Files
firmware/src/mesh/Router.cpp
T
Jonathan BennettandClaude Opus 5 d3b4b343e7 Send an ack over PKC when no channel can carry it (#11891)
* Send an ack over PKC when no channel can carry it

PKI needs only the two keys, so a DM can reach us over a channel we do not
carry. Its ack is a ROUTING packet, which wouldEncryptWithPKC() excludes, so
today it is channel-encoded, fails at setActiveByIndex() with NO_CHANNEL, and is
never sent. The sender sees nothing and retransmits to exhaustion for a message
that was in fact delivered.

Fall back to PKC for exactly that case. This is the one place an ack is
deliberately made opaque to relays; normally that costs next-hop learning and
intermediate retransmission cancel, which is why ROUTING is PKC-excluded in
general, but here there is no readable alternative to lose, because without this
the ack does not exist.

The predicate is scoped as tightly as that argument reaches: a unicast ROUTING
packet we originate, carrying a request_id, to a destination whose key we hold,
under the same ham/sim/private-key preconditions PKC always has, and only when
the channel index does not resolve. It tests channels.getHash() rather than
setActiveByIndex() so it has no side effect; generateHash already returns -1 for
an invalid key, so the two agree on which indexes are unusable.

Four cases in test_packet_signing pin the corners: the fallback fires, it does
not paper over an ack with no destination key, it does not catch a non-ack on
the same unusable channel, and an ack on a channel that does resolve still goes
out readable.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012hLcVif8GDEmA2k77hmFG8

* Short-circuit the fallback so an out-of-range channel logs no error

wouldEncryptWithPKC() reaches channels.getName(chIndex) before its portnum
exclusion, and getByIndex() logs "Invalid channel index" on the way past. With
the general predicate tested first, an ack on an out-of-range index printed that
error and then went on to encode successfully. Test ackFallback first so the
case that is about to succeed never asks.

Also record why the range check leads inside the predicate: getHash() is a bare
hashes[i] with no bounds test of its own.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012hLcVif8GDEmA2k77hmFG8

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-09-20 05:21:06 +00:00

1776 lines
80 KiB
C++

#include "Router.h"
#include "Channels.h"
#include "CryptoEngine.h"
#include "MeshRadio.h"
#include "MeshService.h"
#include "NodeDB.h"
#include "PositionPrecision.h"
#include "UptimeClock.h"
#include "gps/RTC.h"
#include "configuration.h"
#include "main.h"
#include "mesh-pb-constants.h"
#include "meshUtils.h"
#include "modules/RoutingModule.h"
#include <ErriezCRC32.h>
#include <pb_decode.h>
#include <pb_encode.h>
#if USERPREFS_BLOCK_POSITION_ON_EVENT_CHANNEL && !MESHTASTIC_EXCLUDE_GPS
#include "modules/PositionModule.h"
#endif
#if HAS_TRAFFIC_MANAGEMENT
#endif
#if HAS_VARIABLE_HOPS
#include "modules/HopScalingModule.h"
#endif
#if !MESHTASTIC_EXCLUDE_MQTT
#include "mqtt/MQTT.h"
#endif
#include "Default.h"
#if ARCH_PORTDUINO
#include "Throttle.h"
#include "platform/portduino/PortduinoGlue.h"
#include "serialization/MeshPacketSerializer.h"
#endif
// The size checks below budget for the tag that encryptPacketCCM actually appends, so the
// two constants must not drift apart.
static_assert(MESHTASTIC_AEAD_OVERHEAD == CryptoEngine::AEAD_TAG_SIZE,
"MESHTASTIC_AEAD_OVERHEAD must match CryptoEngine::AEAD_TAG_SIZE");
#define MAX_RX_FROMRADIO \
4 // max number of packets destined to our queue, we dispatch packets quickly so it doesn't need to be big
// I think this is right, one packet for each of the three fifos + one packet being currently assembled for TX or RX
// And every TX packet might have a retransmission packet or an ack alive at any moment
#ifdef ARCH_PORTDUINO
// Portduino (native) targets can use dynamic memory pools with runtime-configurable sizes
#define MAX_PACKETS \
(MAX_RX_TOPHONE + MAX_RX_FROMRADIO + 2 * MAX_TX_QUEUE + \
2) // max number of packets which can be in flight (either queued from reception or queued for sending)
// Live in-flight packet bytes are tracked under "pktpool(live)" in the MemAudit breakdown
static MemoryDynamic<meshtastic_MeshPacket> dynamicPool("pktpool(live)");
Allocator<meshtastic_MeshPacket> &packetPool = dynamicPool;
#elif defined(ARCH_STM32WL) || defined(BOARD_HAS_PSRAM)
// On STM32 and boards with PSRAM, there isn't enough heap left over for the rest of the firmware if we allocate this statically.
// For now, make it dynamic again.
#define MAX_PACKETS \
(MAX_RX_TOPHONE + MAX_RX_FROMRADIO + 2 * MAX_TX_QUEUE + \
2) // max number of packets which can be in flight (either queued from reception or queued for sending)
// Live in-flight packet bytes are tracked under "pktpool(live)" in the MemAudit breakdown
static MemoryDynamic<meshtastic_MeshPacket> dynamicPool("pktpool(live)");
Allocator<meshtastic_MeshPacket> &packetPool = dynamicPool;
#else
// Embedded targets use static memory pools with compile-time constants
#define MAX_PACKETS_STATIC \
(MAX_RX_TOPHONE + MAX_RX_FROMRADIO + 2 * MAX_TX_QUEUE + \
2) // max number of packets which can be in flight (either queued from reception or queued for sending)
// Static pool RAM is BSS, not heap; "pktpool(live)" still shows in-flight packet bytes
static MemoryPool<meshtastic_MeshPacket, MAX_PACKETS_STATIC> staticPool("pktpool(live)");
Allocator<meshtastic_MeshPacket> &packetPool = staticPool;
#endif
static uint8_t bytes[MAX_LORA_PAYLOAD_LEN + 1] __attribute__((__aligned__));
static ChannelIndex getEffectiveChannelIndex(const meshtastic_MeshPacket *p)
{
ChannelIndex chIndex = p->channel;
if (nodeDB && isFromUs(p) && !chIndex && !p->pki_encrypted && !isBroadcast(p->to)) {
const meshtastic_NodeInfoLite *node = nodeDB->getMeshNode(p->to);
if (node)
chIndex = node->channel;
}
return chIndex;
}
bool isBlockedEventCoordinatePacket(const meshtastic_MeshPacket *p)
{
#if USERPREFS_BLOCK_POSITION_ON_EVENT_CHANNEL
if (p->pki_encrypted || willUsePki(p)) {
return false;
}
// From us, to us: never leaves the device (sendLocal delivers it locally). This is how the phone
// hands a GPS-less node its fix and time, so it shares nothing and must not be blocked.
if (isFromUs(p) && isToUs(p)) {
return false;
}
if (p->which_payload_variant == meshtastic_MeshPacket_decoded_tag) {
return isCoordinatePortnum(p->decoded.portnum) && channels.isEventChannel(getEffectiveChannelIndex(p));
}
return false;
#else
(void)p;
return false;
#endif
}
#if USERPREFS_BLOCK_POSITION_ON_EVENT_CHANNEL && !MESHTASTIC_EXCLUDE_GPS
// A remote node's unicast position request to us. Only the reply is generated for these; the packet
// itself is still dropped by the caller.
static bool isEventChannelPositionRequestForUs(const meshtastic_MeshPacket *p)
{
return p->which_payload_variant == meshtastic_MeshPacket_decoded_tag &&
p->decoded.portnum == meshtastic_PortNum_POSITION_APP && p->decoded.want_response && isToUs(p) && !isFromUs(p);
}
#endif
bool coerceCoordinatePacketToPositionChannel(meshtastic_MeshPacket *p)
{
#if USERPREFS_BLOCK_POSITION_ON_EVENT_CHANNEL
if (!isBlockedEventCoordinatePacket(p))
return false;
uint8_t positionChannel;
if (!findPositionChannel(positionChannel))
return false;
LOG_DEBUG("Coerce coordinate packet 0x%08x from event channel to position channel %u", p->id, positionChannel);
p->channel = positionChannel;
return true;
#else
(void)p;
return false;
#endif
}
bool willUsePki(const meshtastic_MeshPacket *p)
{
#if !(MESHTASTIC_EXCLUDE_PKI)
if (p->which_payload_variant != meshtastic_MeshPacket_decoded_tag || !isFromUs(p))
return false;
bool haveDestKey = false;
if (p->decoded.portnum == meshtastic_PortNum_KEY_VERIFICATION_APP) {
meshtastic_NodeInfoLite_public_key_t destKey = {0, {0}};
haveDestKey = nodeDB->copyPublicKey(p->to, destKey);
if (!haveDestKey && p->pki_encrypted)
haveDestKey = crypto->getPendingPublicKey(p->to, destKey);
}
return wouldEncryptWithPKC(p, getEffectiveChannelIndex(p), haveDestKey);
#else
(void)p;
return false;
#endif
}
struct RoutingAuthCache {
bool valid = false;
// Deliberately NOT initialized in-class as this eats flash space.
meshtastic_Config_SecurityConfig_PacketSignaturePolicy policy;
meshtastic_MeshPacket wire = meshtastic_MeshPacket_init_zero;
meshtastic_MeshPacket authenticated = meshtastic_MeshPacket_init_zero;
};
static RoutingAuthCache routingAuthCache;
static concurrency::Lock *routingAuthCacheLock;
static uint32_t routingAuthEvaluations;
static bool routingAuthCacheMatches(const meshtastic_MeshPacket &packet)
{
if (!routingAuthCacheLock)
return false;
concurrency::LockGuard guard(routingAuthCacheLock);
if (!routingAuthCache.valid)
return false;
if (routingAuthCache.policy != config.security.packet_signature_policy ||
memcmp(&routingAuthCache.wire, &packet, sizeof(packet)) != 0) {
routingAuthCache.valid = false;
return false;
}
return true;
}
static void storeRoutingAuthCache(const meshtastic_MeshPacket &wire, const meshtastic_MeshPacket &authenticated)
{
concurrency::LockGuard guard(routingAuthCacheLock);
routingAuthCache.wire = wire;
routingAuthCache.authenticated = authenticated;
routingAuthCache.policy = config.security.packet_signature_policy;
routingAuthCache.valid = true;
}
static bool applyRoutingAuthCache(meshtastic_MeshPacket *packet)
{
if (!routingAuthCacheLock)
return false;
concurrency::LockGuard guard(routingAuthCacheLock);
if (!routingAuthCache.valid || routingAuthCache.policy != config.security.packet_signature_policy ||
memcmp(&routingAuthCache.wire, packet, sizeof(*packet)) != 0) {
routingAuthCache.valid = false;
return false;
}
*packet = routingAuthCache.authenticated;
routingAuthCache.valid = false;
return true;
}
static void clearRoutingAuthCache()
{
if (!routingAuthCacheLock)
return;
concurrency::LockGuard guard(routingAuthCacheLock);
routingAuthCache.valid = false;
}
#ifdef PIO_UNIT_TESTING
uint32_t routingAuthEvaluationCount()
{
return routingAuthEvaluations;
}
void resetRoutingAuthEvaluationCount()
{
routingAuthEvaluations = 0;
if (routingAuthCacheLock) {
concurrency::LockGuard guard(routingAuthCacheLock);
routingAuthCache.valid = false;
}
}
#endif
/**
* Constructor
*
* Currently we only allow one interface, that may change in the future
*/
Router::Router() : concurrency::OSThread("Router"), fromRadioQueue(MAX_RX_FROMRADIO)
{
// This is called pre main(), don't touch anything here, the following code is not safe
/* LOG_DEBUG("Size of NodeInfo %d", sizeof(NodeInfo));
LOG_DEBUG("Size of SubPacket %d", sizeof(SubPacket));
LOG_DEBUG("Size of MeshPacket %d", sizeof(MeshPacket)); */
fromRadioQueue.setReader(this);
// init Lockguard for crypt operations
assert(!cryptLock);
cryptLock = new concurrency::Lock();
if (!routingAuthCacheLock)
routingAuthCacheLock = new concurrency::Lock();
// Runtime default for the auth-cache snapshot policy. Keep it here, saves flash.
routingAuthCache.policy = meshtastic_Config_SecurityConfig_PacketSignaturePolicy_PACKET_SIGNATURE_POLICY_BALANCED;
}
bool Router::shouldDecrementHopLimit(const meshtastic_MeshPacket *p)
{
// First hop MUST always decrement to prevent retry issues
if (getHopsAway(*p) == 0) {
return true; // Always decrement on first hop
}
// Check if both local device and previous relay are routers (including CLIENT_BASE)
bool localIsRouter =
IS_ONE_OF(config.device.role, meshtastic_Config_DeviceConfig_Role_ROUTER, meshtastic_Config_DeviceConfig_Role_ROUTER_LATE,
meshtastic_Config_DeviceConfig_Role_CLIENT_BASE);
// If local device isn't a router, always decrement
if (!localIsRouter) {
return true;
}
// router_preserve_hops: not suitable right now - removed from config until
// the right heuristics for when to preserve vs. exhaust hops are established.
// #if HAS_TRAFFIC_MANAGEMENT
// if (moduleConfig.has_traffic_management &&
// moduleConfig.traffic_management.router_preserve_hops && ...) { ... }
// #endif
// For subsequent hops, preserve hop_limit only when the previous relay is UNAMBIGUOUSLY a favorite
// router. The relay_node byte is just the last byte of a 32-bit node number, so on a dense mesh it
// collides; the old "first matching node wins" scan could preserve hops for the wrong node
// (non-deterministic, depends on NodeDB order). resolveLastByte() reports a collision instead, and
// we re-check the favorite/router predicate on the single resolved node. On ambiguity/none we
// decrement (the safe default).
NodeNum resolved = 0;
if (nodeDB->resolveUniqueLastByte(p->relay_node, /*requireDirectNeighbor=*/false, &resolved)) {
const meshtastic_NodeInfoLite *node = nodeDB->getMeshNode(resolved);
if (node && nodeInfoLiteIsFavorite(node) && nodeInfoLiteHasUser(node) &&
IS_ONE_OF(node->role, meshtastic_Config_DeviceConfig_Role_ROUTER, meshtastic_Config_DeviceConfig_Role_ROUTER_LATE,
meshtastic_Config_DeviceConfig_Role_CLIENT_BASE)) {
LOG_DEBUG("Unique favorite relay router 0x%08x from last byte 0x%x", resolved, p->relay_node);
return false; // Don't decrement hop_limit
}
}
// No unambiguous favorite router match found, decrement hop_limit
return true;
}
/**
* do idle processing
* Mostly looking in our incoming rxPacket queue and calling handleReceived.
*/
int32_t Router::runOnce()
{
meshtastic_MeshPacket *mp;
while ((mp = fromRadioQueue.dequeuePtr(0)) != NULL) {
// printPacket("handle fromRadioQ", mp);
perhapsHandleReceived(mp);
}
// LOG_DEBUG("Sleep forever");
return INT32_MAX; // Wait a long time - until we get woken for the message queue
}
/**
* RadioInterface calls this to queue up packets that have been received from the radio. The router is now responsible for
* freeing the packet
*/
void Router::enqueueReceivedMessage(meshtastic_MeshPacket *p)
{
// Try enqueue until successful
while (!fromRadioQueue.enqueue(p, 0)) {
meshtastic_MeshPacket *old_p;
old_p = fromRadioQueue.dequeuePtr(0); // Dequeue and discard the oldest packet
if (old_p) {
printPacket("fromRadioQ full, drop oldest!", old_p);
packetPool.release(old_p);
}
}
// Nasty hack because our threading is primitive. interfaces shouldn't need to know about routers FIXME
setReceivedMessage();
}
/// Generate a unique packet id
// FIXME, move this someplace better
PacketId generatePacketId()
{
static uint32_t rollingPacketId; // Note: trying to keep this in noinit didn't help for working across reboots
static bool didInit = false;
if (!didInit) {
didInit = true;
// pick a random initial sequence number at boot (to prevent repeated reboots always starting at 0)
// Note: we mask the high order bit to ensure that we never pass a 'negative' number to random
rollingPacketId = random(UINT32_MAX & 0x7fffffff);
LOG_DEBUG("Initial packet id %u", rollingPacketId);
}
rollingPacketId++;
rollingPacketId &= ID_COUNTER_MASK; // Mask out the top 22 bits
PacketId id = rollingPacketId | random(UINT32_MAX & 0x7fffffff) << 10; // top 22 bits
LOG_TRACE("Partially randomized packet id 0x%08x", id);
return id;
}
RxTimeStamp computeRxTimeStamp()
{
const bool haveTime = getRTCQuality() >= RTCQualityFromNet;
return {haveTime ? getValidTime(RTCQualityFromNet) : Time::getUptimeSecs(), haveTime};
}
void stampRxTime(meshtastic_MeshPacket *p)
{
const RxTimeStamp ts = computeRxTimeStamp();
p->rx_time = ts.time;
p->has_rx_time = ts.valid;
}
meshtastic_MeshPacket *Router::allocForSending()
{
meshtastic_MeshPacket *p = packetPool.allocZeroed();
if (!p)
return nullptr;
p->which_payload_variant = meshtastic_MeshPacket_decoded_tag; // Assume payload is decoded at start.
p->from = nodeDB->getNodeNum();
p->to = NODENUM_BROADCAST;
p->hop_limit = Default::getConfiguredOrDefaultHopLimit(config.lora.hop_limit);
p->id = generatePacketId();
// Just in case we process the packet locally - make sure it has a timestamp.
stampRxTime(p);
return p;
}
/**
* Send an ack or a nak packet back towards whoever sent idFrom
*/
void Router::sendAckNak(meshtastic_Routing_Error err, NodeNum to, PacketId idFrom, ChannelIndex chIndex, uint8_t hopLimit,
bool ackWantsAck, const meshtastic_MeshPacket *relaySource)
{
routingModule->sendAckNak(err, to, idFrom, chIndex, hopLimit, ackWantsAck, relaySource);
}
void Router::abortSendAndNak(meshtastic_Routing_Error err, meshtastic_MeshPacket *p)
{
LOG_ERROR("Error=%d, return NAK and drop packet", err);
sendAckNak(err, getFrom(p), p->id, p->channel);
packetPool.release(p);
}
void Router::setReceivedMessage()
{
// LOG_DEBUG("set interval to ASAP");
setInterval(0); // Run ASAP, so we can figure out our correct sleep time
runASAP = true;
}
meshtastic_QueueStatus Router::getQueueStatus()
{
if (!iface) {
meshtastic_QueueStatus qs;
qs.res = qs.mesh_packet_id = qs.free = qs.maxlen = 0;
return qs;
} else
return iface->getQueueStatus();
}
ErrorCode Router::sendLocal(meshtastic_MeshPacket *p, RxSource src)
{
if (p->to == 0) {
LOG_ERROR("Packet received with to=0");
}
// No need to deliver externally if the destination is the local node
if (isToUs(p)) {
printPacket("Enqueued local", p);
// Preserve the trusted origin explicitly. Queueing used to erase src and make a local
// phone/module packet indistinguishable from remote already-decoded ingress.
deliverLocal(p, src);
return ERRNO_SHOULD_RELEASE;
} else if (!iface) {
// We must be sending to remote nodes also, fail if no interface found
abortSendAndNak(meshtastic_Routing_Error_NO_INTERFACE, p);
return ERRNO_NO_INTERFACES;
} else {
// Coordinates never go out on the event channel: any local originator (phone, module, UI) that aimed
// one there is moved onto the position channel instead. Before the loopback below so the local copy
// carries the channel it will actually be sent on.
coerceCoordinatePacketToPositionChannel(p);
// If we are sending a broadcast, we also treat it as if we just received it ourself
// this allows local apps (and PCs) to see broadcasts sourced locally. Only the loopback
// handleReceived is deferred when nested; send(p) below still transmits immediately.
if (isBroadcast(p->to)) {
deliverLocal(p, src);
}
// don't override if a channel was requested and no need to set it when PKI is enforced
if (!p->channel && !p->pki_encrypted && !isBroadcast(p->to)) {
ChannelIndex chIndex = getEffectiveChannelIndex(p);
if (chIndex) {
p->channel = chIndex;
LOG_TRACE("localSend to channel %d", p->channel);
}
}
// If someone asks for acks on broadcast, we need the hop limit to be at least one, so that first node that receives our
// message will rebroadcast. But asking for hop_limit 0 in that context means the client app has no preference on hop
// counts and we want this message to get through the whole mesh, so use the default.
if (src == RX_SRC_USER && p->want_ack && p->hop_limit == 0) {
p->hop_limit = Default::getConfiguredOrDefaultHopLimit(config.lora.hop_limit);
}
return send(p);
}
}
/**
* Send a packet on a suitable interface. This routine will
* later free() the packet to pool. This routine is not allowed to stall.
* If the txmit queue is full it might return an error.
*/
ErrorCode Router::send(meshtastic_MeshPacket *p)
{
if (isToUs(p)) {
LOG_ERROR("BUG! send() with packet for local node");
packetPool.release(p);
return meshtastic_Routing_Error_BAD_REQUEST;
} // should have already been handled by sendLocal
// Abort sending if we are violating the duty cycle
float effectiveDutyCycle = getEffectiveDutyCycle();
if (!config.lora.override_duty_cycle && effectiveDutyCycle < 100) {
float hourlyTxPercent = airTime->utilizationTXPercent();
if (hourlyTxPercent > effectiveDutyCycle) {
uint8_t silentMinutes = airTime->getSilentMinutes(hourlyTxPercent, effectiveDutyCycle);
LOG_WARN("Duty cycle limit exceeded, abort send, retry in %d mins", silentMinutes);
meshtastic_ClientNotification *cn = clientNotificationPool.allocZeroed();
if (cn) {
cn->has_reply_id = true;
cn->reply_id = p->id;
cn->level = meshtastic_LogRecord_Level_WARNING;
cn->time = getValidTime(RTCQualityFromNet);
snprintf(cn->message, sizeof(cn->message), "Duty cycle limit exceeded. You can send again in %d mins",
silentMinutes);
service->sendClientNotification(cn);
}
meshtastic_Routing_Error err = meshtastic_Routing_Error_DUTY_CYCLE_LIMIT;
if (isFromUs(p)) { // only send NAK to API, not to the mesh
abortSendAndNak(err, p);
} else {
packetPool.release(p);
}
return err;
}
}
// PacketId nakId = p->decoded.which_ackVariant == SubPacket_fail_id_tag ? p->decoded.ackVariant.fail_id : 0;
// assert(!nakId); // I don't think we ever send 0hop naks over the wire (other than to the phone), test that assumption with
// assert
// Never set the want_ack flag on broadcast packets sent over the air.
if (isBroadcast(p->to))
p->want_ack = false;
// Up until this point we might have been using 0 for the from address (if it started with the phone), but when we send over
// the lora we need to make sure we have replaced it with our local address
p->from = getFrom(p);
p->relay_node = nodeDB->getLastByteOfNodeNum(getNodeNum()); // set the relayer to us
#if HAS_VARIABLE_HOPS
// Apply HopScaling hop recommendation to routine outgoing broadcasts
if (isFromUs(p) && isBroadcast(p->to) && hopScalingModule && p->which_payload_variant == meshtastic_MeshPacket_decoded_tag) {
switch (p->decoded.portnum) {
case meshtastic_PortNum_POSITION_APP:
case meshtastic_PortNum_TELEMETRY_APP:
case meshtastic_PortNum_NODEINFO_APP:
case meshtastic_PortNum_NEIGHBORINFO_APP: {
uint8_t variableHopLimit = hopScalingModule->getLastRequiredHop();
// Never exceed user-configured hop_limit
if (variableHopLimit < p->hop_limit) {
p->hop_limit = variableHopLimit;
}
break;
}
default:
break;
}
}
#endif
// If we are the original transmitter, set the hop limit with which we start
if (isFromUs(p))
p->hop_start = p->hop_limit;
// If the packet hasn't yet been encrypted, do so now (it might already be encrypted if we are just forwarding it)
if (!(p->which_payload_variant == meshtastic_MeshPacket_encrypted_tag ||
p->which_payload_variant == meshtastic_MeshPacket_decoded_tag)) {
// Error returns from here own the packet, as the position-precision path below does.
packetPool.release(p);
return meshtastic_Routing_Error_BAD_REQUEST;
}
fixPriority(p); // Before encryption, fix the priority if it's unset
// Position precision is an originator-only privacy policy. Relays keep
// p->from as the original sender, so do not rewrite their POSITION_APP payload.
if (isBlockedEventCoordinatePacket(p)) {
LOG_DEBUG("Suppress coordinate send on event (everyone) channel");
packetPool.release(p);
return meshtastic_Routing_Error_NOT_AUTHORIZED;
}
if (isFromUs(p)) {
if (!applyPositionPrecisionForChannel(*p, p->channel)) {
LOG_ERROR("Drop malformed position packet before send");
packetPool.release(p);
return meshtastic_Routing_Error_BAD_REQUEST;
}
}
// If the packet is not yet encrypted, do so now
if (p->which_payload_variant == meshtastic_MeshPacket_decoded_tag) {
ChannelIndex chIndex = p->channel; // keep as a local because we are about to change it
DEBUG_HEAP_BEFORE;
meshtastic_MeshPacket *p_decoded = packetPool.allocCopy(*p);
DEBUG_HEAP_AFTER("Router::send", p_decoded);
auto encodeResult = perhapsEncode(p);
if (encodeResult != meshtastic_Routing_Error_NONE) {
packetPool.release(p_decoded);
p->channel = 0; // Reset the channel to 0, so we don't use the failing hash again
abortSendAndNak(encodeResult, p);
return encodeResult; // FIXME - this isn't a valid ErrorCode
}
#if !MESHTASTIC_EXCLUDE_MQTT
// Only publish to MQTT if we're the original transmitter of the packet
if (moduleConfig.mqtt.enabled && isFromUs(p) && mqtt && p_decoded) {
mqtt->onSend(*p, *p_decoded, chIndex);
}
#endif
packetPool.release(p_decoded);
}
#if HAS_UDP_MULTICAST
if (udpHandler && config.network.enabled_protocols & meshtastic_Config_NetworkConfig_ProtocolFlags_UDP_BROADCAST) {
udpHandler->onSend(const_cast<meshtastic_MeshPacket *>(p));
}
#endif
// Only already-encrypted frames (relayed, phone-sourced) reach here oversized; perhapsEncode()
// bounds everything it encodes. No NAK: p->channel is a wire hash by now, not an index.
if (p->encrypted.size > MAX_RADIO_PAYLOAD_LEN) {
LOG_WARN("Drop 0x%08x: payload %u exceeds radio capacity %u", p->id, (unsigned)p->encrypted.size,
(unsigned)MAX_RADIO_PAYLOAD_LEN);
packetPool.release(p);
return meshtastic_Routing_Error_TOO_LARGE;
}
assert(iface); // This should have been detected already in sendLocal (or we just received a packet from outside)
return iface->send(p);
}
/** Attempt to cancel a previously sent packet. Returns true if a packet was found we could cancel */
bool Router::cancelSending(NodeNum from, PacketId id)
{
if (iface && iface->cancelSending(from, id)) {
// We are not a relayer of this packet anymore
removeRelayer(nodeDB->getLastByteOfNodeNum(nodeDB->getNodeNum()), id, from);
return true;
}
return false;
}
/** Attempt to find a packet in the TxQueue. Returns true if the packet was found. */
bool Router::findInTxQueue(NodeNum from, PacketId id)
{
return iface->findInTxQueue(from, id);
}
/**
* Every (non duplicate) packet this node receives will be passed through this method. This allows subclasses to
* update routing tables etc... based on what we overhear (even for messages not destined to our node)
*/
void Router::sniffReceived(const meshtastic_MeshPacket *p, const meshtastic_Routing *c)
{
// FIXME, update nodedb here for any packet that passes through us
}
#if !(MESHTASTIC_EXCLUDE_PKI) && !(MESHTASTIC_EXCLUDE_XEDDSA)
/** Size a decoded Data as the sender's signedDataFits() gate would have, with padding stripped:
* unknown fields inside Data.payload survive in payload.size and would otherwise let a forger
* inflate an unsigned broadcast past the signable budget. Returns false only if sizing failed.
* Sizing only what this build's schema decodes, so a signable type that later grows needs its
* legitimate maximum re-checked against the budget or honest unsigned broadcasts get dropped. */
static bool canonicalSignableSize(meshtastic_Data *d, size_t *size)
{
const pb_msgdesc_t *fields = nullptr;
switch (d->portnum) {
case meshtastic_PortNum_POSITION_APP:
fields = &meshtastic_Position_msg;
break;
case meshtastic_PortNum_TELEMETRY_APP:
fields = &meshtastic_Telemetry_msg;
break;
case meshtastic_PortNum_WAYPOINT_APP:
fields = &meshtastic_Waypoint_msg;
break;
case meshtastic_PortNum_NODEINFO_APP:
fields = &meshtastic_User_msg;
break;
default:
break;
}
if (fields) {
// Scratch kept off the stack: these decoded structs are large for the smaller MCU targets.
// Safe as file-static state because both callers of checkXeddsaReceivePolicy hold cryptLock.
static union {
// cppcheck-suppress unusedStructMember ; written by pb_decode through &inner
meshtastic_Position position;
// cppcheck-suppress unusedStructMember ; written by pb_decode through &inner
meshtastic_Telemetry telemetry;
// cppcheck-suppress unusedStructMember ; written by pb_decode through &inner
meshtastic_Waypoint waypoint;
// cppcheck-suppress unusedStructMember ; written by pb_decode through &inner
meshtastic_User user;
} inner;
memset(&inner, 0, sizeof(inner));
size_t canonicalPayload;
if (pb_decode_from_bytes(d->payload.bytes, d->payload.size, fields, &inner) &&
pb_get_encoded_size(&canonicalPayload, fields, &inner) && canonicalPayload <= d->payload.size) {
// Only the length matters when sizing a bytes field, so swap it in place instead of
// copying the whole Data; restored below because modules still need the real payload.
const pb_size_t prevSize = d->payload.size;
d->payload.size = (pb_size_t)canonicalPayload;
const bool sized = pb_get_encoded_size(size, &meshtastic_Data_msg, d);
d->payload.size = prevSize;
return sized;
}
}
return pb_get_encoded_size(size, &meshtastic_Data_msg, d);
}
enum class NodeInfoBootstrapResult { NOT_APPLICABLE, VERIFIED, INVALID };
static NodeInfoBootstrapResult verifyFirstContactNodeInfo(meshtastic_MeshPacket *p)
{
if (p->decoded.portnum != meshtastic_PortNum_NODEINFO_APP)
return NodeInfoBootstrapResult::NOT_APPLICABLE;
meshtastic_User user = meshtastic_User_init_zero;
if (!pb_decode_from_bytes(p->decoded.payload.bytes, p->decoded.payload.size, &meshtastic_User_msg, &user) ||
user.public_key.size != 32 || crc32Buffer(user.public_key.bytes, user.public_key.size) != p->from ||
!crypto->xeddsa_verify(user.public_key.bytes, p->from, p->id, p->decoded.portnum, p->decoded.payload.bytes,
p->decoded.payload.size, p->decoded.xeddsa_signature.bytes)) {
return NodeInfoBootstrapResult::INVALID;
}
meshtastic_NodeInfoLite *node = nodeDB->getOrCreateMeshNode(p->from);
if (!node)
return NodeInfoBootstrapResult::INVALID;
node->public_key.size = user.public_key.size;
memcpy(node->public_key.bytes, user.public_key.bytes, user.public_key.size);
nodeInfoLiteSetBit(node, NODEINFO_BITFIELD_HAS_XEDDSA_SIGNED_MASK, true);
p->xeddsa_signed = true;
LOG_DEBUG("Verified first-contact XEdDSA NodeInfo from 0x%08x", p->from);
return NodeInfoBootstrapResult::VERIFIED;
}
bool checkXeddsaReceivePolicy(meshtastic_MeshPacket *p)
{
const auto policy = config.security.packet_signature_policy;
const bool strict = policy == meshtastic_Config_SecurityConfig_PacketSignaturePolicy_PACKET_SIGNATURE_POLICY_STRICT;
const bool compatible = policy == meshtastic_Config_SecurityConfig_PacketSignaturePolicy_PACKET_SIGNATURE_POLICY_COMPATIBLE;
// Only a signature we verify below may mark this packet signed; never trust an inbound flag.
p->xeddsa_signed = false;
if (p->decoded.xeddsa_signature.size == XEDDSA_SIGNATURE_SIZE) {
meshtastic_NodeInfoLite_public_key_t senderKey = {0, {0}};
meshtastic_NodeInfoLite *node = nodeDB->getMeshNode(p->from);
// Authoritative keys only: verifying against an opportunistic cache key would let a planted
// key mark its own node a signer, the trust loop #11116 closed on the decrypt path.
if (nodeDB->copyPublicKeyAuthoritative(p->from, senderKey)) {
p->xeddsa_signed =
crypto->xeddsa_verify(senderKey.bytes, p->from, p->id, p->decoded.portnum, p->decoded.payload.bytes,
p->decoded.payload.size, p->decoded.xeddsa_signature.bytes);
if (p->xeddsa_signed) {
// Learn this node as a signer, so a later unsigned signable broadcast from it is dropped
// A warm-tier key must be re-admitted before setting the signer bit; otherwise Balanced
// forgets downgrade protection as soon as the node is evicted from the hot store.
if (!node)
node = nodeDB->getOrCreateMeshNode(p->from);
if (!node)
return false;
nodeInfoLiteSetBit(node, NODEINFO_BITFIELD_HAS_XEDDSA_SIGNED_MASK, true);
LOG_TRACE("Verified XEdDSA signature from 0x%08x", p->from);
} else {
LOG_WARN("XEdDSA signature verify failed from 0x%08x, drop", p->from);
return false;
}
} else {
const auto bootstrap = verifyFirstContactNodeInfo(p);
if (bootstrap == NodeInfoBootstrapResult::INVALID) {
LOG_WARN("Invalid first-contact XEdDSA NodeInfo from 0x%08x, drop", p->from);
return false;
}
if (bootstrap == NodeInfoBootstrapResult::VERIFIED)
return true;
LOG_DEBUG("No public key for 0x%08x, can't verify XEdDSA signature", p->from);
if (strict)
return false;
}
} else if (p->decoded.xeddsa_signature.size != 0) {
// A signature field that is neither empty nor a full 64 bytes is malformed - honest
// senders emit only those two sizes (perhapsEncode sets 0 or XEDDSA_SIGNATURE_SIZE). Drop
// it: a crafted partial signature would otherwise land in the unsigned branch below while
// its bytes inflated the size estimate, letting a forged broadcast dodge the downgrade drop.
LOG_WARN("Malformed XEdDSA signature (%u bytes) from 0x%08x, drop", (unsigned)p->decoded.xeddsa_signature.size, p->from);
return false;
} else {
if (p->pki_encrypted)
return true;
if (strict) {
LOG_WARN("Drop unsigned packet from 0x%08x in Strict signature mode", p->from);
return false;
}
if (compatible)
return true;
// Balanced rejects only what a signer always signs: non-PKI broadcasts whose signed encoding
// would have fit, plus unicasts on ham where licensed senders sign too. Mirrors perhapsEncode.
if (nodeDB->isKnownXeddsaSigner(p->from) && (isBroadcast(p->to) || owner.is_licensed)) {
size_t canonicalSize;
if (!canonicalSignableSize(&p->decoded, &canonicalSize))
return true; // can't size it; never drop on a sizing failure
if (canonicalSize + XEDDSA_SIGNATURE_FIELD_BYTES + MESHTASTIC_HEADER_LENGTH <= MAX_LORA_PAYLOAD_LEN) {
LOG_WARN("Drop unsigned packet from 0x%08x that previously signed", p->from);
return false;
}
}
}
return true;
}
#endif
RoutingAuthVerdict passesRoutingAuthGate(meshtastic_MeshPacket *p)
{
// Routing still needs the original encrypted representation for byte-for-byte relay and for
// MQTT uplink. Authenticate a copy here; handleReceived() performs the normal in-place decode
// only after stateful routing filters have completed.
if (routingAuthCacheMatches(*p))
return RoutingAuthVerdict::ACCEPT;
meshtastic_MeshPacket wire = *p;
meshtastic_MeshPacket authCandidate = *p;
routingAuthEvaluations++;
if (authCandidate.which_payload_variant == meshtastic_MeshPacket_decoded_tag) {
// Already-decoded remote ingress (notably Portduino SimRadio) did not pass through a
// decryptor. Never trust serialized local authentication metadata on that boundary.
authCandidate.pki_encrypted = false;
authCandidate.public_key.size = 0;
#if !(MESHTASTIC_EXCLUDE_PKI) && !(MESHTASTIC_EXCLUDE_XEDDSA)
concurrency::LockGuard g(cryptLock);
if (!checkXeddsaReceivePolicy(&authCandidate)) {
LOG_WARN("Already-decoded packet rejected by signature policy");
return RoutingAuthVerdict::REJECT;
}
#endif
p->xeddsa_signed = authCandidate.xeddsa_signed;
wire = *p;
storeRoutingAuthCache(wire, authCandidate);
return RoutingAuthVerdict::ACCEPT;
}
const DecodeState state = perhapsDecode(&authCandidate);
if (state == DecodeState::DECODE_POLICY_REJECT) {
LOG_WARN("Packet rejected by signature policy");
return RoutingAuthVerdict::REJECT;
}
if (state == DecodeState::DECODE_FATAL) {
LOG_WARN("Fatal decode error, drop packet");
return RoutingAuthVerdict::REJECT;
}
if (state == DecodeState::DECODE_FAILURE) {
// One-byte hash collisions are indistinguishable from tampering, so relay opaquely
// instead of blackholing; isFromUs stays REJECT to keep forged senders off the ACK path.
if (!isToUs(p) && !isFromUs(p)) {
LOG_WARN("Decryptable packet failed decoding, relay opaquely");
return RoutingAuthVerdict::OPAQUE_RELAY_ONLY;
}
LOG_WARN("Decryptable packet failed decoding, drop");
return RoutingAuthVerdict::REJECT;
}
// Only an explicit unknown-channel result remains eligible for opaque relay.
if (state == DecodeState::DECODE_OPAQUE)
return RoutingAuthVerdict::OPAQUE_RELAY_ONLY;
storeRoutingAuthCache(wire, authCandidate);
return RoutingAuthVerdict::ACCEPT;
}
#if !(MESHTASTIC_EXCLUDE_PKI)
// The fallback costs three X25519 ops before the AEAD tag is checked. Budget is global because p->from is
// attacker-controlled; successful runs refund, and their key is then persisted for the fast path.
#define ADMIN_KEY_FALLBACK_BURST 8
#define ADMIN_KEY_FALLBACK_REFILL_MS 250
static uint32_t adminKeyFallbackTokens = ADMIN_KEY_FALLBACK_BURST;
static uint32_t adminKeyFallbackRefillMs = 0;
#ifdef PIO_UNIT_TESTING
// The refill stamp is a timestamp, so it is only meaningful against the clock that produced it.
// A suite that swaps between the real and the virtual clock leaves a stamp from the other
// timebase, and the next unsigned subtraction reads as a near-infinite gap: the bucket silently
// refills to full. Re-stamp when the clock changes.
void resetAdminKeyFallbackBudget()
{
adminKeyFallbackTokens = ADMIN_KEY_FALLBACK_BURST;
adminKeyFallbackRefillMs = Time::getMillis();
}
#endif
static bool adminKeyFallbackAllowed()
{
bool haveAdminKey = false;
for (int i = 0; i < 3; i++) {
if (config.security.admin_key[i].size == 32) {
haveAdminKey = true;
break;
}
}
if (!haveAdminKey)
return false; // nothing to try, so do not spend a token
// Injectable clock so the budget can be tested without sleeping, and without racing a slow host.
uint32_t now = Time::getMillis();
if (adminKeyFallbackRefillMs == 0)
adminKeyFallbackRefillMs = now;
uint32_t elapsed = now - adminKeyFallbackRefillMs;
if (elapsed >= ADMIN_KEY_FALLBACK_REFILL_MS) {
uint32_t refill = elapsed / ADMIN_KEY_FALLBACK_REFILL_MS;
adminKeyFallbackRefillMs += refill * ADMIN_KEY_FALLBACK_REFILL_MS;
if (refill >= ADMIN_KEY_FALLBACK_BURST - adminKeyFallbackTokens)
adminKeyFallbackTokens = ADMIN_KEY_FALLBACK_BURST;
else
adminKeyFallbackTokens += refill;
}
if (adminKeyFallbackTokens == 0)
return false;
adminKeyFallbackTokens--;
return true;
}
static void adminKeyFallbackRefund()
{
if (adminKeyFallbackTokens < ADMIN_KEY_FALLBACK_BURST)
adminKeyFallbackTokens++;
}
#endif
DecodeState perhapsDecode(meshtastic_MeshPacket *p)
{
concurrency::LockGuard g(cryptLock);
if (config.device.rebroadcast_mode == meshtastic_Config_DeviceConfig_RebroadcastMode_KNOWN_ONLY &&
!nodeInfoLiteHasUser(nodeDB->getMeshNode(p->from))) {
LOG_DEBUG("Node 0x%08x not in nodeDB, Rebroadcast KNOWN_ONLY ignores packet", p->from);
return DecodeState::DECODE_FAILURE;
}
if (p->which_payload_variant == meshtastic_MeshPacket_decoded_tag)
return DecodeState::DECODE_SUCCESS; // If packet was already decoded just return
// Authentication metadata is local-only. Re-establish it below only after successful PKI decryption.
p->pki_encrypted = false;
p->public_key.size = 0;
size_t rawSize = p->encrypted.size;
if (rawSize > sizeof(bytes)) {
LOG_ERROR("Packet too large to decrypt (rawSize=%d > 256)", rawSize);
return DecodeState::DECODE_FATAL;
}
bool decrypted = false;
bool pkiAttempted = false;
bool licensedPkiCandidate = false;
bool matchedChannel = false;
ChannelIndex chIndex = 0;
#if !(MESHTASTIC_EXCLUDE_PKI)
meshtastic_NodeInfoLite *ourNode = nullptr;
const bool pkiCandidate = p->channel == 0 && isToUs(p) && p->to > 0 && !isBroadcast(p->to) &&
rawSize > MESHTASTIC_PKC_OVERHEAD && (ourNode = nodeDB->getMeshNode(p->to)) != nullptr &&
ourNode->public_key.size > 0;
if (pkiCandidate && owner.is_licensed) {
licensedPkiCandidate = true;
} else if (pkiCandidate) {
pkiAttempted = true;
LOG_TRACE("Attempt PKI decryption");
// Resolve the sender's key only for actual PKI-decrypt candidates, not every encrypted channel
// packet: copyPublicKeyForDecrypt() can fall through to a linear scan of TrafficManagement's large
// NodeInfo cache. It returns authoritative keys (hot/warm), or a cold-tier cache key only when it is
// key-proven - an unverified TOFU cache key must not back authenticated (pki_encrypted, p->from)
// DM attribution.
meshtastic_NodeInfoLite_public_key_t remotePublic = {0, {0}};
bool haveRemoteKey = nodeDB->copyPublicKeyForDecrypt(p->from, remotePublic);
// A pending key is an unverified identity claim supplied by whoever opened the handshake, so it is
// accepted only for the exchange itself (checked after decode). perhapsEncode applies the same rule.
bool havePendingKey = false;
if (!haveRemoteKey) {
havePendingKey = crypto->getPendingPublicKey(p->from, remotePublic);
haveRemoteKey = havePendingKey;
}
// Try the sender's known key first, then each configured admin key so an authorized admin can
// reach a node that has not yet learned their key. AES-CCM AEAD rejects wrong candidates.
bool viaAdminKey = false;
bool viaPendingKey = false;
if (haveRemoteKey && crypto->decryptCurve25519(p->from, remotePublic, p->id, rawSize, p->encrypted.bytes, bytes)) {
decrypted = true;
viaPendingKey = havePendingKey;
}
if (!decrypted && adminKeyFallbackAllowed()) {
for (int i = 0; i < 3 && !decrypted; i++) {
if (config.security.admin_key[i].size != 32)
continue;
remotePublic.size = 32;
memcpy(remotePublic.bytes, config.security.admin_key[i].bytes, 32);
if (crypto->decryptCurve25519(p->from, remotePublic, p->id, rawSize, p->encrypted.bytes, bytes)) {
decrypted = true;
viaAdminKey = true;
break; // stop after first successful decryption
}
}
if (decrypted)
adminKeyFallbackRefund();
}
if (decrypted) {
LOG_INFO("PKI Decryption worked");
meshtastic_Data decodedtmp;
memset(&decodedtmp, 0, sizeof(decodedtmp));
size_t payloadSize = rawSize - MESHTASTIC_PKC_OVERHEAD;
if (pb_decode_from_bytes(bytes, payloadSize, &meshtastic_Data_msg, &decodedtmp) &&
decodedtmp.portnum != meshtastic_PortNum_UNKNOWN_APP) {
if (viaPendingKey && decodedtmp.portnum != meshtastic_PortNum_KEY_VERIFICATION_APP) {
// The pending key only proves the handshake initiator holds it, not that they are
// p->from. Beyond the exchange it would let them send DMs that look authenticated.
LOG_WARN("Refusing pending-key decrypt of port %u from 0x%08x", (unsigned)decodedtmp.portnum, p->from);
return DecodeState::DECODE_FAILURE;
}
decrypted = true;
rawSize = payloadSize; // commit the overhead subtraction only on full success
LOG_INFO("Packet decrypted using PKI");
p->pki_encrypted = true;
memcpy(p->public_key.bytes, remotePublic.bytes, 32);
p->public_key.size = 32;
p->decoded = decodedtmp;
p->which_payload_variant = meshtastic_MeshPacket_decoded_tag; // change type to decoded
if (viaAdminKey) {
// Persist the admin key for the sender so future packets take the fast path and we can
// PKI-reply; p->from is bound into the AEAD nonce, so the trusted admin authenticated
// it. commitRemoteKey is the bare-key commit primitive: it bypasses updateUser's
// User-payload path deliberately and handles the TrafficManagement write-through.
// AdminChannelProven = possession shown to the admin channel, not via an XEdDSA
// NodeInfo signature, so the key stays TOFU-grade for signing purposes.
nodeDB->commitRemoteKey(p->from, remotePublic.bytes, NodeDB::KeyCommitTrust::AdminChannelProven);
}
} else {
// AEAD already authenticated this ciphertext, so no other candidate could decode it -
// the payload is simply malformed.
LOG_ERROR("PKC Decrypted, but pb_decode failed");
return DecodeState::DECODE_FAILURE;
}
}
}
#endif
// assert(p->which_payloadVariant == MeshPacket_encrypted_tag);
if (!decrypted) {
// Try to find a channel that works with this hash
for (chIndex = 0; chIndex < channels.getNumChannels(); chIndex++) {
// Try to use this hash/channel pair
if (channels.decryptForHash(chIndex, p->channel)) {
matchedChannel = true;
// we have to copy into a scratch buffer, because these bytes are a union with the decoded protobuf. Create a
// fresh copy for each decrypt attempt.
memcpy(bytes, p->encrypted.bytes, rawSize);
size_t decryptedSize = rawSize;
if (channels.isAEADEnabled(chIndex)) {
// AEAD decryption - no CTR fallback
if (rawSize <= MESHTASTIC_AEAD_OVERHEAD) {
LOG_ERROR("Packet too small for AEAD (size=%d)", rawSize);
continue;
}
CryptoKey k = channels.getKey(chIndex);
if (!crypto->decryptPacketCCM(k, p->from, p->to, p->id, rawSize, p->encrypted.bytes, bytes)) {
LOG_WARN("AEAD authentication failed for ch %d", chIndex);
continue; // reject - no fallback to CTR
}
decryptedSize = rawSize - MESHTASTIC_AEAD_OVERHEAD;
} else {
// Standard AES-CTR decryption
crypto->decrypt(p->from, p->id, rawSize, bytes);
}
// printBytes("plaintext", bytes, p->encrypted.size);
// Take those raw bytes and convert them back into a well structured protobuf we can understand
meshtastic_Data decodedtmp;
memset(&decodedtmp, 0, sizeof(decodedtmp));
if (!pb_decode_from_bytes(bytes, decryptedSize, &meshtastic_Data_msg, &decodedtmp)) {
LOG_DEBUG("Invalid protobufs in received mesh packet id=0x%08x (bad psk?)", p->id);
} else if (decodedtmp.portnum == meshtastic_PortNum_UNKNOWN_APP) {
LOG_DEBUG("Invalid portnum (bad psk?)");
#if !(MESHTASTIC_EXCLUDE_PKI)
} else if (!owner.is_licensed && isToUs(p) && decodedtmp.portnum == meshtastic_PortNum_TEXT_MESSAGE_APP) {
LOG_WARN("Rejecting legacy DM");
return DecodeState::DECODE_FAILURE;
#endif
} else {
p->decoded = decodedtmp;
p->which_payload_variant = meshtastic_MeshPacket_decoded_tag; // change type to decoded
decrypted = true;
break;
}
}
}
}
if (decrypted) {
// parsing was successful
p->channel = chIndex; // change to store the index instead of the hash
#if !(MESHTASTIC_EXCLUDE_PKI) && !(MESHTASTIC_EXCLUDE_XEDDSA)
// Run before merging local-only bitfield state into the decoded Data.
if (!checkXeddsaReceivePolicy(p))
return DecodeState::DECODE_POLICY_REJECT;
#endif
if (p->decoded.has_bitfield)
p->decoded.want_response |= p->decoded.bitfield & BITFIELD_WANT_RESPONSE_MASK;
if (isBlockedEventCoordinatePacket(p)) {
// want_response is already merged above: a position request on the event channel is still
// answered (on the position channel) even though its coordinates are dropped.
LOG_DEBUG("Decoded coordinate packet on event channel; suppress payload logging");
return DecodeState::DECODE_SUCCESS;
}
/* Not actually ever used.
// Decompress if needed. jm
if (p->decoded.portnum == meshtastic_PortNum_TEXT_MESSAGE_COMPRESSED_APP) {
// Decompress the payload
char compressed_in[meshtastic_Constants_DATA_PAYLOAD_LEN] = {};
char decompressed_out[meshtastic_Constants_DATA_PAYLOAD_LEN] = {};
int decompressed_len;
memcpy(compressed_in, p->decoded.payload.bytes, p->decoded.payload.size);
decompressed_len = unishox2_decompress_simple(compressed_in, p->decoded.payload.size, decompressed_out);
// LOG_DEBUG("**Decompressed length - %d ", decompressed_len);
memcpy(p->decoded.payload.bytes, decompressed_out, decompressed_len);
// Switch the port from PortNum_TEXT_MESSAGE_COMPRESSED_APP to PortNum_TEXT_MESSAGE_APP
p->decoded.portnum = meshtastic_PortNum_TEXT_MESSAGE_APP;
} */
printPacket("decoded message", p);
#if ARCH_PORTDUINO
if (portduino_config.traceFilename != "" || portduino_config.logoutputlevel == level_trace) {
LOG_TRACE("%s", MeshPacketSerializer::JsonSerialize(p, false).c_str());
} else if (portduino_config.JSONFilename != "") {
if (portduino_config.JSONFileRotate != 0) {
static uint32_t fileage = 0;
if (portduino_config.JSONFileRotate != 0 &&
(fileage == 0 || !Throttle::isWithinTimespanMs(fileage, portduino_config.JSONFileRotate * 60 * 1000))) {
time_t timestamp = time(NULL);
struct tm *timeinfo;
char buffer[80];
timeinfo = localtime(&timestamp);
strftime(buffer, 80, "%Y%m%d-%H%M%S", timeinfo);
std::string datetime(buffer);
if (JSONFile.is_open()) {
JSONFile.close();
}
JSONFile.open(portduino_config.JSONFilename + "_" + datetime, std::ios::out | std::ios::app);
fileage = Time::skipZero(Time::getMillis());
}
}
if (portduino_config.JSONFilter == (_meshtastic_PortNum)0 || portduino_config.JSONFilter == p->decoded.portnum) {
JSONFile << MeshPacketSerializer::JsonSerialize(p, false) << std::endl;
}
}
#endif
return DecodeState::DECODE_SUCCESS;
} else {
LOG_WARN("No channel found for decoding, hash 0x%x", p->channel);
return (matchedChannel || pkiAttempted || licensedPkiCandidate) ? DecodeState::DECODE_FAILURE
: DecodeState::DECODE_OPAQUE;
}
}
#if !(MESHTASTIC_EXCLUDE_PKI) && !(MESHTASTIC_EXCLUDE_XEDDSA)
/** Exact sender-side sign gate: would this Data still fit the LoRa frame with a 64-byte
* signature attached? Sized with the real encoder so it tracks whatever fields are present. */
static bool signedDataFits(meshtastic_Data *d)
{
const pb_size_t prevSize = d->xeddsa_signature.size;
d->xeddsa_signature.size = XEDDSA_SIGNATURE_SIZE;
size_t encodedSize;
const bool sized = pb_get_encoded_size(&encodedSize, &meshtastic_Data_msg, d);
d->xeddsa_signature.size = prevSize;
return sized && encodedSize + MESHTASTIC_HEADER_LENGTH <= MAX_LORA_PAYLOAD_LEN;
}
#endif
#if !(MESHTASTIC_EXCLUDE_PKI)
bool wouldEncryptWithPKC(const meshtastic_MeshPacket *p, ChannelIndex chIndex, bool haveDestKey)
{
// First, only PKC encrypt packets we are originating
return isFromUs(p) &&
#if ARCH_PORTDUINO
// Sim radio via the cli flag skips PKC
!portduino_config.force_simradio &&
#endif
// Don't use PKC with Ham mode
!owner.is_licensed &&
// Don't use PKC on 'serial' or 'gpio' channels unless explicitly requested
!(p->pki_encrypted != true && (strcasecmp(channels.getName(chIndex), Channels::serialChannel) == 0 ||
strcasecmp(channels.getName(chIndex), Channels::gpioChannel) == 0)) &&
// Check for valid keys and single node destination
config.security.private_key.size == 32 && !isBroadcast(p->to) &&
// Some portnums either make no sense to send with PKC
p->decoded.portnum != meshtastic_PortNum_TRACEROUTE_APP && p->decoded.portnum != meshtastic_PortNum_NODEINFO_APP &&
p->decoded.portnum != meshtastic_PortNum_ROUTING_APP && p->decoded.portnum != meshtastic_PortNum_POSITION_APP &&
// We allow Key Verification messages to be sent without a known destination key, since the point of those messages is
// to exchange keys. The first exchange (no usable key yet) falls through to channel encryption; the follow-on packet
// uses the pending key resolved into haveDestKey/destKey above.
// Though possible the first packet each direction should go non-pkc
// to handle the case where the remote node has our key, but we don't have theirs.
!(p->decoded.portnum == meshtastic_PortNum_KEY_VERIFICATION_APP && !haveDestKey);
}
/**
* PKC fallback for an ack that has no channel in common with the sender.
*
* PKI needs only the two keys, so a DM can reach us over a channel we do not carry. Its ack is a
* ROUTING packet, which wouldEncryptWithPKC() excludes, so it would be channel-encoded, fail at
* setActiveByIndex() with NO_CHANNEL, and never be sent - leaving the sender to retransmit to
* exhaustion for a message that was in fact delivered.
*
* This is the one place an ack is deliberately made opaque to relays. Normally that costs next-hop
* learning and intermediate retransmission cancel, which is why ROUTING is PKC-excluded in general;
* here there is no readable alternative to lose, because without this the ack does not exist.
*
* Scoped as tightly as that argument reaches: a unicast ROUTING packet we originate, carrying a
* request_id, to a destination whose key we hold, under the same ham/sim/private-key preconditions
* PKC always has - and only when the channel index does not resolve. It tests channels.getHash()
* rather than setActiveByIndex() so the predicate has no side effect; generateHash already returns
* -1 for an invalid key, so the two agree on which indexes are unusable. The range check has to come
* first and stay first: getHash() is a bare hashes[i] with no bounds test of its own.
*/
static bool ackNeedsPkcFallback(const meshtastic_MeshPacket *p, ChannelIndex chIndex, bool haveDestKey)
{
return isFromUs(p) &&
#if ARCH_PORTDUINO
!portduino_config.force_simradio &&
#endif
!owner.is_licensed && config.security.private_key.size == 32 && haveDestKey && !isBroadcast(p->to) &&
p->decoded.portnum == meshtastic_PortNum_ROUTING_APP && p->decoded.request_id != 0 &&
(chIndex >= MAX_NUM_CHANNELS || channels.getHash(chIndex) < 0);
}
#endif
/** Return 0 for success or a Routing_Error code for failure
*/
meshtastic_Routing_Error perhapsEncode(meshtastic_MeshPacket *p)
{
concurrency::LockGuard g(cryptLock);
int16_t hash;
// If the packet is not yet encrypted, do so now
if (p->which_payload_variant == meshtastic_MeshPacket_decoded_tag) {
if (isFromUs(p)) {
p->decoded.has_bitfield = true;
p->decoded.bitfield |= (config.lora.config_ok_to_mqtt << BITFIELD_OK_TO_MQTT_SHIFT);
p->decoded.bitfield |= (p->decoded.want_response << BITFIELD_WANT_RESPONSE_SHIFT);
// We own signing for packets we originate; discard any signature a client preset.
// Outside the XEdDSA guard: the field exists in the protobuf on every build, and a
// stale/garbage signature transmitted by a non-signing build would hard-fail
// verification at every XEdDSA-enabled receiver that knows our key.
p->decoded.xeddsa_signature.size = 0;
#if !(MESHTASTIC_EXCLUDE_PKI) && !(MESHTASTIC_EXCLUDE_XEDDSA)
// Licensed packets stay plaintext, so sign both broadcasts and unicasts. Normal mode
// continues to sign broadcasts only. Use the exact encoded size: a payload-size heuristic
// where we sign-then-fail-TOO_LARGE breaks packets that
// were deliverable unsigned, and perhapsDecode() applies the mirror-image rule when
// deciding whether an unsigned broadcast from a known signer is a downgrade.
if (!p->pki_encrypted && (owner.is_licensed || isBroadcast(p->to)) && signedDataFits(&p->decoded)) {
if (crypto->xeddsa_sign(p->from, p->id, p->decoded.portnum, p->decoded.payload.bytes, p->decoded.payload.size,
p->decoded.xeddsa_signature.bytes)) {
p->decoded.xeddsa_signature.size = XEDDSA_SIGNATURE_SIZE;
LOG_TRACE("XEdDSA signed packet 0x%08x", p->id);
}
}
#endif
}
size_t numbytes = pb_encode_to_bytes(bytes, sizeof(bytes), &meshtastic_Data_msg, &p->decoded);
/* Not actually used, so save the cycles
// TODO: Allow modules to opt into compression.
if (p->decoded.portnum == meshtastic_PortNum_TEXT_MESSAGE_APP) {
char original_payload[meshtastic_Constants_DATA_PAYLOAD_LEN];
memcpy(original_payload, p->decoded.payload.bytes, p->decoded.payload.size);
char compressed_out[meshtastic_Constants_DATA_PAYLOAD_LEN] = {0};
int compressed_len;
compressed_len = unishox2_compress_simple(original_payload, p->decoded.payload.size, compressed_out);
LOG_DEBUG("Original length - %d ", p->decoded.payload.size);
LOG_DEBUG("Compressed length - %d ", compressed_len);
LOG_DEBUG("Original message - %.*s ", (int)p->decoded.payload.size, p->decoded.payload.bytes);
// If the compressed length is greater than or equal to the original size, don't use the compressed form
if (compressed_len >= p->decoded.payload.size) {
LOG_DEBUG("Not compressing");
// Set the uncompressed payload variant anyway. Shouldn't hurt?
// p->decoded.which_payloadVariant = Data_payload_tag;
// Otherwise we use the compressor
} else {
LOG_DEBUG("Use compressed message");
// Copy the compressed data into the meshpacket
p->decoded.payload.size = compressed_len;
memcpy(p->decoded.payload.bytes, compressed_out, compressed_len);
p->decoded.portnum = meshtastic_PortNum_TEXT_MESSAGE_COMPRESSED_APP;
}
} */
if (numbytes + MESHTASTIC_HEADER_LENGTH > MAX_LORA_PAYLOAD_LEN)
return meshtastic_Routing_Error_TOO_LARGE;
// printBytes("plaintext", bytes, numbytes);
ChannelIndex chIndex = p->channel; // keep as a local because we are about to change it
#if !(MESHTASTIC_EXCLUDE_PKI)
// Resolve the destination's public key: prefer NodeDB (hot store or warm tier - evicted
// long-tail nodes keep their key there), otherwise (for a key-verification follow-on packet
// that explicitly requested PKI) fall back to the not-yet-verified key held during an
// in-progress handshake. This lets us DH-encode the follow-on packet before the peer's key
// has been committed to NodeDB.
meshtastic_NodeInfoLite_public_key_t destKey = {0, {0}};
bool haveDestKey = nodeDB->copyPublicKey(p->to, destKey);
if (!haveDestKey && p->pki_encrypted && p->decoded.portnum == meshtastic_PortNum_KEY_VERIFICATION_APP &&
crypto->getPendingPublicKey(p->to, destKey)) {
haveDestKey = true;
}
const bool ackFallback = ackNeedsPkcFallback(p, chIndex, haveDestKey);
if (ackFallback)
LOG_INFO("No usable channel %d for ack of 0x%08x, send it over PKC", chIndex, p->decoded.request_id);
// We may want to retool things so we can send a PKC packet when the client specifies a key and nodenum, even if the node
// is not in the local nodedb
//
// ackFallback is tested first so an out-of-range chIndex short-circuits: wouldEncryptWithPKC
// reaches channels.getName(chIndex) before its portnum exclusion, and getByIndex() logs
// "Invalid channel index" on the way past. Without the short-circuit this path would print
// an error and then go on to encode the packet successfully.
if (ackFallback || wouldEncryptWithPKC(p, chIndex, haveDestKey)) {
LOG_DEBUG("Use PKI");
if (numbytes + MESHTASTIC_HEADER_LENGTH + MESHTASTIC_PKC_OVERHEAD > MAX_LORA_PAYLOAD_LEN)
return meshtastic_Routing_Error_TOO_LARGE;
// Check for a usable public key for the destination (NodeDB or a pending key-verification key)
if (!haveDestKey) {
LOG_WARN("Unknown public key for 0x%08x (portnum %d), refuse legacy DM", p->to, p->decoded.portnum);
return meshtastic_Routing_Error_PKI_SEND_FAIL_PUBLIC_KEY;
}
if (p->pki_encrypted && !memfll(p->public_key.bytes, 0, 32) && memcmp(p->public_key.bytes, destKey.bytes, 32) != 0) {
LOG_WARN("Client public key differs from requested: 0x%02x, stored key begins 0x%02x", *p->public_key.bytes,
*destKey.bytes);
return meshtastic_Routing_Error_PKI_FAILED;
}
// On failure encrypted.bytes holds no ciphertext, so continuing would put the plaintext
// on the air labelled pki_encrypted.
if (!crypto->encryptCurve25519(p->to, getFrom(p), destKey, p->id, numbytes, bytes, p->encrypted.bytes)) {
LOG_WARN("PKI encryption failed for 0x%08x", p->to);
return meshtastic_Routing_Error_PKI_FAILED;
}
numbytes += MESHTASTIC_PKC_OVERHEAD;
p->channel = 0;
p->pki_encrypted = true;
} else
#endif
{
if (p->pki_encrypted == true) {
// Client specifically requested PKI encryption
return meshtastic_Routing_Error_PKI_FAILED;
}
const bool useAead = channels.isAEADEnabled(chIndex);
if (useAead && numbytes + MESHTASTIC_HEADER_LENGTH + MESHTASTIC_AEAD_OVERHEAD > MAX_LORA_PAYLOAD_LEN)
return meshtastic_Routing_Error_TOO_LARGE;
hash = channels.setActiveByIndex(chIndex);
// Now that we are encrypting the packet channel should be the hash (no longer the index)
p->channel = hash;
if (hash < 0)
return meshtastic_Routing_Error_NO_CHANNEL;
if (useAead) {
// AEAD (AES-CCM) authenticated encryption path
CryptoKey k = channels.getKey(chIndex);
if (!crypto->encryptPacketCCM(k, getFrom(p), p->to, p->id, numbytes, bytes, p->encrypted.bytes)) {
LOG_ERROR("AEAD encryption failed for ch %d", chIndex);
return meshtastic_Routing_Error_BAD_REQUEST;
}
numbytes += MESHTASTIC_AEAD_OVERHEAD;
} else {
// Standard AES-CTR encryption path
crypto->encryptPacket(getFrom(p), p->id, numbytes, bytes);
memcpy(p->encrypted.bytes, bytes, numbytes);
}
}
// Copy back into the packet and set the variant type
p->encrypted.size = numbytes;
p->which_payload_variant = meshtastic_MeshPacket_encrypted_tag;
}
return meshtastic_Routing_Error_NONE;
}
NodeNum Router::getNodeNum()
{
return nodeDB->getNodeNum();
}
bool Router::enqueueDeferredLocal(meshtastic_MeshPacket *p, RxSource src)
{
if (deferredLocalCount >= deferredLocalCapacity)
return false;
uint8_t tail = (deferredLocalHead + deferredLocalCount) % deferredLocalCapacity;
deferredLocalQueue[tail].p = p;
deferredLocalQueue[tail].src = src;
deferredLocalCount++;
return true;
}
bool Router::dequeueDeferredLocal(DeferredLocal &out)
{
if (deferredLocalCount == 0)
return false;
out = deferredLocalQueue[deferredLocalHead];
deferredLocalHead = (deferredLocalHead + 1) % deferredLocalCapacity;
deferredLocalCount--;
return true;
}
void Router::deliverLocal(meshtastic_MeshPacket *p, RxSource src)
{
// Top level: handle synchronously, exactly as before the depth guard existed.
bool nested;
{
concurrency::LockGuard g(&deferredLock);
nested = handleDepth > 0;
}
if (!nested) {
handleReceived(p, src);
return;
}
// Nested: a module sent this from inside callModules(). Defer a copy so the outermost
// handleReceived() drains it once the current dispatch unwinds, instead of stacking another
// handleReceived() frame on top of the module handler (nRF52 stack overflow on config save).
meshtastic_MeshPacket *copy = packetPool.allocCopy(*p);
if (copy) {
// Re-check depth under the lock that also gates the drain's decrement, so a drain finishing
// while we allocated cannot leave this copy stranded in the ring.
bool stillNested = false, queued = false;
{
concurrency::LockGuard g(&deferredLock);
stillNested = handleDepth > 0;
if (stillNested)
queued = enqueueDeferredLocal(copy, src);
}
if (queued)
return;
if (!stillNested) {
// The drain finished first, so nothing would pick this up. Go through handleReceived()
// rather than dispatchReceived() so a loopback from its modules still defers.
handleReceived(copy, src);
packetPool.release(copy);
return;
}
}
// Pool exhausted or queue full: drop the deferral. Leak-free and degraded but safe - the
// packet still followed its normal non-loopback path (SHOULD_RELEASE, or the TX path for a
// broadcast). Mirrors sendToPhone()'s degrade-on-exhaustion behavior.
if (copy)
packetPool.release(copy);
LOG_WARN("Deferred local queue full/alloc failed, drop loopback of 0x%08x", p->id);
#ifdef PIO_UNIT_TESTING
deferredLocalDropped++;
#endif
}
/**
* Handle any packet that is received by an interface on this node.
* Note: some packets may merely being passed through this node and will be forwarded elsewhere.
*/
void Router::handleReceived(meshtastic_MeshPacket *p, RxSource src)
{
{
concurrency::LockGuard g(&deferredLock);
handleDepth++;
#ifdef PIO_UNIT_TESTING
if (handleDepth > maxHandleDepthObserved)
maxHandleDepthObserved = handleDepth;
#endif
}
dispatchReceived(p, src);
// Decide "am I the last frame" and drop the depth in one critical section. Splitting them lets
// two frames both read the same pre-decrement value, skip the drain, and strand the ring.
for (;;) {
DeferredLocal d;
{
concurrency::LockGuard g(&deferredLock);
if (handleDepth > 1) {
// Another frame is still live and will own the drain once it is last.
handleDepth--;
return;
}
if (!dequeueDeferredLocal(d)) {
// Last frame and nothing queued, so zero is reached only with the ring empty.
handleDepth--;
return;
}
}
// Depth stays at 1 across the drain, so a loopback from these modules defers instead of
// recursing, and dispatch runs outside the lock.
dispatchReceived(d.p, d.src);
packetPool.release(d.p);
}
}
void Router::dispatchReceived(meshtastic_MeshPacket *p, RxSource src)
{
bool skipHandle = false;
// Store a copy of the encrypted packet for MQTT.
// Kept as a local (not a class member) so each dispatch owns its own copy. A shared member was
// historically overwritten without release when a module's reply re-entered this path through
// MeshService::sendToMesh -> Router::sendLocal (issues #9632, #10101, #8729). Nested local
// sends are now deferred rather than synchronously re-entrant (see the drain in
// handleReceived()), so this no longer strictly needs to be a local, but it is kept per-call.
DEBUG_HEAP_BEFORE;
meshtastic_MeshPacket *p_encrypted = packetPool.allocCopy(*p);
DEBUG_HEAP_AFTER("Router::handleReceived", p_encrypted);
// Consume the decoded/authenticated handoff after preserving the exact encrypted packet and
// before mutating any packet fields that participate in the exact cache match.
if (src == RX_SRC_RADIO)
applyRoutingAuthCache(p);
// Keep the decoded working packet and encrypted MQTT copy on the same local arrival timestamp.
// See computeRxTimeStamp() for the placeholder/has_rx_time semantics.
const RxTimeStamp rxStamp = computeRxTimeStamp();
p->rx_time = rxStamp.time;
p->has_rx_time = rxStamp.valid;
if (p_encrypted) {
p_encrypted->rx_time = rxStamp.time;
p_encrypted->has_rx_time = rxStamp.valid;
}
// Take those raw bytes and convert them back into a well structured protobuf we can understand
auto decodedState = perhapsDecode(p);
if (decodedState == DecodeState::DECODE_FATAL || decodedState == DecodeState::DECODE_POLICY_REJECT ||
decodedState == DecodeState::DECODE_FAILURE) {
// Fatal decoding error, we can't do anything with this packet
LOG_WARN(decodedState == DecodeState::DECODE_POLICY_REJECT
? "Packet rejected by signature policy"
: (decodedState == DecodeState::DECODE_FATAL ? "Fatal decode error, drop packet"
: "Decryptable packet failed decoding, drop"));
// A policy rejection is attacker-controlled input and must not cancel a valid pending
// transmission with the same (from, id). Preserve the pre-existing fatal-decode behavior.
if (decodedState == DecodeState::DECODE_FATAL)
cancelSending(p->from, p->id);
skipHandle = true;
} else if (decodedState == DecodeState::DECODE_SUCCESS) {
// parsing was successful, queue for our recipient
if (src == RX_SRC_LOCAL)
printPacket("handleReceived(LOCAL)", p);
else if (src == RX_SRC_USER)
printPacket("handleReceived(USER)", p);
else
printPacket("handleReceived(REMOTE)", p);
#if MESHTASTIC_PREHOP_DROP
// Pre-hop firmware drop, post-decode half: a packet whose hop_start is still missing/unknown comes
// from pre-hop firmware - keep it out of module processing, admin handling, phone delivery, MQTT
// and rebroadcast.
if (shouldSkipHandleForPostDecodeHop(*p)) {
logHopStartDrop(*p, "post-decode pre-hop drop");
cancelSending(p->from, p->id);
skipHandle = true;
}
#endif
// Neighbor info module is disabled, ignore expensive neighbor info packets
if (p->which_payload_variant == meshtastic_MeshPacket_decoded_tag &&
p->decoded.portnum == meshtastic_PortNum_NEIGHBORINFO_APP &&
(!moduleConfig.has_neighbor_info || !moduleConfig.neighbor_info.enabled)) {
LOG_DEBUG("Neighbor info module disabled, ignore packet");
cancelSending(p->from, p->id);
skipHandle = true;
}
#if !MESHTASTIC_EXCLUDE_BEACON
// Beacon listening is disabled: drop beacon packets so they are neither surfaced to the
// phone nor handled on-device (same pattern as the disabled neighbor-info case above).
if (p->which_payload_variant == meshtastic_MeshPacket_decoded_tag &&
p->decoded.portnum == meshtastic_PortNum_MESH_BEACON_APP &&
(!moduleConfig.has_mesh_beacon ||
!(moduleConfig.mesh_beacon.flags & meshtastic_ModuleConfig_MeshBeaconConfig_Flags_FLAG_LISTEN_ENABLED))) {
LOG_DEBUG("Beacon listening disabled, ignore packet");
cancelSending(p->from, p->id);
skipHandle = true;
}
#endif
bool shouldIgnoreNonstandardPorts =
config.device.rebroadcast_mode == meshtastic_Config_DeviceConfig_RebroadcastMode_CORE_PORTNUMS_ONLY;
#if USERPREFS_EVENT_MODE
shouldIgnoreNonstandardPorts = true;
#endif
if (shouldIgnoreNonstandardPorts && p->which_payload_variant == meshtastic_MeshPacket_decoded_tag &&
!IS_ONE_OF(p->decoded.portnum, meshtastic_PortNum_TEXT_MESSAGE_APP, meshtastic_PortNum_TEXT_MESSAGE_COMPRESSED_APP,
meshtastic_PortNum_POSITION_APP, meshtastic_PortNum_NODEINFO_APP, meshtastic_PortNum_ROUTING_APP,
meshtastic_PortNum_TELEMETRY_APP, meshtastic_PortNum_ADMIN_APP, meshtastic_PortNum_ALERT_APP,
meshtastic_PortNum_KEY_VERIFICATION_APP, meshtastic_PortNum_WAYPOINT_APP,
meshtastic_PortNum_STORE_FORWARD_APP, meshtastic_PortNum_TRACEROUTE_APP,
meshtastic_PortNum_STORE_FORWARD_PLUSPLUS_APP)) {
LOG_DEBUG("Ignore packet on non-standard portnum for CORE_PORTNUMS_ONLY");
cancelSending(p->from, p->id);
skipHandle = true;
}
#if USERPREFS_BLOCK_POSITION_ON_EVENT_CHANNEL
// Discard coordinate-bearing packets that arrive on the event ("everyone")
// channel: don't process, store in NodeDB, or rebroadcast them.
if (!skipHandle && isBlockedEventCoordinatePacket(p)) {
// A position request addressed to us is still answered, on our position channel at that
// channel's precision, so "request position" from a node that only shares the event channel
// with us resolves where positions actually live. The requester's own coordinates are
// still dropped: not stored, not forwarded to the phone, not relayed, not published.
// Builds without the position module (MESHTASTIC_EXCLUDE_GPS, e.g. repeaters) have nothing
// to answer with, and neither the symbol nor the global exists to link against.
#if !MESHTASTIC_EXCLUDE_GPS
if (isEventChannelPositionRequestForUs(p) && positionModule)
positionModule->replyOnPositionChannel(*p);
#endif
LOG_DEBUG("Drop coordinate packet on event (everyone) channel");
cancelSending(p->from, p->id);
skipHandle = true;
}
#endif
} else {
printPacket("packet decoding failed or skipped (no PSK?)", p);
}
// call modules here
// If this could be a spoofed packet, don't let the modules see it.
if (!skipHandle) {
MeshModule::callModules(*p, src);
#if !MESHTASTIC_EXCLUDE_MQTT
if (p_encrypted == nullptr) {
LOG_WARN("p_encrypted null, skip MQTT publish");
} else {
// Mark as pki_encrypted if it is not yet decoded and MQTT encryption is also enabled, hash matches and it's a DM not
// to us (because we would be able to decrypt it)
if (decodedState == DecodeState::DECODE_OPAQUE && moduleConfig.mqtt.encryption_enabled && p->channel == 0x00 &&
!isBroadcast(p->to) && !isToUs(p))
p_encrypted->pki_encrypted = true;
// After potentially altering it, publish received message to MQTT if we're not the original transmitter of the packet
if ((decodedState == DecodeState::DECODE_SUCCESS || p_encrypted->pki_encrypted) && moduleConfig.mqtt.enabled &&
!isFromUs(p) && mqtt) {
if (decodedState == DecodeState::DECODE_SUCCESS && p->decoded.portnum == meshtastic_PortNum_TRACEROUTE_APP &&
moduleConfig.mqtt.encryption_enabled) {
// For TRACEROUTE_APP packets release the original encrypted packet and encrypt a new from the changed packet
// Only release the original after successful allocation to avoid losing an incomplete but valid packet
auto *p_encrypted_new = packetPool.allocCopy(*p);
if (p_encrypted_new) {
auto encodeResult = perhapsEncode(p_encrypted_new);
if (encodeResult != meshtastic_Routing_Error_NONE) {
// Encryption failed, release the new packet and fall back to sending the original encrypted packet to
// MQTT
LOG_WARN("New TR packet encrypt failed, send original TR to MQTT");
packetPool.release(p_encrypted_new);
p_encrypted_new = nullptr;
} else {
// Successfully re-encrypted, release the original encrypted packet and use the new one for MQTT
packetPool.release(p_encrypted);
p_encrypted = p_encrypted_new;
}
} else {
// Allocation failed, log a warning and fall back to sending the original encrypted packet to MQTT
LOG_WARN("Alloc encrypted TR packet failed, send original TR to MQTT");
}
}
mqtt->onSend(*p_encrypted, *p, p->channel);
}
}
#endif
}
packetPool.release(p_encrypted); // Release the encrypted packet (release() handles nullptr)
}
void Router::perhapsHandleReceived(meshtastic_MeshPacket *p)
{
#if ARCH_PORTDUINO
// Even ignored packets get logged in the trace
if (portduino_config.traceFilename != "" || portduino_config.logoutputlevel == level_trace) {
// Store the arrival timestamp for the phone before it's traced.
stampRxTime(p);
LOG_TRACE("%s", MeshPacketSerializer::JsonSerializeEncrypted(p).c_str());
}
#endif
// assert(radioConfig.has_preferences);
if (is_in_repeated(config.lora.ignore_incoming, p->from)) {
clearRoutingAuthCache();
LOG_DEBUG("Ignore msg, 0x%08x in ignore list", p->from);
packetPool.release(p);
return;
}
meshtastic_NodeInfoLite const *node = nodeDB->getMeshNode(p->from);
if (nodeInfoLiteIsIgnored(node)) {
clearRoutingAuthCache();
LOG_DEBUG("Ignore msg, 0x%08x is ignored", p->from);
packetPool.release(p);
return;
}
if (p->from == NODENUM_BROADCAST) {
clearRoutingAuthCache();
LOG_DEBUG("Ignore msg from broadcast address");
packetPool.release(p);
return;
}
if (config.lora.ignore_mqtt && p->via_mqtt) {
clearRoutingAuthCache();
LOG_DEBUG("Msg came in via MQTT from 0x%08x", p->from);
packetPool.release(p);
return;
}
if (shouldDropPacketForPreHop(*p)) {
clearRoutingAuthCache();
logHopStartDrop(*p, "pre-hop drop");
packetPool.release(p);
return;
}
// Decrypt and authenticate before Reliable/Flooding/NextHop filters can update retry
// timers, packet history, implicit ACK state, cancellation, or relay queues. A packet for
// an unknown channel passes as opaque traffic and retains the existing relay behavior.
const auto authVerdict = passesRoutingAuthGate(p);
if (authVerdict == RoutingAuthVerdict::REJECT) {
packetPool.release(p);
return;
}
if (authVerdict == RoutingAuthVerdict::OPAQUE_RELAY_ONLY) {
// A packet we originated but cannot decrypt (a PKI DM we sent, overheard being rebroadcast)
// is opaque to us and would otherwise skip shouldFilterReceived entirely, so the implicit
// ACK that marks a DM "Delivered to mesh" never fires. The ACK is header-only (from/id), so
// generate it here from the still-encrypted packet before opaque relay.
if (isFromUs(p))
perhapsGenerateImplicitAckForOwnOverheard(p);
relayOpaquePacket(p);
packetPool.release(p);
return;
}
if (shouldFilterReceived(p)) {
clearRoutingAuthCache();
LOG_DEBUG("Incoming msg filtered from 0x%08x", p->from);
packetPool.release(p);
return;
}
// Note: we avoid calling shouldFilterReceived if we are supposed to ignore certain nodes - because some overrides might
// cache/learn of the existence of nodes (i.e. FloodRouter) that they should not
handleReceived(p);
packetPool.release(p);
}