It turns out clients pass 0.10 instead of 1.0 as the protocol version.
Distinguish some handlers for 1.0 and 1.1 protocols.
Log protocol version request
Add tests of new library function
Closes#104
DEFAULT_PORTS now a coin property
A Peer object maintains peer information
Revamp LocalRPC "peers" call to show a lot more information
Have lib/jsonrpc.py take care of handling request timeouts
Save and restore peers to a file
Loosen JSON RPC rules so we work with electrum-server and beancurd which don't follow the spec.
Handle incoming server.add_peer requests
Send server.add_peer registrations if peer doesn't have us or correct ports
Verify peers at regular intervals, forget stale peers, verify new peers or those with updated ports
If connecting via one port fails, try the other
Add socks.py for SOCKS4 and SOCKS5 proxying, so Tor servers can now be reached by TCP and SSL
Put full licence boilerplate in lib/ files
Disable IRC advertising on testnet
Serve a Tor banner file if it seems like a connection came from your tor proxy (see ENVIONMENT.rst)
Retry tor proxy hourly, and peers that are about to turn stale
Report more onion peers to a connection that seems to be combing from your tor proxy
Only report good peers to server.peers.subscribe; always report self if valid
Handle peers on the wrong network robustly
Default to 127.0.0.1 rather than localhost for Python <= 3.5.2 compatibility
Put peer name in logs of connections to it
Update docs
Aim to be easier for clients to use, because to do peer
discovery we must act as a client to other servers.
Split out JSON session concept from the asyncio protocol
concept. This makes the JSON RPC support more easily
testable and usable as a stand-alone library.
In addition, support JSON RPC v1 and v2, and auto-detection
of peer's version.
This involved a rewrite of the JSON RPC layer.
I took the opportunity to clean up the handling of requests
in general. It should now be easy to return nice help from
the docstrings.
Closes issue #99
Pause serving sessions whose socket buffer is full (anti-DoS)
Serve requests in batches of 8
Don't store the session in the request
RPC has priority 0; every other session at least 1
Periodically consolidate small session groups into 1
Add new RPC command requests.
Adjust sessions RPC command to show pending requests.
Only keep first 15 chars in version string.
Set socket timeout
Try more forceful closing of socket
Fixes#67
Batch requests are now processed and encoded incrementally.
Their bandwidth usage is also incrementally added, and so overlarge
responses are rejected as soon as they become too large.
JSON text is also more memory efficient than the python data
structures they represent.
Show it in the logs with each session-specific message.
Show the sessions ID in the sessions list with an L suffix if logging.
Add RPC commands to toggle logging of, and disconnect, a session.
Closes#55
This accomodates large tx sends of up to 500k bytes.
However they are unlikely to be propogated by your daemon as the
default relay limit it imposes is 100K.
- incoming buffered network requests limited to 150,000 bytes
which I believe is ample for a genuine client
- if exceeded, the connection is dropped
- raise outgoing data limit for RPC connections to 5 MB - expect
sessions calls can be long and connection is implicitly trusted
- similarly raise incoming buffered data limit to 5 MB for RPC
connections
