Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
tree: 725f626aae
Fetching contributors…

Octocat-spinner-32-eaf2f5

Cannot retrieve contributors at this time

file 410 lines (318 sloc) 16.04 kb
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
/**
* Copyright (C) 2008 10gen Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License, version 3,
* as published by the Free Software Foundation.
*
* 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 Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/

#include "pch.h"

#include "mongo/s/balance.h"

#include "mongo/client/distlock.h"
#include "mongo/client/dbclientcursor.h"
#include "mongo/db/jsobj.h"
#include "mongo/db/cmdline.h"
#include "mongo/s/chunk.h"
#include "mongo/s/cluster_constants.h"
#include "mongo/s/config.h"
#include "mongo/s/grid.h"
#include "mongo/s/server.h"
#include "mongo/s/shard.h"

namespace mongo {

    Balancer balancer;

    Balancer::Balancer() : _balancedLastTime(0), _policy( new BalancerPolicy() ) {}

    Balancer::~Balancer() {
    }

    int Balancer::_moveChunks(const vector<CandidateChunkPtr>* candidateChunks,
                              bool secondaryThrottle,
                              bool waitForDelete)
    {
        int movedCount = 0;

        for ( vector<CandidateChunkPtr>::const_iterator it = candidateChunks->begin(); it != candidateChunks->end(); ++it ) {
            const CandidateChunk& chunkInfo = *it->get();

            DBConfigPtr cfg = grid.getDBConfig( chunkInfo.ns );
            verify( cfg );

            ChunkManagerPtr cm = cfg->getChunkManager( chunkInfo.ns );
            verify( cm );

            ChunkPtr c = cm->findIntersectingChunk( chunkInfo.chunk.min );
            if ( c->getMin().woCompare( chunkInfo.chunk.min ) || c->getMax().woCompare( chunkInfo.chunk.max ) ) {
                // likely a split happened somewhere
                cm = cfg->getChunkManager( chunkInfo.ns , true /* reload */);
                verify( cm );

                c = cm->findIntersectingChunk( chunkInfo.chunk.min );
                if ( c->getMin().woCompare( chunkInfo.chunk.min ) || c->getMax().woCompare( chunkInfo.chunk.max ) ) {
                    log() << "chunk mismatch after reload, ignoring will retry issue " << chunkInfo.chunk.toString() << endl;
                    continue;
                }
            }

            BSONObj res;
            if (c->moveAndCommit(Shard::make(chunkInfo.to),
                                 Chunk::MaxChunkSize,
                                 secondaryThrottle,
                                 waitForDelete,
                                 res)) {
                movedCount++;
                continue;
            }

            // the move requires acquiring the collection metadata's lock, which can fail
            log() << "balancer move failed: " << res << " from: " << chunkInfo.from << " to: " << chunkInfo.to
                  << " chunk: " << chunkInfo.chunk << endl;

            if ( res["chunkTooBig"].trueValue() ) {
                // reload just to be safe
                cm = cfg->getChunkManager( chunkInfo.ns );
                verify( cm );
                c = cm->findIntersectingChunk( chunkInfo.chunk.min );
                
                log() << "forcing a split because migrate failed for size reasons" << endl;
                
                res = BSONObj();
                c->singleSplit( true , res );
                log() << "forced split results: " << res << endl;
                
                if ( ! res["ok"].trueValue() ) {
                    log() << "marking chunk as jumbo: " << c->toString() << endl;
                    c->markAsJumbo();
                    // we increment moveCount so we do another round right away
                    movedCount++;
                }

            }
        }

        return movedCount;
    }

    void Balancer::_ping( DBClientBase& conn, bool waiting ) {
        WriteConcern w = conn.getWriteConcern();
        conn.setWriteConcern( W_NONE );

        conn.update( ConfigNS::mongos ,
                     BSON( MongosFields::name(_myid) ) ,
                     BSON( "$set" << BSON( MongosFields::ping(jsTime()) <<
                                           MongosFields::up((int)(time(0)-_started)) <<
                                           MongosFields::waiting(waiting) ) ) ,
                     true );

        conn.setWriteConcern( w);
    }

    bool Balancer::_checkOIDs() {
        vector<Shard> all;
        Shard::getAllShards( all );

        map<int,Shard> oids;

        for ( vector<Shard>::iterator i=all.begin(); i!=all.end(); ++i ) {
            Shard s = *i;
            BSONObj f = s.runCommand( "admin" , "features" , true );
            if ( f["oidMachine"].isNumber() ) {
                int x = f["oidMachine"].numberInt();
                if ( oids.count(x) == 0 ) {
                    oids[x] = s;
                }
                else {
                    log() << "error: 2 machines have " << x << " as oid machine piece " << s.toString() << " and " << oids[x].toString() << endl;
                    s.runCommand( "admin" , BSON( "features" << 1 << "oidReset" << 1 ) , true );
                    oids[x].runCommand( "admin" , BSON( "features" << 1 << "oidReset" << 1 ) , true );
                    return false;
                }
            }
            else {
                log() << "warning: oidMachine not set on: " << s.toString() << endl;
            }
        }
        return true;
    }

    void Balancer::_doBalanceRound( DBClientBase& conn, vector<CandidateChunkPtr>* candidateChunks ) {
        verify( candidateChunks );

        //
        // 1. Check whether there is any sharded collection to be balanced by querying
        // the ShardsNS::collections collection
        //

        auto_ptr<DBClientCursor> cursor = conn.query(ConfigNS::collection, BSONObj());
        vector< string > collections;
        while ( cursor->more() ) {
            BSONObj col = cursor->nextSafe();

            // sharded collections will have a shard "key".
            if ( ! col[CollectionFields::key()].eoo() &&
                 ! col[CollectionFields::noBalance()].trueValue() ){
                collections.push_back( col[CollectionFields::name()].String() );
            }
            else if( col[CollectionFields::noBalance()].trueValue() ){
                LOG(1) << "not balancing collection " << col[CollectionFields::name()].String()
                       << ", explicitly disabled" << endl;
            }

        }
        cursor.reset();

        if ( collections.empty() ) {
            LOG(1) << "no collections to balance" << endl;
            return;
        }

        //
        // 2. Get a list of all the shards that are participating in this balance round
        // along with any maximum allowed quotas and current utilization. We get the
        // latter by issuing db.serverStatus() (mem.mapped) to all shards.
        //
        // TODO: skip unresponsive shards and mark information as stale.
        //

        vector<Shard> allShards;
        Shard::getAllShards( allShards );
        if ( allShards.size() < 2) {
            LOG(1) << "can't balance without more active shards" << endl;
            return;
        }
        
        ShardInfoMap shardInfo;
        for ( vector<Shard>::const_iterator it = allShards.begin(); it != allShards.end(); ++it ) {
            const Shard& s = *it;
            ShardStatus status = s.getStatus();
            shardInfo[ s.getName() ] = ShardInfo( s.getMaxSize(),
                                                  status.mapped(),
                                                  s.isDraining(),
                                                  status.hasOpsQueued(),
                                                  s.tags()
                                                  );
        }

        //
        // 3. For each collection, check if the balancing policy recommends moving anything around.
        //

        for (vector<string>::const_iterator it = collections.begin(); it != collections.end(); ++it ) {
            const string& ns = *it;

            map< string,vector<BSONObj> > shardToChunksMap;
            cursor = conn.query(ConfigNS::chunk,
                                QUERY(ChunkFields::ns(ns)).sort(ChunkFields::min()));

            while ( cursor->more() ) {
                BSONObj chunk = cursor->nextSafe();
                vector<BSONObj>& chunks = shardToChunksMap[chunk[ChunkFields::shard()].String()];
                chunks.push_back( chunk.getOwned() );
            }
            cursor.reset();

            if (shardToChunksMap.empty()) {
                LOG(1) << "skipping empty collection (" << ns << ")";
                continue;
            }
            
            for ( vector<Shard>::iterator i=allShards.begin(); i!=allShards.end(); ++i ) {
                // this just makes sure there is an entry in shardToChunksMap for every shard
                Shard s = *i;
                shardToChunksMap[s.getName()].size();
            }

            DistributionStatus status( shardInfo, shardToChunksMap );

            // load tags
            conn.ensureIndex(ConfigNS::tag,
                             BSON(TagFields::ns() << 1 << TagFields::min() << 1),
                             true);

            cursor = conn.query(ConfigNS::tag,
                                QUERY(TagFields::ns(ns)).sort(TagFields::min()));

            while ( cursor->more() ) {
                BSONObj tag = cursor->nextSafe();
                uassert(16356 , str::stream() << "tag ranges not valid for: " << ns ,
                        status.addTagRange(TagRange(tag[TagFields::min()].Obj().getOwned(),
                                                    tag[TagFields::max()].Obj().getOwned(),
                                                    tag[TagFields::tag()].String())));

            }
            cursor.reset();
            
            CandidateChunk* p = _policy->balance( ns, status, _balancedLastTime );
            if ( p ) candidateChunks->push_back( CandidateChunkPtr( p ) );
        }
    }

    bool Balancer::_init() {
        try {

            log() << "about to contact config servers and shards" << endl;

            // contact the config server and refresh shard information
            // checks that each shard is indeed a different process (no hostname mixup)
            // these checks are redundant in that they're redone at every new round but we want to do them initially here
            // so to catch any problem soon
            Shard::reloadShardInfo();
            _checkOIDs();

            log() << "config servers and shards contacted successfully" << endl;

            StringBuilder buf;
            buf << getHostNameCached() << ":" << cmdLine.port;
            _myid = buf.str();
            _started = time(0);

            log() << "balancer id: " << _myid << " started at " << time_t_to_String_short(_started) << endl;

            return true;

        }
        catch ( std::exception& e ) {
            warning() << "could not initialize balancer, please check that all shards and config servers are up: " << e.what() << endl;
            return false;

        }
    }

    void Balancer::run() {

        // this is the body of a BackgroundJob so if we throw here we're basically ending the balancer thread prematurely
        while ( ! inShutdown() ) {

            if ( ! _init() ) {
                log() << "will retry to initialize balancer in one minute" << endl;
                sleepsecs( 60 );
                continue;
            }

            break;
        }

        int sleepTime = 30;

        // getConnectioString and dist lock constructor does not throw, which is what we expect on while
        // on the balancer thread
        ConnectionString config = configServer.getConnectionString();
        DistributedLock balanceLock( config , "balancer" );

        while ( ! inShutdown() ) {

            try {

                scoped_ptr<ScopedDbConnection> connPtr(
                        ScopedDbConnection::getInternalScopedDbConnection(config.toString(), 30));
                ScopedDbConnection& conn = *connPtr;

                // ping has to be first so we keep things in the config server in sync
                _ping( conn.conn() );

                // use fresh shard state
                Shard::reloadShardInfo();

                // refresh chunk size (even though another balancer might be active)
                Chunk::refreshChunkSize();

                BSONObj balancerConfig;
                // now make sure we should even be running
                if ( ! grid.shouldBalance( "", &balancerConfig ) ) {
                    LOG(1) << "skipping balancing round because balancing is disabled" << endl;

                    // Ping again so scripts can determine if we're active without waiting
                    _ping( conn.conn(), true );

                    conn.done();
                    
                    sleepsecs( sleepTime );
                    continue;
                }

                sleepTime = balancerConfig[SettingsFields::shortBalancerSleep()].trueValue() ? 30 :
                                                                                               6;
                
                uassert( 13258 , "oids broken after resetting!" , _checkOIDs() );

                {
                    dist_lock_try lk( &balanceLock , "doing balance round" );
                    if ( ! lk.got() ) {
                        LOG(1) << "skipping balancing round because another balancer is active" << endl;

                        // Ping again so scripts can determine if we're active without waiting
                        _ping( conn.conn(), true );

                        conn.done();
                        
                        sleepsecs( sleepTime ); // no need to wake up soon
                        continue;
                    }
                    
                    LOG(1) << "*** start balancing round" << endl;

                    if (balancerConfig["_waitForDelete"].trueValue()) {
                        LOG(1) << "balancer chunk moves will wait for cleanup" << endl;
                    }

                    if (balancerConfig["_secondaryThrottle"].trueValue()) {
                        LOG(1) << "balancer chunk moves will wait for secondaries" << endl;
                    }

                    vector<CandidateChunkPtr> candidateChunks;
                    _doBalanceRound( conn.conn() , &candidateChunks );
                    if ( candidateChunks.size() == 0 ) {
                        LOG(1) << "no need to move any chunk" << endl;
                        _balancedLastTime = 0;
                    }
                    else {
                        _balancedLastTime = _moveChunks(&candidateChunks,
                                balancerConfig[SettingsFields::secondaryThrottle()].trueValue(),
                                balancerConfig["_waitForDelete"].trueValue());
                    }

                    LOG(1) << "*** end of balancing round" << endl;
                }

                // Ping again so scripts can determine if we're active without waiting
                _ping( conn.conn(), true );
                
                conn.done();

                sleepsecs( _balancedLastTime ? sleepTime / 6 : sleepTime );
            }
            catch ( std::exception& e ) {
                log() << "caught exception while doing balance: " << e.what() << endl;

                // Just to match the opening statement if in log level 1
                LOG(1) << "*** End of balancing round" << endl;

                sleepsecs( sleepTime ); // sleep a fair amount b/c of error
                continue;
            }
        }

    }

} // namespace mongo
Something went wrong with that request. Please try again.