privacore-open-source-searc.../Multicast.cpp

1025 lines
36 KiB
C++

#include "Multicast.h"
#include "UdpSlot.h"
#include "UdpServer.h"
#include "Hostdb.h"
#include "Stats.h"
#include "Conf.h"
#include "Loop.h" // registerSleepCallback()
#include "ScopedLock.h"
#include "Process.h"
#include "ip.h"
#include "Mem.h"
#include "Msg0.h" //msg+MSG0RDBIDOFFSET
#include "Errno.h"
#include "fctypes.h"
#include "hash.h"
#include <cerrno>
// TODO: if we're ordered to close and we still are waiting on stuff
// to send we should send as much as we can and save the remaining
// slots to disk for sending later??
void Multicast::constructor() {
m_msg = NULL;
m_readBuf = NULL;
m_inUse = false;
for(int i=0; i<MAX_HOSTS_PER_GROUP; i++)
m_host[i].reset();
}
void Multicast::destructor() {
reset();
}
Multicast::Multicast()
: m_msg(NULL),
m_msgSize(0),
m_msgType((msg_type_t)-1),
m_ownMsg(false),
m_slot(NULL),
m_inUse(false),
m_replyingHost(NULL),
m_replyLaunchTime(0),
m_mtx(),
m_state(NULL), m_state2(NULL),
m_callback(NULL),
m_totalTimeout(0),
m_startTime(0),
m_numReplies(0),
m_host(),
m_numHosts(0),
m_readBuf(NULL),
m_readBufSize(0),
m_readBufMaxSize(0),
m_ownReadBuf(false),
m_registeredSleep(false),
m_niceness(0),
m_lastLaunch(0),
m_freeReadBuf(false),
m_key(0),
m_sentToTwin(false)
{
constructor();
}
Multicast::~Multicast() {
reset();
}
// free the send/read (request/reply) bufs we pirated from a UdpSlot or
// got from the caller
void Multicast::reset ( ) {
// if this is called while we are shutting down and Scraper has a
// MsgE out it cores
if ( m_inUse && ! g_process.m_exiting ) {
log(LOG_ERROR, "multicast: Resetting multicast which is in use. msgType=0x%02x", (int)m_msgType);
g_process.shutdownAbort(true);
}
if ( m_msg && m_ownMsg )
mfree ( m_msg , m_msgSize , "Multicast" );
if ( m_readBuf && m_ownReadBuf && m_freeReadBuf )
mfree ( m_readBuf , m_readBufMaxSize , "Multicast" );
m_msg = NULL;
m_readBuf = NULL;
m_inUse = false;
m_replyingHost = NULL;
for(int i=0; i<MAX_HOSTS_PER_GROUP; i++)
m_host[i].reset();
}
// . an individual transaction's udpSlot is not be removed because we might
// get it a reply from it later after it's timeout
// . returns false and sets g_errno on error
// . caller can now pass in his own reply buffer
// . if "freeReplyBuf" is true that means it needs to be freed at some point
// otherwise, it's probably on the stack or part of a larger allocate class.
bool Multicast::send(char *msg, int32_t msgSize, msg_type_t msgType, bool ownMsg, uint32_t shardNum, bool sendToWholeGroup_,
int32_t key, void *state, void *state2, void (*callback)(void *state, void *state2),
int64_t totalTimeout, int32_t niceness, int32_t firstHostId, bool freeReplyBuf) {
// make sure not being re-used!
if ( m_inUse ) {
log( LOG_ERROR, "multicast: Attempt to re-use active multicast");
g_process.shutdownAbort(true);
}
reset();
// it is now in use
m_inUse = true;
// set the parameters in this class
m_msg = msg;
m_ownMsg = ownMsg;
m_ownReadBuf = true;
m_freeReadBuf = freeReplyBuf;
m_msgSize = msgSize;
m_msgType = msgType;
m_state = state;
m_state2 = state2;
m_callback = callback;
m_totalTimeout = totalTimeout; // in milliseconds
m_niceness = niceness;
// this can't be -1 i guess
if ( totalTimeout <= 0 ) { g_process.shutdownAbort(true); }
m_startTime = gettimeofdayInMilliseconds();
m_numReplies = 0;
m_readBuf = NULL;
m_readBufSize = 0;
m_readBufMaxSize = 0;
m_registeredSleep = false;
m_sentToTwin = false;
m_key = key;
// clear m_retired, m_errnos, m_slots
for(int i=0; i<MAX_HOSTS_PER_GROUP; i++)
m_host[i].reset();
getCandidateHostList(shardNum, msgType, msg, msgSize);
// . pick the fastest host in the group
// . this should pick the fastest one we haven't already sent to yet
if ( ! sendToWholeGroup_ ) {
bool retVal = sendToHostLoop(key,firstHostId);
// on error, un-use this class
if ( ! retVal ) {
m_inUse = false;
}
return retVal;
} else {
// . send to ALL hosts in this group if sendToWholeGroup is true
// . blocks forever until sends to all hosts are successfull
sendToWholeGroup();
// . sendToWholeGroup() always blocks, but we return true if no g_errno
// . we actually keep looping until all hosts get the msg w/o error
return true;
}
}
//Get the list of hosts in the shard, filtering out unwanted hosts
//Used for filtering out no-spider hosts when sending msg0-get-list with db=spiderdb.
void Multicast::getCandidateHostList(uint32_t shardNum, msg_type_t msgType, const char *msg, int32_t /*msgSize*/) {
bool requireQueryEnabled = false; //logical preparation for when we eliminate dbs on spider-only hosts
bool requireSpiderEnabled = false;
if(msgType==msg_type_0) {
rdbid_t rdbid = (rdbid_t)*(msg+MSG0RDBIDOFFSET);
if(rdbid==RDB_SPIDERDB_DEPRECATED) {
//spiderdb is only present on hosts with spidering enabled
requireSpiderEnabled = true;
}
}
int32_t numCandidateHosts = 0;
Host *candidateHosts = g_hostdb.getShard(shardNum, &numCandidateHosts);
m_numHosts = 0;
for(int i=0; i<numCandidateHosts; i++) {
if(requireQueryEnabled && !candidateHosts[i].m_queryEnabled)
continue;
if(requireSpiderEnabled && !candidateHosts[i].m_spiderEnabled)
continue;
if(candidateHosts[i].m_retired)
continue;
m_host[m_numHosts].m_hostPtr = candidateHosts+i;
m_numHosts++;
}
}
///////////////////////////////////////////////////////
// //
// GROUP SEND //
// //
///////////////////////////////////////////////////////
// . keeps calling itself back on any error
// . resends to host/ip's that had error forever
// . callback only called when all hosts transmission are successful
// . it does not send to hosts whose m_errnos is 0
// . TODO: deal with errors from g_udpServer::sendRequest() better
// . returns false and sets g_errno on error
void Multicast::sendToWholeGroup() {
ScopedLock sl(m_mtx);
// see if anyone gets an error
bool hadError = false;
// . cast the msg to ALL hosts in the m_hosts group of hosts
for ( int32_t i = 0 ; i < m_numHosts ; i++ ) {
// cancel any errors
g_errno = 0;
// get the host
Host *h = m_host[i].m_hostPtr;
// if we got a nice reply from him skip him
//slots[i] && m_host[i].m_slot->doneReading() ) continue;
if ( m_host[i].m_retired ) continue;
// . timeout is in seconds
// . timeout is just the time remaining for the whole groupcast
// int32_t timeout = m_startTime + m_totalTimeout - getTime();
// . since we now must get non-error replies from ALL hosts
// in the group we no longer have a "totalTimeout" per se
// reset the g_errno for host #i
m_host[i].m_errno = 0;
// send to the same port as us!
int16_t destPort = h->m_port;
// if from hosts2.conf pick the best ip!
//int32_t bestIp = h->m_ip;
int32_t bestIp = g_hostdb.getBestHosts2IP ( h );
// retire the host to prevent resends
m_host[i].m_retired = true;
int32_t hid = h->m_hostId;
// . send to a single host
// . this creates a transaction control slot, "udpSlot"
// . returns false and sets g_errno on error
if (g_udpServer.sendRequest(m_msg, m_msgSize, m_msgType, bestIp, destPort, hid, &m_host[i].m_slot, this, gotReply2, m_totalTimeout, m_niceness)) {
continue;
}
// g_errno must have been set, remember it
m_host[i].m_errno = g_errno;
// we had an error
hadError = true;
// bring him out of retirement to try again later in time
m_host[i].m_retired = false;
// log the error
log(LOG_WARN, "multicast: Got error sending add data request (0x%02x) to host #%d: %s. Sleeping one second and retrying.",
(int)m_msgType, h->m_hostId, mstrerror(g_errno));
// . clear it, we'll try again
// . if we don't clear Msg1::addList(), which returns
// true if it did not block, false if it did, will pick up
// on it and wierd things might happen.
g_errno = 0;
// continue if we're already registered for sleep callbacks
if ( m_registeredSleep ) continue;
// otherwise register for sleep callback to try again
g_loop.registerSleepCallback(5000, this, sleepWrapper2, "Multicast::sleepWrapper2", m_niceness);
m_registeredSleep = true;
}
// if we had an error then we'll be called again in a second
if ( hadError ) return;
// otherwise, unregister sleep callback if we had no error
if ( m_registeredSleep ) {
g_loop.unregisterSleepCallback ( this , sleepWrapper2 );
m_registeredSleep = false;
}
}
void Multicast::sleepWrapper2(int bogusfd, void *state) {
Multicast *THIS = static_cast<Multicast*>(state);
// try another round of sending to see if hosts had errors or not
THIS->sendToWholeGroup();
}
void Multicast::gotReply2(void *state, UdpSlot *slot) {
Multicast *THIS = static_cast<Multicast*>(state);
THIS->gotReply2(slot);
}
// . otherwise, we were sending to a whole group so ALL HOSTS must produce a
// successful reply
// . we keep re-trying forever until they do
void Multicast::gotReply2 ( UdpSlot *slot ) {
ScopedLock sl(m_mtx);
// don't ever let UdpServer free this send buf (it is m_msg)
slot->m_sendBufAlloc = NULL;
// save this for msg4 logic that calls injection callback
m_slot = slot;
// . log the error
// . ETRYAGAIN often happens when we are falling too far behind in
// our merging (see Rdb.cpp) and we enter urgent merge mode
// . it may also happen if tree is too full and is being dumped to disk
//if ( g_errno && g_errno != ETRYAGAIN )
// log("net: Got error reply sending to a host during a "
// "group send: %s.", mstrerror(g_errno) );
// set m_errnos for this slot
int32_t i;
for(i = 0; i < m_numHosts; i++)
if(m_host[i].m_slot == slot)
break;
if ( i == m_numHosts ) {
// if it matched no slot that's wierd
log(LOG_LOGIC,"multicast: Not our slot=%p", slot);
return;
}
// set m_errnos to g_errno, if any
m_host[i].m_errno = g_errno;
// if g_errno was not set we have a legit reply
if ( ! g_errno )
m_numReplies++;
// reset g_errno in case we do more sending
g_errno = 0;
// . if we got all the legit replies we're done, call the callback
// . all slots should be destroyed by UdpServer in this case
if ( m_numReplies >= m_numHosts ) {
// allow us to be re-used now, callback might relaunch
m_inUse = false;
if ( m_callback ) {
sl.unlock();
m_callback ( m_state , m_state2 );
}
return;
}
// if this guy had no error then wait for more callbacks
if(!m_host[i].m_errno)
return;
// bring this slot out of retirement so we can send to him again
m_host[i].m_retired = false;
// do indeed log the try again things, cuz we have gotten into a
// nasty loop with them that took me a while to track down
const Host *h = slot->m_host;
if ( h )
logDebug(g_conf.m_logDebugMulticast, "multicast: Got error sending request to hostId %d (msgType=0x%02x transId=%d): %s. Retrying.",
h->m_hostId, (int)slot->getMsgType(), slot->getTransId(), mstrerror(m_host[i].m_errno) );
else {
char ipbuf[16];
logDebug(g_conf.m_logDebugMulticast, "multicast: Got error sending request to %s:%d (msgType=0x%02x transId=%d): %s. Retrying.",
iptoa(slot->getIp(),ipbuf), (int32_t)slot->getPort(), (int)slot->getMsgType(), slot->getTransId(),
mstrerror(m_host[i].m_errno) );
}
// . let's sleep for a second before retrying the send
// . the g_errno could be ETRYAGAIN which happens if we're trying to
// add data but the other host is temporarily full
// . continue if we're already registered for sleep callbacks
if(m_registeredSleep)
return;
// . otherwise register for sleep callback to try again
// . sleepWrapper2() will call sendToWholeGroup() for us
g_loop.registerSleepCallback(5000/*ms*/, this, sleepWrapper2, "Multicast::sleepWrapper2", m_niceness);
m_registeredSleep = true;
}
///////////////////////////////////////////////////////
// //
// PICK & SEND //
// //
///////////////////////////////////////////////////////
// . returns false and sets g_errno on error
// . returns true if managed to send to one host ok (initially at least)
// . uses key to pick the first host to send to (for consistency)
// . after we pick a host and launch the request to him the sleepWrapper1
// will call this at regular intervals, so be careful,
bool Multicast::sendToHostLoop(int32_t key, int32_t firstHostId) {
// erase any errors we may have got
g_errno = 0 ;
for (;;) {
// what if this host is dead?!?!?
int32_t i = pickBestHost(key, firstHostId);
// . if no more hosts return FALSE
// . we need to return false to the caller of us below
if (i < 0) {
g_errno = ENOHOSTS;
return false;
}
// do not resend to retired hosts
if( m_host[i].m_retired ) {
g_errno = ENOHOSTS;
return false;
}
// . send to this guy, if we haven't yet
// . returns false and sets g_errno on error
// . if it returns true, we sent ok, so we should return true
// . will return false if the whole thing is timed out and g_errno
// will be set to ETIMEDOUT
// . i guess ENOSLOTS means the udp server has no slots available
// for sending, so its pointless to try to send to another host
if (sendToHost(i)) {
return true;
}
// . if no more slots, we're done, don't loop!
// . pointless as well if no time left in the multicast
// . or if shutting down the server! otherwise it loops forever and
// won't exit when sending a msg20 request. i've seen this...
if (g_errno == ENOSLOTS || g_errno == EUDPTIMEDOUT || g_errno == ESHUTTINGDOWN) {
return false;
}
// otherwise try another host and hope for the best
g_errno = 0;
key = 0;
// what kind of error leads us here? EBUFTOOSMALL or EBADENGINEER...
}
}
// . pick the fastest host from m_hosts based on avg roundtrip time for ACKs
// . skip hosts in our m_retired[] list of hostIds
// . returns -1 if none left to pick
int32_t Multicast::pickBestHost ( uint32_t key , int32_t firstHostId ) {
// bail if no hosts
if ( m_numHosts == 0 ) {
return -1;
}
// . if firstHostId not -1, try it first
// . Msg0 uses this only to select hosts on same machine for now
// . Msg20 now uses this to try to make sure the lower half of docids
// go to one twin and the upper half to the other. this makes the
// tfndb page cache twice as effective when getting summaries.
if ( firstHostId >= 0 ) {
//log("got first hostId!!!!");
// find it in group
int32_t i;
for ( i = 0 ; i < m_numHosts ; i++ )
if ( m_host[i].m_hostPtr->m_hostId == firstHostId ) break;
// if not found bitch
if ( i >= m_numHosts ) {
log(LOG_LOGIC,"multicast: HostId %" PRId32" not in group.", firstHostId );
g_process.shutdownAbort(true);
}
// if we got a match and it's not dead, return it
if(g_hostdb.mayWeSendRequestToHost(m_host[i].m_hostPtr,m_msgType) &&
!g_hostdb.isDead(m_host[i].m_hostPtr))
return i;
}
// round robin selection
int32_t numDead = 0;
int32_t dead = -1;
int32_t n = 0;
//int32_t count = 0;
bool balance = g_conf.m_doStripeBalancing;
// always turn off stripe balancing for all but these msgTypes
if ( m_msgType != msg_type_39 ) {
balance = false;
}
// . pick the guy in our "stripe" first if we are doing these msgs
// . this will prevent a ton of msg39s from hitting one host and
// "spiking" it.
if ( balance ) n = g_hostdb.m_myHost->m_stripe;
// . if key is not zero, use it to select a host in this group
// . if the host we want is dead then do it the old way
// . ignore the key if balance is true though! MDW
if ( key != 0 && ! balance ) {
// often the groupId was selected based on the key, so lets
// randomize everything up a bit
uint32_t i = hashLong ( key ) % m_numHosts;
// if he's not dead or retired use him right away
if ( ! m_host[i].m_retired &&
g_hostdb.mayWeSendRequestToHost(m_host[i].m_hostPtr,m_msgType) &&
! g_hostdb.isDead ( m_host[i].m_hostPtr ) )
return i;
}
// no no no we need to randomize the order that we try them
Host *fh = m_host[n].m_hostPtr;
// if this host is not dead, use him
if ( ! m_host[n].m_retired &&
g_hostdb.mayWeSendRequestToHost(m_host[n].m_hostPtr,m_msgType) &&
! g_hostdb.isDead(fh) )
return n;
// . ok now select the kth available host
// . make a list of the candidates
int32_t cand[32];
int32_t nc = 0;
for ( int32_t i = 0 ; i < m_numHosts ; i++ ) {
// get the host
Host *h = m_host[i].m_hostPtr;
// count those that are dead
if ( g_hostdb.isDead ( h ) )
numDead++;
// skip host if we've retired it
if ( m_host[i].m_retired ) continue;
// if this host is not dead use him
if(!g_hostdb.mayWeSendRequestToHost(m_host[i].m_hostPtr,m_msgType))
continue;
if ( !g_hostdb.isDead(h) )
cand[nc++] = i;
// pick a dead that isn't retired
dead = i;
}
// if a host was alive and untried, use him next
if ( nc > 0 ) {
int32_t k = ((uint32_t)m_key) % nc;
return cand[k];
}
// . come here if all hosts were DEAD
// . try sending to a host that is dead, but not retired now
// . if all deadies are retired this will return -1
// . sometimes a host can appear to be dead even though it was
// just under severe load
if ( numDead == m_numHosts ) return dead;
// otherwise, they weren't all dead so don't send to a deadie
return -1;
}
// . returns false and sets error on g_errno
// . returns true if kicked of the request (m_msg)
// . sends m_msg to host "h"
bool Multicast::sendToHost ( int32_t i ) {
// sanity check
if ( i >= m_numHosts ) { g_process.shutdownAbort(true); }
// sanity check , bitch if retired
if ( m_host[i].m_retired ) {
log(LOG_LOGIC,"multicast: Host #%d is retired. Bad engineer.",i);
//g_process.shutdownAbort(true);
return true;
}
// . add this host to our retired list so we don't try again
// . only used by pickBestHost() and sendToHost()
m_host[i].m_retired = true;
// what time is it now?
int64_t nowms = gettimeofdayInMilliseconds();
// save the time
m_host[i].m_launchTime = nowms;
// sometimes clock is updated on us
if ( m_startTime > nowms )
m_startTime = nowms;
// . timeout is in seconds
// . timeout is just the time remaining for the whole groupcast
int64_t timeRemaining = m_startTime + m_totalTimeout - nowms;
// . if timeout is negative then reset start time so we try forever
// . no, this could be called by a re-route in sleepWrapper1 in which
// case we really should timeout.
// . this can happen if sleepWrapper found a timeout before UdpServer
// got its timeout.
if ( timeRemaining <= 0 ) {
//m_startTime = getTime();; timeout = m_totalTimeout;}
//g_errno = ETIMEDOUT;
// this can happen if the udp reply timed out!!! like if a
// host is under severe load... with Msg23::getLinkText()
// or Msg22::getTitleRec() timing out on us. basically, our
// msg23 request tried to send a msg22 request which timed out
// on it so it sent us back this error.
if ( g_errno != EUDPTIMEDOUT )
log(LOG_INFO,"multicast: had negative timeout, %" PRId64". "
"startTime=%" PRId64" totalTimeout=%" PRId64" "
"now=%" PRId64". msgType=0x%02x "
"niceness=%" PRId32" clock updated?",
timeRemaining,m_startTime,m_totalTimeout,
nowms, (int)m_msgType,
(int32_t)m_niceness);
timeRemaining = m_totalTimeout;
}
// get the host
Host *h = m_host[i].m_hostPtr;
// send to the same port as us!
int16_t destPort = h->m_port;
// if from hosts2.conf pick the best ip!
int32_t bestIp = g_hostdb.getBestHosts2IP ( h );
// sanity check
//if ( g_hostdb.isDead(h) ) {
// log("net: trying to send to dead host.");
// g_process.shutdownAbort(true);
//}
// don't set hostid if we're sending to a remote cluster
int32_t hid = h->m_hostId;
// if sending to a proxy keep this set to -1
if ( h->m_type != HT_GRUNT ) hid = -1;
// max resends. if we resend a request dgram this many times and
// got no ack, bail out with g_errno set to ENOACK. this is better
// than the timeout because it takes like 20 seconds to mark a
// host as dead and takes "timeRemaining" seconds to timeout the
// request
int32_t maxResends = -1;
// . only use for nicness 0
// . it uses a backoff scheme, increments delay for first few resends:
// . it starts of at 33ms, then 66, then 132, then 200 from there out
if ( m_niceness == 0 ) maxResends = 4;
// . send to a single host
// . this creates a transaction control slot, "udpSlot"
// . return false and sets g_errno on error
// . returns true on successful launch and calls callback on completion
ScopedLock sl(m_mtx);
if (!g_udpServer.sendRequest(m_msg, m_msgSize, m_msgType, bestIp, destPort, hid, &m_host[i].m_slot, this, gotReply1, timeRemaining, m_niceness, NULL, maxResends)) {
log(LOG_WARN, "multicast: Had error sending msgtype 0x%02x to host #%d: %s. Not retrying.",
(int)m_msgType, h->m_hostId, mstrerror(g_errno));
// i've seen ENOUDPSLOTS available msg here along with oom
// condition...
//g_process.shutdownAbort(true);
return false;
}
// mark it as outstanding
m_host[i].m_inProgress = true;
// set our last launch date
m_lastLaunch = nowms ; // gettimeofdayInMilliseconds();
// . let's sleep so we have a chance to launch to another host in
// the same group in case this guy takes too long
// . don't re-register if we already did
if(!m_registeredSleep) {
int wait = calculateTimeout();
if(wait>=0) {
// . otherwise register for sleep callback to try again
// . sleepCallback1Wrapper() will call sendToHostLoop() for us
g_loop.registerSleepCallback(wait, this, sleepCallback1Wrapper, "Multicast::sleepCallback1Wrapper", m_niceness);
m_registeredSleep = true;
}
}
// successful launch
return true;
}
int Multicast::calculateTimeout() {
// . don't relaunch any niceness 1 stuff for a while
// . it often gets suspended due to query traffic
//if ( m_niceness > 0 && elapsed < 800000 ) return;
if ( m_niceness > 0 ) return -1;
// . Msg36 is used to get the length of an IndexList (termFreq)
// and is very fast, all in memory, don't wait more than 50ms
// . if we do re-route this is sucks cuz we'll get slightly different
// termFreqs which impact the total results count as well as summary
// generation since it's based on termFreq, not too mention the
// biggest impact being ordering of search results since the
// score weight is based on termFreq as well
// . but unfortunately, this scheme doesn't increase the ping time
// of dead hosts that much!!
// . NOTE: 2/26/04: i put most everything to 8000 ms since rerouting
// too much on an already saturated network of drives just
// excacerbates the problem. this stuff was originally put here
// to reroute for when a host went down... let's keep it that way
switch ( m_msgType ) {
// msg to get a summary from a query (calls msg22)
// put to 5 seconds now since some hosts freezeup still it seems
// and i haven't seen a summary generation of 5 seconds
case msg_type_20:
return 5000;
// msg 0x20 calls this to get the title rec
case msg_type_22:
return 1000;
// . msg to get an index list over the net
// . this limit should really be based on length of the index list
case msg_type_0:
// this should just be for when a host goes down, not for
// performance reasons, cuz we do pretty good load balancing
// and when things get saturated, rerouting excacerbates it
return 8000;
// msg to get docIds from a query, may take a while
case msg_type_39: {
// how many docsids request? first 4 bytes of request.
int32_t docsWanted = 10;
int32_t nqterms = 0;
if ( m_msg ) {
docsWanted = *(int32_t *)(m_msg);
nqterms = *(int32_t *)(m_msg+4);
}
// never re-route if it has a rerank, those take forever
// . how many milliseconds of waiting before we re-route?
// . 100 ms per doc wanted, but if they all end up
// clustering then docsWanted is no indication of the
// actual number of titleRecs (or title keys) read
// . it may take a while to do dup removal on 1 million docs
int32_t wait = 5000 + 100 * docsWanted;
// those big UOR queries should not get re-routed all the time
if ( nqterms > 0 ) {
wait += 1000 * nqterms;
}
if ( wait < 8000 ) {
wait = 8000;
}
return wait;
}
// don't relaunch anything else unless over 8 secs
default:
return 8000;
}
}
// this is called every 50 ms so we have the chance to launch our request
// to a more responsive host
void Multicast::sleepCallback1Wrapper ( int bogusfd , void *state ) {
Multicast *that = static_cast<Multicast*>(state);
that->sleepCallback1();
}
void Multicast::sleepCallback1() {
// . if our last launch was less than X seconds ago, wait another tick
// . we often send out 2+ requests and end up getting one reply before
// the others and that results in us getting unwanted dgrams...
// . increasing this delay here results in fewer wasted requests but
// if a host goes down you don't want a user to wait too long
// . after a host goes down it's ping takes a few secs to decrease
// . if a host is shutdown properly it will broadcast a msg to
// all hosts using Hostdb::broadcast() informing them that it's
// going down so they know to stop sending to it and mark him as
// dead
int64_t now = gettimeofdayInMilliseconds();
// watch out for someone advancing the system clock
if ( m_lastLaunch > now ) m_lastLaunch = now;
// get elapsed time since we started the send
int32_t elapsed = now - m_lastLaunch;
//log("elapsed = %" PRId32" type=0x%02x",elapsed,m_msgType);
// TODO: if the host went dead on us, re-route
// find out which host timedout
Host *hd = NULL;
for(int i=0; i<m_numHosts && !hd; i++)
if(m_host[i].m_retired)
hd = m_host[i].m_hostPtr;
// 11/21/06: now we only reroute if the host we sent to is marked as
// dead unless it is a msg type that takes little reply generation time
if ( hd && ! g_hostdb.isDead(hd) ) {
return;
}
// . do a loop over all hosts in the group
// . if a whole group of twins is down this will loop forever here
// every Xms, based the sleepWrapper timer for the msgType
if ( g_conf.m_logDebugQuery ) {
for (int32_t i = 0 ; i < m_numHosts ; i++ ) {
if ( ! m_host[i].m_slot ) continue;
// transaction is not in progress ifm_host[i].m_errno is set
const char *ee = "";
if ( m_host[i].m_errno ) ee = mstrerror(m_host[i].m_errno);
char ipbuf[16];
logf(LOG_DEBUG, "multicast: sleepWrapper1: tried host %s:%d %s",
iptoa(m_host[i].m_slot->getIp(),ipbuf), (int32_t)m_host[i].m_slot->getPort(), ee);
}
}
// log msg that we are trying to re-route
//log("Multicast::sleepWrapper1: trying to re-route msgType=0x%02x "
// "to new host", m_msgType );
// . otherwise, launch another request if we can
// . returns true if we successfully sent to another host
// . returns false and sets g_errno if no hosts left or other error
if ( sendToHostLoop(0,-1) ) {
// log msg that we were successful
int32_t hid = -1;
if ( hd ) hid = hd->m_hostId;
log(LOG_WARN, "multicast: sleepWrapper1: rerouted msgType=0x%02x from host #%d to new host after waiting %d ms",
(int)m_msgType, hid,elapsed);
// . mark it in the stats for PageStats.cpp
// . this is timeout based rerouting
g_stats.m_reroutes[(int)m_msgType][m_niceness?1:0]++;
return;
}
// if we registered the sleep callback we must have launched a
// request to a host so let gotReplyWrapperM1() deal with closeUpShop()
// . let replyWrapper1 be called if we got one launched
// . it should then call closeUpShop()
//if ( m_numLaunched ) return;
// otherwise, no outstanding requests and we failed to send to another
// host, probably because :
// 1. Msg34 timed out on all hosts
// 2. there were no udp slots available (which is bad)
//log("Multicast:: re-route failed for msgType=%02x. abandoning.",
// m_msgType );
// . the next send failed to send to a host, so close up shop
// . this is probably because the Msg34s timed out and we could not
// find a next "best host" to send to because of that
//closeUpShop ( NULL );
// . we were not able to send to another host, maybe it was dead or
// there are no hosts left!
// . i guess keep sleeping until host comes back up or transaction
// is cancelled
//log("Multicast::sleepWrapper1: re-route of msgType=0x%02x failed",
// m_msgType);
}
void Multicast::gotReply1(void *state, UdpSlot *slot) {
Multicast *THIS = static_cast<Multicast*>(state);
THIS->gotReply1(slot);
}
// come here if we've got a reply from a host that's not part of a group send
void Multicast::gotReply1 ( UdpSlot *slot ) {
ScopedLock sl(m_mtx);
// don't ever let UdpServer free this send buf (it is m_msg)
slot->m_sendBufAlloc = NULL;
// remove the slot from m_slots so it doesn't get nuked in
// gotSlot(slot) routine above
int32_t i = 0;
// careful! we might have recycled a slot!!! start with top and go down
// because UdpServer might give us the same slot ptr on our 3rd try
// that we had on our first try!
for ( i = 0 ; i < m_numHosts ; i++ ) {
// skip if not in progress
if ( ! m_host[i].m_inProgress ) continue;
// slot must match
if ( m_host[i].m_slot == slot ) break;
}
// if it matched no slot that's wierd
if ( i >= m_numHosts ) {
log(LOG_LOGIC,"multicast: multicast: Not our slot 2.");
g_process.shutdownAbort(true);
}
// setm_host[i].m_errno, if any
if ( g_errno ) {
m_host[i].m_errno = g_errno;
}
// mark it as no longer in progress
m_host[i].m_inProgress = false;
Host *h = m_host[i].m_hostPtr;
// save the host we got a reply from
m_replyingHost = h;
m_replyLaunchTime = m_host[i].m_launchTime;
if ( m_sentToTwin ) {
logDebug(g_conf.m_logDebugMulticast, "multicast: Twin msgType=0x%" PRIx32" (this=%p) reply: %s.",
(int32_t) m_msgType, this, mstrerror(g_errno));
}
sl.unlock();
// on error try sending the request to another host
// return if we kicked another request off ok
if ( g_errno ) {
// do not log not found on an external network
// Only log those errors that are real errors as warning
if ( g_errno != ENOTFOUND &&
g_errno != ENOLINKTEXT_AREATAG ) {
// log the error
Host *h = g_hostdb.getUdpHost(slot->getIp(), slot->getPort());
if (h) {
log(LOG_WARN, "multicast: Got error in reply from hostId %d (msgType=0x%02x transId=%d nice=%d): %s.",
h->m_hostId, (int)slot->getMsgType(), slot->getTransId(),
m_niceness,
mstrerror(g_errno));
} else {
char ipbuf[16];
log(LOG_WARN, "multicast: Got error in reply from %s:%d (msgType=0x%02x transId=%d nice =%d): %s.",
iptoa(slot->getIp(),ipbuf), (int32_t) slot->getPort(),
(int)slot->getMsgType(), slot->getTransId(), m_niceness,
mstrerror(g_errno));
}
}
// . try to send to another host
// . on successful sending return, we'll be called on reply
// . this also returns false if no new hosts left to send to
// . only try another host if g_errno is NOT ENOTFOUND cuz
// we have quite a few missing clustRecs and titleRecs
// and doing a second lookup will decrease query response
// . if the Msg22 lookup cannot find the titleRec for indexing
// purposes, it should check any twin hosts because this
// is very important... if this is for query time, however,
// then accept the ENOTFOUND without spawning another request
// . but if the record is really not there we waste seeks!
// . EBADENGINEER is now used by titledb's Msg22 when a docid
// is in tfndb but not in titledb (or id2 is invalid)
// . it is more important that we serve the title rec than
// the performance gain. if you want the performance gain
// then you should repair your index to avoid this. therefore
// send to twin on ENOTFOUND
// . often, though, we are restring to indexdb root so after
// doing a lot of deletes there will be a lot of not founds
// that are really not found (not corruption) so don't do it
// anymore
// . let's go for accuracy even for queries
// . until i fix the bug of losing titlerecs for some reason
// probably during merges now, we reroute on ENOTFOUND.
bool sendToTwin = true;
if ( g_errno == EBADENGINEER ) sendToTwin = false;
if ( g_errno == EMSGTOOBIG ) sendToTwin = false;
if ( g_errno == E2BIG ) sendToTwin = false;
if ( g_errno == EUNCOMPRESSERROR ) sendToTwin = false;
if ( g_errno == ENOLINKTEXT_AREATAG) sendToTwin = false;
// ok, let's give up on ENOTFOUND, because the vast majority
// of time it seems it is really not on the twin either...
if ( g_errno == ENOTFOUND ) sendToTwin = false;
// do not send to twin if we are out of time
time_t now = getTime();
int32_t timeRemaining = m_startTime + m_totalTimeout - now;
if ( timeRemaining <= 0 ) sendToTwin = false;
// send to the twin
if ( sendToTwin && sendToHostLoop(0,-1) ) {
log(LOG_INFO, "multicast: Trying to send request msgType=0x%" PRIx32" to a twin. (this=%p)",
(int32_t)m_msgType,this);
m_sentToTwin = true;
// . keep stats
// . this is error based rerouting
// . this can be timeouts as well, if the
// receiver sent a request itself and that
// timed out...
g_stats.m_reroutes[(int)m_msgType][m_niceness?1:0]++;
return;
}
// . otherwise we've failed on all hosts
// . re-instate g_errno,might have been set by sendToHostLoop()
g_errno =m_host[i].m_errno;
}
closeUpShop ( slot );
}
void Multicast::closeUpShop ( UdpSlot *slot ) {
// sanity check
if (!m_inUse) {
g_process.shutdownAbort(true);
}
// destroy the OTHER slots we've spawned that are in progress
destroySlotsInProgress ( slot );
if (slot) {
// . now we have a good reply... but not if g_errno is set
// . save the reply of this slot here
// . this is bad if we got an g_errno above, it will set the slot's
// readBuf to NULL up there, and that will make m_readBuf NULL here
// causing a mem leak. i fixed by adding an mfree on m_replyBuf
// in Multicast::reset() routine.
// . i fixed again by ensuring we do not set m_ownReadBuf to false
// in getBestReply() below if m_readBuf is NULL
m_readBuf = slot->m_readBuf;
m_readBufSize = slot->m_readBufSize;
m_readBufMaxSize = slot->m_readBufMaxSize;
// . if the slot had an error, propagate it so it will be set when
// we call the callback.
if (!g_errno) {
g_errno = slot->getErrno();
}
// don't let UdpServer free the readBuf now that we point to it
slot->m_readBuf = NULL;
slot->m_readBufSize = 0;
slot->m_readBufMaxSize = 0;
// save slot so msg4 knows what slot replied in udpserver
// for doing its flush callback logic
m_slot = slot;
}
// unregister our sleep wrapper if we did
if ( m_registeredSleep ) {
g_loop.unregisterSleepCallback(this, sleepCallback1Wrapper);
m_registeredSleep = false;
}
// allow us to be re-used now, callback might relaunch
m_inUse = false;
// now call the user callback if it exists
if ( m_callback ) {
m_callback ( m_state , m_state2 );
}
}
// destroy all slots that may be in progress (except "slot")
void Multicast::destroySlotsInProgress ( UdpSlot *slot ) {
// do a loop over all hosts in the group
for (int32_t i = 0 ; i < m_numHosts ; i++ ) {
// . destroy all slots but this one that are in progress
// . we'll be destroyed when we return from the cback
if ( ! m_host[i].m_slot ) continue;
// transaction is not in progress ifm_host[i].m_errno is set
if ( m_host[i].m_errno ) continue;
// dont' destroy us, it'll happen when we return
if ( m_host[i].m_slot == slot ) continue;
// must be in progress
if ( ! m_host[i].m_inProgress ) continue;
// don't free his sendBuf, readBuf is ok to free, however
m_host[i].m_slot->m_sendBufAlloc = NULL;
// destroy this slot that's in progress
g_udpServer.destroySlot ( m_host[i].m_slot );
// do not re-destroy. consider no longer in progress.
m_host[i].m_inProgress = false;
}
}
// we set *freeReply to true if you'll need to free it
char *Multicast::getBestReply(int32_t *replySize, int32_t *replyMaxSize, bool *freeReply, bool steal) {
*replySize = m_readBufSize;
*replyMaxSize = m_readBufMaxSize;
if(steal) {
m_freeReadBuf = false;
}
*freeReply = m_freeReadBuf;
// this can be NULL if we destroyed the slot in progress only to
// try another host who was dead!
if ( m_readBuf ) m_ownReadBuf = false;
return m_readBuf;
}