-
Notifications
You must be signed in to change notification settings - Fork 36
/
messaging.rs
561 lines (498 loc) · 22.1 KB
/
messaging.rs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
// Copyright 2022 MaidSafe.net limited.
//
// This SAFE Network Software is licensed to you under The General Public License (GPL), version 3.
// Unless required by applicable law or agreed to in writing, the SAFE Network Software distributed
// under the GPL Licence is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. Please review the Licences for the specific language governing
// permissions and limitations relating to use of the SAFE Network Software.
use super::{MsgResponse, QueryResult, Session};
use crate::{Error, Result};
use sn_interface::{
messaging::{
data::{DataQuery, DataQueryVariant, QueryResponse},
ClientAuth, Dst, MsgId, MsgKind, WireMsg,
},
network_knowledge::supermajority,
types::{ChunkAddress, Peer},
};
use bytes::Bytes;
use rand::{rngs::OsRng, seq::SliceRandom};
use std::collections::BTreeSet;
use tokio::task::JoinSet;
use tracing::{debug, error, trace, warn};
use xor_name::XorName;
// Number of Elders subset to send queries to
pub(crate) const NUM_OF_ELDERS_SUBSET_FOR_QUERIES: usize = 3;
impl Session {
#[instrument(skip(self), level = "debug", name = "session setup conns")]
/// Make a best effort to pre connect to only relevant nodes for a set of dst addresses
/// This should reduce the number of connections attempts to the same elder set
pub(crate) async fn setup_connections_to_relevant_nodes(
&self,
dst_addresses: Vec<XorName>,
) -> Result<()> {
let mut relevant_elders = BTreeSet::new();
// TODO: get relevant nodes
for address in dst_addresses {
let (_, elders) = self.get_cmd_elders(address).await?;
for elder in elders {
let _existed = relevant_elders.insert(elder);
}
}
let mut tasks = vec![];
for peer in relevant_elders {
let session = self.clone();
let task = async move {
let connect_now = true;
// We don't retry here.. if we fail it will be retried on a per message basis
let _ = session
.peer_links
.get_or_create_link(&peer, connect_now, None)
.await;
};
tasks.push(task);
}
let _ = futures::future::join_all(tasks).await;
Ok(())
}
#[instrument(skip(self, auth, payload), level = "debug", name = "session send cmd")]
pub(crate) async fn send_cmd(
&self,
dst_address: XorName,
auth: ClientAuth,
payload: Bytes,
) -> Result<()> {
let endpoint = self.endpoint.clone();
// TODO: Consider other approach: Keep a session per section!
let (section_pk, elders) = self.get_cmd_elders(dst_address).await?;
let elders_len = elders.len();
let msg_id = MsgId::new();
debug!(
"Sending cmd w/id {msg_id:?}, from {}, to {elders_len} Elders w/ dst: {dst_address:?}",
endpoint.public_addr(),
);
let dst = Dst {
name: dst_address,
section_key: section_pk,
};
let kind = MsgKind::Client(auth);
let wire_msg = WireMsg::new_msg(msg_id, payload, kind, dst);
let send_cmd_tasks = self.send_msg(elders.clone(), wire_msg).await?;
trace!("Cmd msg {msg_id:?} sent");
// We wait for ALL the Acks get received.
// The AE messages are handled by the tasks, hence no extra wait is required.
self.we_have_sufficient_acks_for_cmd(msg_id, elders.clone(), send_cmd_tasks)
.await
}
/// Checks for acks for a given msg.
/// Returns Ok if we've sufficient to call this cmd a success
async fn we_have_sufficient_acks_for_cmd(
&self,
msg_id: MsgId,
elders: Vec<Peer>,
mut send_cmd_tasks: JoinSet<MsgResponse>,
) -> Result<()> {
debug!("----> Init of check for acks for {msg_id:?}");
let expected_acks = elders.len();
let mut received_acks = BTreeSet::default();
let mut received_errors = BTreeSet::default();
let mut failures = BTreeSet::default();
while let Some(msg_resp) = send_cmd_tasks.join_next().await {
debug!("Handling msg_resp sent to ack wait channel: {msg_resp:?}");
let (src, result) = match msg_resp {
Ok(MsgResponse::CmdResponse(src, response)) => (src, response.result().clone()),
Ok(MsgResponse::QueryResponse(src, resp)) => {
debug!("Unexpected query response received from {src:?} for {msg_id:?} when awaiting a CmdAck: {resp:?}");
let _ = received_errors.insert(src);
continue;
}
Ok(MsgResponse::Failure(src, error)) => {
debug!("Failure occurred with msg {msg_id:?} from {src:?}: {error:?}");
let _ = failures.insert(src);
continue;
}
Err(join_err) => {
warn!("Join failure occurred with msg {msg_id:?}: {join_err:?}");
continue;
}
};
match result {
Ok(()) => {
let preexisting = !received_acks.insert(src) || received_errors.contains(&src);
debug!(
"ACK from {src:?} read from set for {msg_id:?} - preexisting??: {preexisting:?}",
);
if received_acks.len() >= expected_acks {
trace!("{msg_id:?} Good! We're at or above {expected_acks} expected_acks");
return Ok(());
}
}
Err(error) => {
let _ = received_errors.insert(src);
error!(
"Received error {error:?} of cmd {msg_id:?} from {src:?}, so far {} respones and {} of them are errors",
received_acks.len() + received_errors.len(), received_errors.len()
);
// exit if too many errors:
if failures.len() + received_errors.len() >= expected_acks {
error!("Received majority of error response for cmd {msg_id:?}: {error:?}");
return Err(Error::CmdError {
source: error,
msg_id,
});
}
}
}
}
debug!("ACKs for {msg_id:?} received from: {received_acks:?}");
debug!("CmdErrors for {msg_id:?} received from: {received_errors:?}");
debug!("Failures for {msg_id:?} with: {failures:?}");
let missing_responses: Vec<Peer> = elders
.iter()
.cloned()
.filter(|p| {
let addr = &p.addr();
!received_acks.contains(addr)
&& !received_errors.contains(addr)
&& !failures.contains(addr)
})
.collect();
debug!(
"Insufficient CmdAcks returned for {msg_id:?}: {}/{expected_acks}. \
Missing Responses from: {missing_responses:?}",
received_acks.len()
);
Err(Error::InsufficientAcksReceived {
msg_id,
expected: expected_acks,
received: received_acks.len(),
})
}
#[instrument(
skip(self, auth, payload),
level = "debug",
name = "session send query"
)]
#[allow(clippy::too_many_arguments)]
/// Send a `ClientMsg` to the network awaiting for the response.
pub(crate) async fn send_query(
&self,
query: DataQuery,
auth: ClientAuth,
payload: Bytes,
dst_section_info: Option<(bls::PublicKey, Vec<Peer>)>,
) -> Result<QueryResult> {
let endpoint = self.endpoint.clone();
let chunk_addr = if let DataQueryVariant::GetChunk(address) = query.variant {
Some(address)
} else {
None
};
let dst = query.variant.dst_name();
let (section_pk, elders) = if let Some(section_info) = dst_section_info {
section_info
} else {
self.get_query_elders(dst).await?
};
let elders_len = elders.len();
let msg_id = MsgId::new();
debug!(
"Sending query message {msg_id:?}, from {}, {query:?} to \
the {elders_len} Elders closest to data name: {elders:?}",
endpoint.public_addr(),
);
let dst = Dst {
name: dst,
section_key: section_pk,
};
let kind = MsgKind::Client(auth);
let wire_msg = WireMsg::new_msg(msg_id, payload, kind, dst);
let send_query_tasks = self.send_msg(elders.clone(), wire_msg).await?;
// TODO:
// We are now simply accepting the very first valid response we receive,
// but we may want to revisit this to compare multiple responses and validate them,
// similar to what we used to do up to the following commit:
// https://github.com/maidsafe/sn_client/blob/9091a4f1f20565f25d3a8b00571cc80751918928/src/connection_manager.rs#L328
//
// For Chunk responses we already validate its hash matches the xorname requested from,
// so we don't need more than one valid response to prevent from accepting invalid responses
// from byzantine nodes, however for mutable data (non-Chunk responses) we will
// have to review the approach.
self.check_query_responses(msg_id, elders.clone(), chunk_addr, send_query_tasks)
.await
}
async fn check_query_responses(
&self,
msg_id: MsgId,
elders: Vec<Peer>,
chunk_addr: Option<ChunkAddress>,
mut send_query_tasks: JoinSet<MsgResponse>,
) -> Result<QueryResult> {
let mut discarded_responses: usize = 0;
let mut error_response = None;
let mut valid_response = None;
let elders_len = elders.len();
while let Some(msg_resp) = send_query_tasks.join_next().await {
let (peer_address, response) = match msg_resp {
Ok(MsgResponse::QueryResponse(src, resp)) => (src, resp),
Ok(MsgResponse::CmdResponse(src, resp)) => {
debug!("Unexpected cmd response received from {src:?} for {msg_id:?} when awaiting a QueryResponse: {resp:?}");
discarded_responses += 1;
continue;
}
Ok(MsgResponse::Failure(src, error)) => {
debug!("Failure occurred with msg {msg_id:?} from {src:?}: {error:?}");
discarded_responses += 1;
continue;
}
Err(join_err) => {
warn!("Join failure occurred with msg {msg_id:?}: {join_err:?}");
continue;
}
};
// let's see if we have a positive response...
debug!("Response to {msg_id:?}: {response:?}");
match *response {
QueryResponse::GetChunk(Ok(chunk)) => {
if let Some(chunk_addr) = chunk_addr {
// We are dealing with Chunk query responses, thus we validate its hash
// matches its xorname, if so, we don't need to await for more responses
debug!("Chunk QueryResponse received is: {chunk:#?}");
if chunk_addr.name() == chunk.name() {
trace!("Valid Chunk received for {msg_id:?}");
valid_response = Some(QueryResponse::GetChunk(Ok(chunk)));
break;
} else {
// the Chunk content doesn't match its XorName,
// this is suspicious and it could be a byzantine node
warn!("We received an invalid Chunk response from one of the nodes for {msg_id:?}");
discarded_responses += 1;
}
}
}
QueryResponse::GetRegister(Err(_))
| QueryResponse::ReadRegister(Err(_))
| QueryResponse::GetRegisterPolicy(Err(_))
| QueryResponse::GetRegisterOwner(Err(_))
| QueryResponse::GetRegisterUserPermissions(Err(_))
| QueryResponse::GetChunk(Err(_)) => {
debug!(
"QueryResponse error #{discarded_responses} for {msg_id:?} received \
from {peer_address:?} (but may be overridden by a non-error response \
from another elder): {:#?}",
&response
);
error_response = Some(*response);
discarded_responses += 1;
}
QueryResponse::GetRegister(Ok(ref register)) => {
debug!("okay got register from {peer_address:?}");
// TODO: properly merge all registers
if let Some(QueryResponse::GetRegister(Ok(prior_response))) = &valid_response {
if register.size() > prior_response.size() {
debug!("longer register");
// keep this new register
valid_response = Some(*response);
}
} else {
valid_response = Some(*response);
}
}
QueryResponse::ReadRegister(Ok(_)) => {
debug!("okay _read_ register from {peer_address:?}");
if valid_response.is_none() {
valid_response = Some(*response);
}
}
QueryResponse::SpentProofShares(Ok(ref spentproof_set)) => {
debug!("okay _read_ spentproofs from {peer_address:?}");
// TODO: properly merge all registers
if let Some(QueryResponse::SpentProofShares(Ok(prior_response))) =
&valid_response
{
if spentproof_set.len() > prior_response.len() {
debug!("longer spentproof response retrieved");
// keep this new register
valid_response = Some(*response);
}
} else {
valid_response = Some(*response);
}
}
response => {
// we got a valid response
valid_response = Some(response)
}
}
}
// we've looped over all responses...
// if any are valid, lets return it
if let Some(response) = valid_response {
debug!("Valid response in!!!: {response:?}");
return Ok(QueryResult { response });
// otherwise, if we've got an error in
// we can return that too
} else if let Some(response) = error_response {
if discarded_responses > elders_len / 2 {
return Ok(QueryResult { response });
}
}
Err(Error::NoResponse {
msg_id,
peers: elders,
})
}
/// Get DataSection elders details. Resort to own section if DataSection is not available.
/// Takes a random subset (NUM_OF_ELDERS_SUBSET_FOR_QUERIES) of the avialable elders as targets
pub(crate) async fn get_query_elders(
&self,
dst: XorName,
) -> Result<(bls::PublicKey, Vec<Peer>)> {
let sap = self.network.read().await.closest(&dst, None).cloned();
let (section_pk, mut elders) = if let Some(sap) = &sap {
(sap.section_key(), sap.elders_vec())
} else {
return Err(Error::NoNetworkKnowledge(dst));
};
elders.shuffle(&mut OsRng);
// We select the NUM_OF_ELDERS_SUBSET_FOR_QUERIES closest Elders we are querying
let elders: Vec<_> = elders
.into_iter()
.take(NUM_OF_ELDERS_SUBSET_FOR_QUERIES)
.collect();
let elders_len = elders.len();
if elders_len < NUM_OF_ELDERS_SUBSET_FOR_QUERIES && elders_len > 1 {
return Err(Error::InsufficientElderConnections {
connections: elders_len,
required: NUM_OF_ELDERS_SUBSET_FOR_QUERIES,
});
}
Ok((section_pk, elders))
}
async fn get_cmd_elders(&self, dst_address: XorName) -> Result<(bls::PublicKey, Vec<Peer>)> {
let a_close_sap = self
.network
.read()
.await
.closest(&dst_address, None)
.cloned();
// Get DataSection elders details.
if let Some(sap) = a_close_sap {
let sap_elders = sap.elders_vec();
let section_pk = sap.section_key();
trace!("SAP elders found {sap_elders:?}");
// Supermajority of elders is expected.
let targets_count = supermajority(sap_elders.len());
// any SAP that does not hold elders_count() is indicative of a broken network (after genesis)
if sap_elders.len() < targets_count {
error!(
"Insufficient knowledge to send to address {dst_address:?}, \
elders for this section: {sap_elders:?} ({targets_count} needed), \
section PK is: {section_pk:?}"
);
return Err(Error::InsufficientElderKnowledge {
connections: sap_elders.len(),
required: targets_count,
section_pk,
});
}
Ok((section_pk, sap_elders))
} else {
Err(Error::NoNetworkKnowledge(dst_address))
}
}
#[instrument(skip_all, level = "trace")]
pub(super) async fn send_msg(
&self,
nodes: Vec<Peer>,
wire_msg: WireMsg,
) -> Result<JoinSet<MsgResponse>> {
let msg_id = wire_msg.msg_id();
debug!("---> Send msg {msg_id:?} going out.");
let bytes = wire_msg.serialize()?;
let mut tasks = JoinSet::new();
for (peer_index, peer) in nodes.into_iter().enumerate() {
let session = self.clone();
let bytes = bytes.clone();
let _abort_handle = tasks.spawn(async move {
let mut connect_now = false;
debug!("Trying to send msg {msg_id:?} to {peer:?}");
loop {
let link = session
.peer_links
.get_or_create_link(&peer, connect_now, Some(msg_id))
.await;
match link.send_bi(bytes.clone(), msg_id).await {
Ok(recv_stream) => {
debug!(
"That's {msg_id:?} sent to {peer:?}... starting receive listener"
);
// let's listen for responses on the bi-stream
break session
.recv_stream_listener(msg_id, peer, peer_index, recv_stream)
.await;
}
Err(error) if !connect_now => {
// Let's retry (only once) to reconnect to this peer and send the msg.
error!(
"Failed to send {msg_id:?} to {peer:?} on a new \
bi-stream: {error:?}. Creating a new connection to retry once ..."
);
session.peer_links.remove_link_from_peer_links(&peer).await;
connect_now = true;
continue;
}
Err(error) => {
error!("Error sending {msg_id:?} bidi to {peer:?}: {error:?}");
session.peer_links.remove_link_from_peer_links(&peer).await;
break MsgResponse::Failure(
peer.addr(),
Error::FailedToInitateBiDiStream { msg_id, error },
);
}
}
}
});
}
Ok(tasks)
}
}
#[cfg(test)]
mod tests {
use super::*;
use sn_interface::{
network_knowledge::SectionTree,
test_utils::{prefix, TestKeys, TestSapBuilder},
};
use eyre::Result;
use qp2p::Config;
use std::net::{Ipv4Addr, SocketAddr};
use xor_name::Prefix;
fn new_network_network_contacts() -> (SectionTree, bls::SecretKey, bls::PublicKey) {
let (genesis_sap, genesis_sk_set, ..) = TestSapBuilder::new(Prefix::default()).build();
let genesis_sk = genesis_sk_set.secret_key();
let genesis_pk = genesis_sk.public_key();
let genesis_sap = TestKeys::get_section_signed(&genesis_sk, genesis_sap);
let tree = SectionTree::new(genesis_sap).expect("SAP belongs to the genesis prefix");
(tree, genesis_sk, genesis_pk)
}
#[tokio::test(flavor = "multi_thread")]
async fn cmd_sent_to_all_elders() -> Result<()> {
let elders_len = 5;
let prefix = prefix("0");
let (sap, secret_key_set, ..) = TestSapBuilder::new(prefix).elder_count(elders_len).build();
let sap0 = TestKeys::get_section_signed(&secret_key_set.secret_key(), sap);
let (mut network_contacts, _genesis_sk, _) = new_network_network_contacts();
assert!(network_contacts.insert_without_chain(sap0));
let session = Session::new(
Config::default(),
SocketAddr::from((Ipv4Addr::UNSPECIFIED, 0)),
network_contacts,
)?;
let mut rng = rand::thread_rng();
let result = session.get_cmd_elders(XorName::random(&mut rng)).await?;
assert_eq!(result.0, secret_key_set.public_keys().public_key());
assert_eq!(result.1.len(), elders_len);
Ok(())
}
}