1 unchannelization or T1 channel
concatenation - that way it would be clear that it is a process performed on a
channelized T1 circuit.
So the question that raises is "what does it mean to "concatenate" the
channels"? Does it mean taking 8 bits from channel 1, appending 8 bits from
channel 2, then 8 bits from channel 3, ... 8 bits from channel 24, the next 8
bits from channel 1, the next 8 bits from channel 2, ...?
I'm assuming that
1. Only T1 payload or content is transferred to the Ethernet Frame (no T1
Framing or Overhead)
2. The Packets always start with DS0 Channel-1 (unless the T1 slips due to
incorrect clocking)
3. and continues through to DS0 channel-24
from comment 8 means that, *IF* that's what "concatenation" means, we can just
treat the bytes in the payload portion of the Ethernet packets as a bit stream,
and don't have to worry about T1-layer framing, superframes, extended
superframes, etc. (as per "no T1 Framing or Overhead").
You are receiving this mail because:
- You are watching all bug changes.
