ZeroTierOne/node/Topology.cpp
2014-02-11 15:02:21 -08:00

303 lines
8.2 KiB
C++

/*
* ZeroTier One - Global Peer to Peer Ethernet
* Copyright (C) 2012-2013 ZeroTier Networks LLC
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* --
*
* ZeroTier may be used and distributed under the terms of the GPLv3, which
* are available at: http://www.gnu.org/licenses/gpl-3.0.html
*
* If you would like to embed ZeroTier into a commercial application or
* redistribute it in a modified binary form, please contact ZeroTier Networks
* LLC. Start here: http://www.zerotier.com/
*/
#include <algorithm>
#include "Topology.hpp"
#include "NodeConfig.hpp"
#include "CMWC4096.hpp"
#define ZT_PEER_WRITE_BUF_SIZE 131072
namespace ZeroTier {
Topology::Topology(const RuntimeEnvironment *renv,bool enablePermanentIdCaching) :
_r(renv),
_amSupernode(false)
{
if (enablePermanentIdCaching)
_idCacheBase = (_r->homePath + ZT_PATH_SEPARATOR_S + "iddb.d");
_loadPeers();
}
Topology::~Topology()
{
clean();
_dumpPeers();
}
void Topology::setSupernodes(const std::map< Identity,std::vector<InetAddress> > &sn)
{
Mutex::Lock _l(_supernodes_m);
_supernodes = sn;
_supernodeAddresses.clear();
_supernodePeers.clear();
uint64_t now = Utils::now();
for(std::map< Identity,std::vector<InetAddress> >::const_iterator i(sn.begin());i!=sn.end();++i) {
if (i->first != _r->identity) {
SharedPtr<Peer> p(getPeer(i->first.address()));
if (!p)
p = addPeer(SharedPtr<Peer>(new Peer(_r->identity,i->first)));
for(std::vector<InetAddress>::const_iterator j(i->second.begin());j!=i->second.end();++j)
p->setPathAddress(*j,true);
p->setLastUsed(now);
_supernodePeers.push_back(p);
}
_supernodeAddresses.insert(i->first.address());
}
_amSupernode = (_supernodes.find(_r->identity) != _supernodes.end());
}
SharedPtr<Peer> Topology::addPeer(const SharedPtr<Peer> &peer)
{
if (peer->address() == _r->identity.address()) {
TRACE("BUG: addNewPeer() caught and ignored attempt to add peer for self");
throw std::logic_error("cannot add peer for self");
}
uint64_t now = Utils::now();
Mutex::Lock _l(_activePeers_m);
SharedPtr<Peer> p(_activePeers.insert(std::pair< Address,SharedPtr<Peer> >(peer->address(),peer)).first->second);
p->setLastUsed(now);
saveIdentity(p->identity());
return p;
}
SharedPtr<Peer> Topology::getPeer(const Address &zta)
{
if (zta == _r->identity.address()) {
TRACE("BUG: ignored attempt to getPeer() for self, returned NULL");
return SharedPtr<Peer>();
}
uint64_t now = Utils::now();
Mutex::Lock _l(_activePeers_m);
std::map< Address,SharedPtr<Peer> >::const_iterator ap(_activePeers.find(zta));
if ((ap != _activePeers.end())&&(ap->second)) {
ap->second->setLastUsed(now);
return ap->second;
}
return SharedPtr<Peer>();
}
Identity Topology::getIdentity(const Address &zta)
{
SharedPtr<Peer> p(getPeer(zta));
if (p)
return p->identity();
if (_idCacheBase.length()) {
std::string idcPath(_idCacheBase + ZT_PATH_SEPARATOR_S + zta.toString());
std::string ids;
if (Utils::readFile(idcPath.c_str(),ids)) {
try {
return Identity(ids);
} catch ( ... ) {} // ignore invalid IDs
}
}
return Identity();
}
void Topology::saveIdentity(const Identity &id)
{
if ((id)&&(_idCacheBase.length())) {
std::string idcPath(_idCacheBase + ZT_PATH_SEPARATOR_S + id.address().toString());
if (!Utils::fileExists(idcPath.c_str()))
Utils::writeFile(idcPath.c_str(),id.toString(false));
}
}
SharedPtr<Peer> Topology::getBestSupernode(const Address *avoid,unsigned int avoidCount,bool strictAvoid) const
{
SharedPtr<Peer> bestSupernode;
unsigned int bestSupernodeLatency = 65536;
uint64_t now = Utils::now();
Mutex::Lock _l(_supernodes_m);
// First look for a best supernode by comparing latencies, but exclude
// supernodes that have not responded to direct messages in order to
// try to exclude any that are dead or unreachable.
for(std::vector< SharedPtr<Peer> >::const_iterator sn(_supernodePeers.begin());sn!=_supernodePeers.end();) {
// Skip explicitly avoided relays
for(unsigned int i=0;i<avoidCount;++i) {
if (avoid[i] == (*sn)->address())
goto keep_searching_for_supernodes;
}
// Skip possibly comatose or unreachable relays
uint64_t lds = (*sn)->lastDirectSend();
uint64_t ldr = (*sn)->lastDirectReceive();
if ((lds)&&(lds > ldr)&&((lds - ldr) > ZT_PEER_RELAY_CONVERSATION_LATENCY_THRESHOLD))
goto keep_searching_for_supernodes;
if ((*sn)->hasActiveDirectPath(now)) {
unsigned int l = (*sn)->latency();
if (bestSupernode) {
if ((l)&&(l < bestSupernodeLatency)) {
bestSupernodeLatency = l;
bestSupernode = *sn;
}
} else {
if (l)
bestSupernodeLatency = l;
bestSupernode = *sn;
}
}
keep_searching_for_supernodes:
++sn;
}
if (bestSupernode) {
bestSupernode->setLastUsed(now);
return bestSupernode;
} else if (strictAvoid)
return SharedPtr<Peer>();
// If we have nothing from above, just pick one without avoidance criteria.
for(std::vector< SharedPtr<Peer> >::const_iterator sn=_supernodePeers.begin();sn!=_supernodePeers.end();++sn) {
if ((*sn)->hasActiveDirectPath(now)) {
unsigned int l = (*sn)->latency();
if (bestSupernode) {
if ((l)&&(l < bestSupernodeLatency)) {
bestSupernodeLatency = l;
bestSupernode = *sn;
}
} else {
if (l)
bestSupernodeLatency = l;
bestSupernode = *sn;
}
}
}
if (bestSupernode)
bestSupernode->setLastUsed(now);
return bestSupernode;
}
void Topology::clean()
{
uint64_t now = Utils::now();
Mutex::Lock _l(_activePeers_m);
Mutex::Lock _l2(_supernodes_m);
for(std::map< Address,SharedPtr<Peer> >::iterator p(_activePeers.begin());p!=_activePeers.end();) {
if (((now - p->second->lastUsed()) >= ZT_PEER_IN_MEMORY_EXPIRATION)&&(!_supernodeAddresses.count(p->second->address())))
_activePeers.erase(p++);
else ++p;
}
}
void Topology::_dumpPeers()
{
Buffer<ZT_PEER_WRITE_BUF_SIZE> buf;
std::string pdpath(_r->homePath + ZT_PATH_SEPARATOR_S + "peers.persist");
Mutex::Lock _l(_activePeers_m);
FILE *pd = fopen(pdpath.c_str(),"wb");
if (!pd)
return;
if (fwrite("ZTPD0",5,1,pd) != 1) {
fclose(pd);
Utils::rm(pdpath);
return;
}
for(std::map< Address,SharedPtr<Peer> >::iterator p(_activePeers.begin());p!=_activePeers.end();++p) {
try {
p->second->serialize(buf);
if (buf.size() >= (ZT_PEER_WRITE_BUF_SIZE / 2)) {
if (fwrite(buf.data(),buf.size(),1,pd) != 1) {
fclose(pd);
Utils::rm(pdpath);
return;
}
buf.clear();
}
} catch ( ... ) {
fclose(pd);
Utils::rm(pdpath);
return;
}
}
if (buf.size()) {
if (fwrite(buf.data(),buf.size(),1,pd) != 1) {
fclose(pd);
Utils::rm(pdpath);
return;
}
}
fclose(pd);
Utils::lockDownFile(pdpath.c_str(),false);
}
void Topology::_loadPeers()
{
Buffer<ZT_PEER_WRITE_BUF_SIZE> buf;
std::string pdpath(_r->homePath + ZT_PATH_SEPARATOR_S + "peers.persist");
Mutex::Lock _l(_activePeers_m);
_activePeers.clear();
FILE *pd = fopen(pdpath.c_str(),"rb");
if (!pd)
return;
try {
char magic[5];
if ((fread(magic,5,1,pd) == 1)&&(!memcmp("ZTPD0",magic,5))) {
long rlen = 0;
do {
long rlen = (long)fread(buf.data() + buf.size(),1,ZT_PEER_WRITE_BUF_SIZE - buf.size(),pd);
if (rlen < 0) rlen = 0;
buf.setSize(buf.size() + (unsigned int)rlen);
unsigned int ptr = 0;
while ((ptr < (ZT_PEER_WRITE_BUF_SIZE / 2))&&(ptr < buf.size())) {
SharedPtr<Peer> p(new Peer());
ptr += p->deserialize(buf,ptr);
_activePeers[p->address()] = p;
saveIdentity(p->identity());
}
if (ptr) {
memmove(buf.data(),buf.data() + ptr,buf.size() - ptr);
buf.setSize(buf.size() - ptr);
}
} while (rlen > 0);
}
} catch ( ... ) {
_activePeers.clear();
}
fclose(pd);
Utils::rm(pdpath);
}
} // namespace ZeroTier