path: root/db/dur.cpp

// @file dur.cpp durability in the storage engine (crash-safeness / journaling)

/**
*    Copyright (C) 2009 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/>.
*/

/*
   phases:

     PREPLOGBUFFER
       we will build an output buffer ourself and then use O_DIRECT
       we could be in read lock for this
       for very large objects write directly to redo log in situ?
     WRITETOJOURNAL
       we could be unlocked (the main db lock that is...) for this, with sufficient care, but there is some complexity
         have to handle falling behind which would use too much ram (going back into a read lock would suffice to stop that).
         for now (1.7.5/1.8.0) we are in read lock which is not ideal.
     WRITETODATAFILES
       apply the writes back to the non-private MMF after they are for certain in redo log
     REMAPPRIVATEVIEW
       we could in a write lock quickly flip readers back to the main view, then stay in read lock and do our real
         remapping. with many files (e.g., 1000), remapping could be time consuming (several ms), so we don't want
         to be too frequent.
       there could be a slow down immediately after remapping as fresh copy-on-writes for commonly written pages will
         be required.  so doing these remaps fractionally is helpful. 

   mutexes:

     READLOCK dbMutex
     LOCK groupCommitMutex
       PREPLOGBUFFER()
     READLOCK mmmutex
       commitJob.reset()
     UNLOCK dbMutex                                     // now other threads can write
       WRITETOJOURNAL()
       WRITETODATAFILES()
     UNLOCK mmmutex
     UNLOCK groupCommitMutex

     on the next write lock acquisition for dbMutex:    // see MongoMutex::_acquiredWriteLock()
       REMAPPRIVATEVIEW()

     @see https://docs.google.com/drawings/edit?id=1TklsmZzm7ohIZkwgeK6rMvsdaR13KjtJYMsfLr175Zc
*/

#include "pch.h"
#include "cmdline.h"
#include "client.h"
#include "dur.h"
#include "dur_journal.h"
#include "dur_commitjob.h"
#include "dur_recover.h"
#include "dur_stats.h"
#include "../util/concurrency/race.h"
#include "../util/mongoutils/hash.h"
#include "../util/mongoutils/str.h"
#include "../util/timer.h"

using namespace mongoutils;

namespace mongo {

    namespace dur {

        void PREPLOGBUFFER(JSectHeader& outParm);
        void WRITETOJOURNAL(JSectHeader h, AlignedBuilder& uncompressed);
        void WRITETODATAFILES(const JSectHeader& h, AlignedBuilder& uncompressed);

        /** declared later in this file
            only used in this file -- use DurableInterface::commitNow() outside
        */
        static void groupCommit();

        CommitJob& commitJob = *(new CommitJob()); // don't destroy

        Stats stats;

        void Stats::S::reset() {
            memset(this, 0, sizeof(*this));
        }

        Stats::Stats() {
            _a.reset();
            _b.reset();
            curr = &_a;
            _intervalMicros = 3000000;
        }

        Stats::S * Stats::other() {
            return curr == &_a ? &_b : &_a;
        }
                        string _CSVHeader();

        string Stats::S::_CSVHeader() { 
            return "cmts  jrnMB\twrDFMB\tcIWLk\tearly\tprpLgB  wrToJ\twrToDF\trmpPrVw";
        }

        string Stats::S::_asCSV() { 
            stringstream ss;
            ss << 
                setprecision(2) << 
                _commits << '\t' << fixed << 
                _journaledBytes / 1000000.0 << '\t' << 
                _writeToDataFilesBytes / 1000000.0 << '\t' << 
                _commitsInWriteLock << '\t' << 
                _earlyCommits <<  '\t' << 
                (unsigned) (_prepLogBufferMicros/1000) << '\t' << 
                (unsigned) (_writeToJournalMicros/1000) << '\t' << 
                (unsigned) (_writeToDataFilesMicros/1000) << '\t' << 
                (unsigned) (_remapPrivateViewMicros/1000);
            return ss.str();
        }

        //int getAgeOutJournalFiles();
        BSONObj Stats::S::_asObj() {
            BSONObjBuilder b;
            b << 
                       "commits" << _commits <<
                       "journaledMB" << _journaledBytes / 1000000.0 <<
                       "writeToDataFilesMB" << _writeToDataFilesBytes / 1000000.0 <<
                       "compression" << _journaledBytes / (_uncompressedBytes+1.0) <<
                       "commitsInWriteLock" << _commitsInWriteLock <<
                       "earlyCommits" << _earlyCommits << 
                       "timeMs" <<
                       BSON( "dt" << _dtMillis <<
                             "prepLogBuffer" << (unsigned) (_prepLogBufferMicros/1000) <<
                             "writeToJournal" << (unsigned) (_writeToJournalMicros/1000) <<
                             "writeToDataFiles" << (unsigned) (_writeToDataFilesMicros/1000) <<
                             "remapPrivateView" << (unsigned) (_remapPrivateViewMicros/1000)
                           );
            /*int r = getAgeOutJournalFiles();
            if( r == -1 )
                b << "ageOutJournalFiles" << "mutex timeout";
            if( r == 0 )
                b << "ageOutJournalFiles" << false;*/
            if( cmdLine.journalCommitInterval != 0 )
                b << "journalCommitIntervalMs" << cmdLine.journalCommitInterval;
            return b.obj();
        }

        BSONObj Stats::asObj() {
            return other()->_asObj();
        }

        void Stats::rotate() {
            unsigned long long now = curTimeMicros64();
            unsigned long long dt = now - _lastRotate;
            if( dt >= _intervalMicros && _intervalMicros ) {
                // rotate
                curr->_dtMillis = (unsigned) (dt/1000);
                _lastRotate = now;
                curr = other();
                curr->reset();
            }
        }

        void NonDurableImpl::setNoJournal(void *dst, void *src, unsigned len) {
            memcpy(dst, src, len);
        }

        void DurableImpl::setNoJournal(void *dst, void *src, unsigned len) {
            // we are at least read locked, so we need not worry about REMAPPRIVATEVIEW herein.
            DEV dbMutex.assertAtLeastReadLocked();

            MemoryMappedFile::makeWritable(dst, len);

            // we enter the RecoveryJob mutex here, so that if WRITETODATAFILES is happening we do not 
            // conflict with it
            scoped_lock lk1( RecoveryJob::get()._mx );

            // we stay in this mutex for everything to work with DurParanoid/validateSingleMapMatches
            //
            // either of these mutexes also makes setNoJournal threadsafe, which is good as we call it from a read 
            // (not a write) lock in class SlaveTracking
            //
            scoped_lock lk( privateViews._mutex() );

            size_t ofs;
            MongoMMF *f = privateViews.find_inlock(dst, ofs);
            assert(f);
            void *w = (((char *)f->view_write())+ofs);
            // first write it to the writable (file) view
            memcpy(w, src, len);
            if( memcmp(w, dst, len) ) {
                // if we get here, a copy-on-write had previously occurred. so write it to the private view too
                // to keep them in sync.  we do this as we do not want to cause a copy on write unnecessarily.
                memcpy(dst, src, len);
            }
        }

        /** base declare write intent function that all the helpers call. */
        void DurableImpl::declareWriteIntent(void *p, unsigned len) {
            commitJob.note(p, len);
        }

        static DurableImpl* durableImpl = new DurableImpl();
        static NonDurableImpl* nonDurableImpl = new NonDurableImpl();
        DurableInterface* DurableInterface::_impl = nonDurableImpl;

        void DurableInterface::enableDurability() {
            assert(_impl == nonDurableImpl);
            _impl = durableImpl;
        }

        void DurableInterface::disableDurability() {
            assert(_impl == durableImpl);
            massert(13616, "can't disable durability with pending writes", !commitJob.hasWritten());
            _impl = nonDurableImpl;
        }

        bool DurableImpl::commitNow() {
            stats.curr->_earlyCommits++;
            groupCommit();
            return true;
        }

        bool DurableImpl::awaitCommit() {
            commitJob._notify.awaitBeyondNow();
            return true;
        }

        /** Declare that a file has been created
            Normally writes are applied only after journaling, for safety.  But here the file
            is created first, and the journal will just replay the creation if the create didn't
            happen because of crashing.
        */
        void DurableImpl::createdFile(string filename, unsigned long long len) {
            shared_ptr<DurOp> op( new FileCreatedOp(filename, len) );
            commitJob.noteOp(op);
        }

        void* DurableImpl::writingPtr(void *x, unsigned len) {
            void *p = x;
            declareWriteIntent(p, len);
            return p;
        }

        /** declare intent to write
            @param ofs offset within buf at which we will write
            @param len the length at ofs we will write
            @return new buffer pointer.
        */
        void* DurableImpl::writingAtOffset(void *buf, unsigned ofs, unsigned len) {
            char *p = (char *) buf;
            declareWriteIntent(p+ofs, len);
            return p;
        }

        void* DurableImpl::writingRangesAtOffsets(void *buf, const vector< pair< long long, unsigned > > &ranges ) {
            char *p = (char *) buf;
            for( vector< pair< long long, unsigned > >::const_iterator i = ranges.begin();
                    i != ranges.end(); ++i ) {
                declareWriteIntent( p + i->first, i->second );
            }
            return p;
        }

        bool DurableImpl::aCommitIsNeeded() const {
            DEV commitJob._nSinceCommitIfNeededCall = 0;
            return commitJob.bytes() > UncommittedBytesLimit;
        }

        bool DurableImpl::commitIfNeeded() {
            if ( ! dbMutex.isWriteLocked() ) // we implicitly commit if needed when releasing write lock
                return false;

            DEV commitJob._nSinceCommitIfNeededCall = 0;
            if (commitJob.bytes() > UncommittedBytesLimit) { // should this also fire if CmdLine::DurAlwaysCommit?
                stats.curr->_earlyCommits++;
                groupCommit();
                return true;
            }
            return false;
        }

        /** Used in _DEBUG builds to check that we didn't overwrite the last intent
            that was declared.  called just before writelock release.  we check a few
            bytes after the declared region to see if they changed.

            @see MongoMutex::_releasedWriteLock

            SLOW
        */
#if 0
        void DurableImpl::debugCheckLastDeclaredWrite() {
            static int n;
            ++n;

            assert(debug && cmdLine.dur);
            if (commitJob.writes().empty())
                return;
            const WriteIntent &i = commitJob.lastWrite();
            size_t ofs;
            MongoMMF *mmf = privateViews.find(i.start(), ofs);
            if( mmf == 0 )
                return;
            size_t past = ofs + i.length();
            if( mmf->length() < past + 8 )
                return; // too close to end of view
            char *priv = (char *) mmf->getView();
            char *writ = (char *) mmf->view_write();
            unsigned long long *a = (unsigned long long *) (priv+past);
            unsigned long long *b = (unsigned long long *) (writ+past);
            if( *a != *b ) {
                for( set<WriteIntent>::iterator it(commitJob.writes().begin()), end((commitJob.writes().begin())); it != end; ++it ) {
                    const WriteIntent& wi = *it;
                    char *r1 = (char*) wi.start();
                    char *r2 = (char*) wi.end();
                    if( r1 <= (((char*)a)+8) && r2 > (char*)a ) {
                        //log() << "it's ok " << wi.p << ' ' << wi.len << endl;
                        return;
                    }
                }
                log() << "journal data after write area " << i.start() << " does not agree" << endl;
                log() << " was:  " << ((void*)b) << "  " << hexdump((char*)b, 8) << endl;
                log() << " now:  " << ((void*)a) << "  " << hexdump((char*)a, 8) << endl;
                log() << " n:    " << n << endl;
                log() << endl;
            }
        }
#endif

        // Functor to be called over all MongoFiles

        class validateSingleMapMatches {
        public:
            validateSingleMapMatches(unsigned long long& bytes) :_bytes(bytes)  {}
            void operator () (MongoFile *mf) {
                if( mf->isMongoMMF() ) {
                    MongoMMF *mmf = (MongoMMF*) mf;
                    const unsigned char *p = (const unsigned char *) mmf->getView();
                    const unsigned char *w = (const unsigned char *) mmf->view_write();

                    if (!p || !w) return; // File not fully opened yet

                    _bytes += mmf->length();

                    assert( mmf->length() == (unsigned) mmf->length() );
                    {
                        scoped_lock lk( privateViews._mutex() ); // see setNoJournal
                        if (memcmp(p, w, (unsigned) mmf->length()) == 0)
                            return; // next file
                    }

                    unsigned low = 0xffffffff;
                    unsigned high = 0;
                    log() << "DurParanoid mismatch in " << mmf->filename() << endl;
                    int logged = 0;
                    unsigned lastMismatch = 0xffffffff;
                    for( unsigned i = 0; i < mmf->length(); i++ ) {
                        if( p[i] != w[i] ) {
                            if( lastMismatch != 0xffffffff && lastMismatch+1 != i )
                                log() << endl; // separate blocks of mismatches
                            lastMismatch= i;
                            if( ++logged < 60 ) {
                                if( logged == 1 )
                                    log() << "ofs % 628 = 0x" << hex << (i%628) << endl; // for .ns files to find offset in record
                                stringstream ss;
                                ss << "mismatch ofs:" << hex << i <<  "\tfilemap:" << setw(2) << (unsigned) w[i] << "\tprivmap:" << setw(2) << (unsigned) p[i];
                                if( p[i] > 32 && p[i] <= 126 )
                                    ss << '\t' << p[i];
                                log() << ss.str() << endl;
                            }
                            if( logged == 60 )
                                log() << "..." << endl;
                            if( i < low ) low = i;
                            if( i > high ) high = i;
                        }
                    }
                    if( low != 0xffffffff ) {
                        std::stringstream ss;
                        ss << "journal error warning views mismatch " << mmf->filename() << ' ' << (hex) << low << ".." << high << " len:" << high-low+1;
                        log() << ss.str() << endl;
                        log() << "priv loc: " << (void*)(p+low) << ' ' << endl;
                        set<WriteIntent>& b = commitJob.writes();
                        (void)b; // mark as unused. Useful for inspection in debugger

                        // should we abort() here so this isn't unnoticed in some circumstances?
                        massert(13599, "Written data does not match in-memory view. Missing WriteIntent?", false);
                    }
                }
            }
        private:
            unsigned long long& _bytes;
        };

        /** (SLOW) diagnostic to check that the private view and the non-private view are in sync.
        */
        void debugValidateAllMapsMatch() {
            if( ! (cmdLine.durOptions & CmdLine::DurParanoid) )
                return;

            unsigned long long bytes = 0;
            Timer t;
            MongoFile::forEach(validateSingleMapMatches(bytes));
            OCCASIONALLY log() << "DurParanoid map check " << t.millis() << "ms for " <<  (bytes / (1024*1024)) << "MB" << endl;
        }

        extern size_t privateMapBytes;

        /** We need to remap the private views periodically. otherwise they would become very large.
            Call within write lock.
        */
        void _REMAPPRIVATEVIEW() {
            // todo: Consider using ProcessInfo herein and watching for getResidentSize to drop.  that could be a way 
            //       to assure very good behavior here.

            static unsigned startAt;
            static unsigned long long lastRemap;

            dbMutex.assertWriteLocked();
            dbMutex._remapPrivateViewRequested = false;
            assert( !commitJob.hasWritten() );

            // we want to remap all private views about every 2 seconds.  there could be ~1000 views so
            // we do a little each pass; beyond the remap time, more significantly, there will be copy on write
            // faults after remapping, so doing a little bit at a time will avoid big load spikes on
            // remapping.
            unsigned long long now = curTimeMicros64();
            double fraction = (now-lastRemap)/2000000.0;
            if( cmdLine.durOptions & CmdLine::DurAlwaysRemap )
                fraction = 1;
            lastRemap = now;

#if defined(_WIN32)
            // Note that this negatively affects performance.
            // We must grab the exclusive lock here because remapThePrivateView() on Windows needs
            // to grab it as well, due to the lack of a non-atomic way to remap a memory mapped file.
            // See SERVER-5723 for performance improvement.
            // See SERVER-5680 to see why this code is necessary.
            RWLockRecursive::Exclusive lk(MongoFile::mmmutex);
#else
            RWLockRecursive::Shared lk(MongoFile::mmmutex);
#endif

            set<MongoFile*>& files = MongoFile::getAllFiles();
            unsigned sz = files.size();
            if( sz == 0 )
                return;

            {
                // be careful not to use too much memory if the write rate is 
                // extremely high
                double f = privateMapBytes / ((double)UncommittedBytesLimit);
                if( f > fraction ) { 
                    fraction = f;
                }
                privateMapBytes = 0;
            }

            unsigned ntodo = (unsigned) (sz * fraction);
            if( ntodo < 1 ) ntodo = 1;
            if( ntodo > sz ) ntodo = sz;

            const set<MongoFile*>::iterator b = files.begin();
            const set<MongoFile*>::iterator e = files.end();
            set<MongoFile*>::iterator i = b;
            // skip to our starting position
            for( unsigned x = 0; x < startAt; x++ ) {
                i++;
                if( i == e ) i = b;
            }
            startAt = (startAt + ntodo) % sz; // mark where to start next time

            for( unsigned x = 0; x < ntodo; x++ ) {
                dassert( i != e );
                if( (*i)->isMongoMMF() ) {
                    MongoMMF *mmf = (MongoMMF*) *i;
                    assert(mmf);
                    if( mmf->willNeedRemap() ) {
                        mmf->willNeedRemap() = false;
                        mmf->remapThePrivateView();
                    }
                    i++;
                    if( i == e ) i = b;
                }
            }
        }
        void REMAPPRIVATEVIEW() {
            Timer t;
            _REMAPPRIVATEVIEW();
            stats.curr->_remapPrivateViewMicros += t.micros();
        }

        // lock order: dbMutex first, then this
        mutex groupCommitMutex("groupCommit");

        bool _groupCommitWithLimitedLocks() {
            scoped_ptr<readlocktry> lk1( new readlocktry("", 500) );
            if( !lk1->got() )
                return false;

            scoped_lock lk2(groupCommitMutex);

            commitJob.beginCommit();

            if( !commitJob.hasWritten() ) {
                // getlasterror request could have came after the data was already committed
                commitJob.notifyCommitted();
                return true;
            }
            JSectHeader h;
            PREPLOGBUFFER(h);

            RWLockRecursive::Shared lk3(MongoFile::mmmutex);

            unsigned abLen = commitJob._ab.len();
            commitJob.reset(); // must be reset before allowing anyone to write
            DEV assert( !commitJob.hasWritten() );

            // release the readlock -- allowing others to now write while we are writing to the journal (etc.)
            lk1.reset();

            // ****** now other threads can do writes ******
            WRITETOJOURNAL(h, commitJob._ab);
            assert( abLen == commitJob._ab.len() ); // a check that no one touched the builder while we were doing work. if so, our locking is wrong.

            // data is now in the journal, which is sufficient for acknowledging getLastError.
            // (ok to crash after that)
            commitJob.notifyCommitted();

            WRITETODATAFILES(h, commitJob._ab);
            assert( abLen == commitJob._ab.len() ); // check again wasn't modded
            commitJob._ab.reset();

            // can't : dbMutex._remapPrivateViewRequested = true;

            return true;
        }

       /** @return true if committed; false if lock acquisition timed out (we only try for a read lock herein and only wait for a certain duration). */
       bool groupCommitWithLimitedLocks() {
           try {
               return _groupCommitWithLimitedLocks();
           }
           catch(DBException& e ) {
               log() << "dbexception in groupCommitLL causing immediate shutdown: " << e.toString() << endl;
               mongoAbort("dur1");
           }
           catch(std::ios_base::failure& e) {
               log() << "ios_base exception in groupCommitLL causing immediate shutdown: " << e.what() << endl;
               mongoAbort("dur2");
           }
           catch(std::bad_alloc& e) {
               log() << "bad_alloc exception in groupCommitLL causing immediate shutdown: " << e.what() << endl;
               mongoAbort("dur3");
           }
           catch(std::exception& e) {
               log() << "exception in dur::groupCommitLL causing immediate shutdown: " << e.what() << endl;
               mongoAbort("dur4");
           }
           return false;
       }

       /** locking: in read lock when called. */
        static void _groupCommit() {
            commitJob.beginCommit();

            if( !commitJob.hasWritten() ) {
                // getlasterror request could have came after the data was already committed
                commitJob.notifyCommitted();
                return;
            }

            // we need to make sure two group commits aren't running at the same time
            // (and we are only read locked in the dbMutex, so it could happen)
            scoped_lock lk(groupCommitMutex);

            JSectHeader h;
            PREPLOGBUFFER(h);

            // todo : write to the journal outside locks, as this write can be slow.
            //        however, be careful then about remapprivateview as that cannot be done 
            //        if new writes are then pending in the private maps.
            WRITETOJOURNAL(h, commitJob._ab);

            // data is now in the journal, which is sufficient for acknowledging getLastError.
            // (ok to crash after that)
            commitJob.notifyCommitted();

            WRITETODATAFILES(h, commitJob._ab);
            debugValidateAllMapsMatch();

            commitJob.reset();
            commitJob._ab.reset();

            // REMAPPRIVATEVIEW
            //
            // remapping private views must occur after WRITETODATAFILES otherwise
            // we wouldn't see newly written data on reads.
            //
            DEV assert( !commitJob.hasWritten() );
            if( !dbMutex.isWriteLocked() ) {
                // this needs done in a write lock (as there is a short window during remapping when each view 
                // might not exist) thus we do it on the next acquisition of that instead of here (there is no 
                // rush if you aren't writing anyway -- but it must happen, if it is done, before any uncommitted 
                // writes occur).  If desired, perhaps this can be eliminated on posix as it may be that the remap 
                // is race-free there.
                //
                dbMutex._remapPrivateViewRequested = true;
            }
            else {
                stats.curr->_commitsInWriteLock++;
                // however, if we are already write locked, we must do it now -- up the call tree someone
                // may do a write without a new lock acquisition.  this can happen when MongoMMF::close() calls
                // this method when a file (and its views) is about to go away.
                //
                REMAPPRIVATEVIEW();
            }
        }

        /** locking: in read lock when called
                     or, for early commits (commitIfNeeded), in write lock
            @see MongoMMF::close()
        */
        static void groupCommit() {
            // we need to be at least read locked on the dbMutex so that we know the write intent data 
            // structures are not changing while we work
            dbMutex.assertAtLeastReadLocked();

            try {
                _groupCommit();
            }
            catch(DBException& e ) { 
                log() << "dbexception in groupCommit causing immediate shutdown: " << e.toString() << endl;
                mongoAbort("gc1");
            }
            catch(std::ios_base::failure& e) { 
                log() << "ios_base exception in groupCommit causing immediate shutdown: " << e.what() << endl;
                mongoAbort("gc2");
            }
            catch(std::bad_alloc& e) { 
                log() << "bad_alloc exception in groupCommit causing immediate shutdown: " << e.what() << endl;
                mongoAbort("gc3");
            }
            catch(std::exception& e) {
                log() << "exception in dur::groupCommit causing immediate shutdown: " << e.what() << endl;
                mongoAbort("gc4");
            }
        }

        static void go() {
            const int N = 10;
            static int n;
            if( privateMapBytes < UncommittedBytesLimit && ++n % N && (cmdLine.durOptions&CmdLine::DurAlwaysRemap)==0 ) {
                // limited locks version doesn't do any remapprivateview at all, so only try this if privateMapBytes
                // is in an acceptable range.  also every Nth commit, we do everything so we can do some remapping;
                // remapping a lot all at once could cause jitter from a large amount of copy-on-writes all at once.
                if( groupCommitWithLimitedLocks() )
                    return;
            }
            else {
                readlocktry lk("", 1000);
                if( lk.got() ) {
                    groupCommit();
                    return;
                }
            }

            // starvation on read locks could occur.  so if read lock acquisition is slow, try to get a
            // write lock instead.  otherwise journaling could be delayed too long (too much data will 
            // not accumulate though, as commitIfNeeded logic will have executed in the meantime if there 
            // has been writes)
            writelock lk;
            groupCommit();
        }

        /** called when a MongoMMF is closing -- we need to go ahead and group commit in that case before its
            views disappear
        */
        void closingFileNotification() {
            if (!cmdLine.dur)
                return;

            if( dbMutex.atLeastReadLocked() ) {
                groupCommit();
            }
            else {
                assert( inShutdown() );
                if( commitJob.hasWritten() ) {
                    log() << "journal warning files are closing outside locks with writes pending" << endl;
                }
            }
        }

        filesystem::path getJournalDir();

        void durThread() {
            Client::initThread("journal");

            bool samePartition = true;
            try {
                const string dbpathDir = boost::filesystem::path(dbpath).native_directory_string();
                samePartition = onSamePartition(getJournalDir().string(), dbpathDir);
            }
            catch(...) {
            }

            while( !inShutdown() ) {
                unsigned ms = cmdLine.journalCommitInterval;
                if( ms == 0 ) { 
                    // use default
                    ms = samePartition ? 100 : 30;
                }

                unsigned oneThird = (ms / 3) + 1; // +1 so never zero

                try {
                    stats.rotate();

                    // we do this in a couple blocks (the invoke()), which makes it a tiny bit faster (only a little) on throughput,
                    // but is likely also less spiky on our cpu usage, which is good.

                    // commit sooner if one or more getLastError j:true is pending
                    sleepmillis(oneThird);
                    for( unsigned i = 1; i <= 2; i++ ) {
                        if( commitJob._notify.nWaiting() )
                            break;
                        commitJob.wi()._deferred.invoke();
                        sleepmillis(oneThird);
                    }

                    go();
                }
                catch(std::exception& e) {
                    log() << "exception in durThread causing immediate shutdown: " << e.what() << endl;
                    mongoAbort("exception in durThread");
                }
            }
            cc().shutdown();
        }

        void recover();

        void releasingWriteLock() {
            // implicit commitIfNeeded check on each write unlock
            DEV commitJob._nSinceCommitIfNeededCall = 0; // implicit commit if needed
            if( commitJob.bytes() > UncommittedBytesLimit || cmdLine.durOptions & CmdLine::DurAlwaysCommit ) {
                stats.curr->_earlyCommits++;
                groupCommit();
            }
        }

        void preallocateFiles();

        /** at startup, recover, and then start the journal threads */
        void startup() {
            if( !cmdLine.dur )
                return;

#if defined(_DURABLEDEFAULTON)
            DEV { 
                if( time(0) & 1 ) {
                    cmdLine.durOptions |= CmdLine::DurAlwaysCommit;
                    log() << "_DEBUG _DURABLEDEFAULTON : forcing DurAlwaysCommit mode for this run" << endl;
                }
                if( time(0) & 2 ) {
                    cmdLine.durOptions |= CmdLine::DurAlwaysRemap;
                    log() << "_DEBUG _DURABLEDEFAULTON : forcing DurAlwaysRemap mode for this run" << endl;
                }
            }
#endif

            DurableInterface::enableDurability();

            journalMakeDir();
            try {
                recover();
            }
            catch(...) {
                log() << "exception during recovery" << endl;
                throw;
            }

            preallocateFiles();

            boost::thread t(durThread);
        }

        void DurableImpl::syncDataAndTruncateJournal() {
            dbMutex.assertWriteLocked();

            // a commit from the commit thread won't begin while we are in the write lock,
            // but it may already be in progress and the end of that work is done outside 
            // (dbMutex) locks. This line waits for that to complete if already underway.
            {
                scoped_lock lk(groupCommitMutex);
            }

            groupCommit();
            MongoFile::flushAll(true);
            journalCleanup();

            assert(!haveJournalFiles()); // Double check post-conditions
        }

    } // namespace dur

} // namespace mongo