Difference between revisions of "Core Protocol"
From ASSS Wiki
(added core packet addendum for reliable packet handling) |
m (moved Core protocol to Core Protocol over redirect) |
(No difference)
|
Latest revision as of 13:59, 16 February 2015
This protocol is UDP and bi-directional. It can be thought of as another layer within an application's network layer. Subgame, ASSS, Continuum, VIE's SubSpace client, the Directory server, and some Billing Servers utilize this protocol.
There is support for reliable ordered packets, large packets and cluster packets.
- The maximum packet size is 520 bytes
- All integers are little endian
- All core packets begin with the type byte 0x00
- Packets can be nested (They are usually processed recursively)
0x01 login offset size comment 0 1 type 0x00 1 1 type 0x01 2 4 client key 6 2 protocol version (vie = 0x01, ctm = 0x11) 0x02 login response offset size comment 0 1 type 0x00 1 1 type 0x02 2 4 ~client key (negative) 0x03 reliable header (see reliable packet addendum below) offset size comment 0 1 type 0x00 1 1 type 0x03 2 4 packet id (ie: ACK_ID) 6 ? payload (514 bytes max) 0x04 reliable acknowledge (see reliable packet addendum below) offset size comment 0 1 type 0x00 1 1 type 0x04 2 4 packet id (ie: ACK_ID) 0x05 sync offset size comment 0 1 type 0x00 1 1 type 0x05 2 4 local time 6 4 packets sent 10 4 packets received 0x06 sync response offset size comment 0 1 type 0x00 1 1 type 0x06 2 4 time from last received 0x05 packet 6 4 server time 0x07 disconnect offset size comment 0 1 type 0x00 1 1 type 0x07 0x08 small chunk offset size comment 0 1 type 0x00 1 1 type 0x08 2 ? payload (expect 472 max) keep appending payloads until 0x09 chunk tail is received 0x09 chunk tail offset size comment 0 1 type 0x00 1 1 type 0x09 2 ? payload (expect 472 or less) append this payload then process contents as a non-core packet 0x0A stream offset size comment 0 1 type 0x00 1 1 type 0x0A 2 4 total length of all segments 6 ? payload keep appending payloads until total length is reached then process contents as a non-core packet. 0x0B cancel stream request offset size comment 0 1 type 0x00 1 1 type 0x0B 0x0C cancel stream acknowledge offset size comment 0 1 type 0x00 1 1 type 0x0C 0x0E cluster offset size comment 0 1 type 0x00 1 1 type 0x0E 2 1 length 3 ? payload with length as above packet repeats from offset 2 until end, process contents as arbitrary packet 0x10 Continuum alternate encryption response (s2c)? offset size comment 0 1 type 0x00 1 1 type 0x10 2 10 ?? 0x11 Continuum alternate encryption response response (c2s)? offset size comment 0 1 type 0x00 1 1 type 0x11 2 6 ??
Reliable Packet Handling Addendum
Here are some tips when implementing a reliable packet handler stack.
For outgoing traffic:
1. All outgoing traffic must be put into a send list until an ACKnowledgement (0x04) is received with the same ACK_ID as the traffic. 2. Sometimes double-sending reliable messages is warranted, given high-bandwidth and high-packetloss. 3. ACK's (0x04) should cause disconnection if you have not sent a message with that ACK_ID yet. 4. ACK_ID starts at 0 for both client and server, ACK_ID is incremented by one every time.
For incoming traffic:
1. ACK (0x04) all incoming Reliable traffic no matter what. ACK the traffic twice if it makes you feel happy. 2. All incoming traffic must be processed in the order of the ACK_ID's. 3. If a packet is lost, all traffic stamped with an ACK_ID higher than the next expected ACK_ID must be placed on a backlog list of packets until the packet in sequence is finally received. 4. If an incoming reliable message's ACK_ID has already been processed, ignore the packet.