[freeradius.git] / src / main / threads.c

/*
 * threads.c    request threading support
 *
 * Version:     $Id$
 *
 *   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 2 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, write to the Free Software
 *   Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
 *
 * Copyright 2000,2006  The FreeRADIUS server project
 * Copyright 2000  Alan DeKok <aland@ox.org>
 */

#include <freeradius-devel/ident.h>
RCSID("$Id$")

#include <freeradius-devel/radiusd.h>
#include <freeradius-devel/rad_assert.h>

/*
 *      Other OS's have sem_init, OS X doesn't.
 */
#ifdef HAVE_SEMAPHORE_H
#include <semaphore.h>
#endif

#ifdef DARWIN
#include <mach/task.h>
#include <mach/semaphore.h>

#undef sem_t
#define sem_t semaphore_t
#undef sem_init
#define sem_init(s,p,c) semaphore_create(mach_task_self(),s,SYNC_POLICY_FIFO,c)
#undef sem_wait
#define sem_wait(s) semaphore_wait(*s)
#undef sem_post
#define sem_post(s) semaphore_signal(*s)
#endif

#ifdef HAVE_SYS_WAIT_H
#include <sys/wait.h>
#endif

#ifdef HAVE_PTHREAD_H

#ifdef HAVE_OPENSSL_CRYPTO_H
#include <openssl/crypto.h>
#endif
#ifdef HAVE_OPENSSL_ERR_H
#include <openssl/err.h>
#endif
#ifdef HAVE_OPENSSL_EVP_H
#include <openssl/evp.h>
#endif

#define SEMAPHORE_LOCKED        (0)
#define SEMAPHORE_UNLOCKED      (1)

#define THREAD_RUNNING          (1)
#define THREAD_CANCELLED        (2)
#define THREAD_EXITED           (3)

#define NUM_FIFOS               RAD_LISTEN_MAX


/*
 *  A data structure which contains the information about
 *  the current thread.
 *
 *  pthread_id     pthread id
 *  thread_num     server thread number, 1...number of threads
 *  semaphore     used to block the thread until a request comes in
 *  status        is the thread running or exited?
 *  request_count the number of requests that this thread has handled
 *  timestamp     when the thread started executing.
 */
typedef struct THREAD_HANDLE {
        struct THREAD_HANDLE *prev;
        struct THREAD_HANDLE *next;
        pthread_t            pthread_id;
        int                  thread_num;
        int                  status;
        unsigned int         request_count;
        time_t               timestamp;
        REQUEST              *request;
} THREAD_HANDLE;

/*
 *      For the request queue.
 */
typedef struct request_queue_t {
        REQUEST           *request;
        RAD_REQUEST_FUNP  fun;
} request_queue_t;

typedef struct thread_fork_t {
        pid_t           pid;
        int             status;
        int             exited;
} thread_fork_t;


/*
 *      A data structure to manage the thread pool.  There's no real
 *      need for a data structure, but it makes things conceptually
 *      easier.
 */
typedef struct THREAD_POOL {
        THREAD_HANDLE *head;
        THREAD_HANDLE *tail;

        int total_threads;
        int active_threads;     /* protected by queue_mutex */
        int max_thread_num;
        int start_threads;
        int max_threads;
        int min_spare_threads;
        int max_spare_threads;
        unsigned int max_requests_per_thread;
        unsigned long request_count;
        time_t time_last_spawned;
        int cleanup_delay;
        int spawn_flag;

#ifdef WNOHANG
        pthread_mutex_t wait_mutex;
        fr_hash_table_t *waiters;
#endif

        /*
         *      All threads wait on this semaphore, for requests
         *      to enter the queue.
         */
        sem_t           semaphore;

        /*
         *      To ensure only one thread at a time touches the queue.
         */
        pthread_mutex_t queue_mutex;

        int             max_queue_size;
        int             num_queued;
        fr_fifo_t       *fifo[NUM_FIFOS];
} THREAD_POOL;

static THREAD_POOL thread_pool;
static int pool_initialized = FALSE;
static time_t last_cleaned = 0;

static void thread_pool_manage(time_t now);

/*
 *      A mapping of configuration file names to internal integers
 */
static const CONF_PARSER thread_config[] = {
        { "start_servers",           PW_TYPE_INTEGER, 0, &thread_pool.start_threads,           "5" },
        { "max_servers",             PW_TYPE_INTEGER, 0, &thread_pool.max_threads,             "32" },
        { "min_spare_servers",       PW_TYPE_INTEGER, 0, &thread_pool.min_spare_threads,       "3" },
        { "max_spare_servers",       PW_TYPE_INTEGER, 0, &thread_pool.max_spare_threads,       "10" },
        { "max_requests_per_server", PW_TYPE_INTEGER, 0, &thread_pool.max_requests_per_thread, "0" },
        { "cleanup_delay",           PW_TYPE_INTEGER, 0, &thread_pool.cleanup_delay,           "5" },
        { "max_queue_size",          PW_TYPE_INTEGER, 0, &thread_pool.max_queue_size,           "65536" },
        { NULL, -1, 0, NULL, NULL }
};


#ifdef HAVE_OPENSSL_CRYPTO_H

/*
 *      If we're linking against OpenSSL, then it is the
 *      duty of the application, if it is multithreaded,
 *      to provide OpenSSL with appropriate thread id
 *      and mutex locking functions
 *
 *      Note: this only implements static callbacks.
 *      OpenSSL does not use dynamic locking callbacks
 *      right now, but may in the futiure, so we will have
 *      to add them at some point.
 */

static pthread_mutex_t *ssl_mutexes = NULL;

static unsigned long ssl_id_function(void)
{
        return (unsigned long) pthread_self();
}

static void ssl_locking_function(int mode, int n, const char *file, int line)
{
        file = file;            /* -Wunused */
        line = line;            /* -Wunused */

        if (mode & CRYPTO_LOCK) {
                pthread_mutex_lock(&(ssl_mutexes[n]));
        } else {
                pthread_mutex_unlock(&(ssl_mutexes[n]));
        }
}

static int setup_ssl_mutexes(void)
{
        int i;

#ifdef HAVE_OPENSSL_EVP_H
        /*
         *      Enable all ciphers and digests.
         */
        OpenSSL_add_all_algorithms();
#endif

        ssl_mutexes = rad_malloc(CRYPTO_num_locks() * sizeof(pthread_mutex_t));
        if (!ssl_mutexes) {
                radlog(L_ERR, "Error allocating memory for SSL mutexes!");
                return 0;
        }

        for (i = 0; i < CRYPTO_num_locks(); i++) {
                pthread_mutex_init(&(ssl_mutexes[i]), NULL);
        }

        CRYPTO_set_id_callback(ssl_id_function);
        CRYPTO_set_locking_callback(ssl_locking_function);

        return 1;
}
#endif

#ifdef WNOHANG
/*
 *      We don't want to catch SIGCHLD for a host of reasons.
 *
 *      - exec_wait means that someone, somewhere, somewhen, will
 *      call waitpid(), and catch the child.
 *
 *      - SIGCHLD is delivered to a random thread, not the one that
 *      forked.
 *
 *      - if another thread catches the child, we have to coordinate
 *      with the thread doing the waiting.
 *
 *      - if we don't waitpid() for non-wait children, they'll be zombies,
 *      and will hang around forever.
 *
 */
static void reap_children(void)
{
        pid_t pid;
        int status;
        thread_fork_t mytf, *tf;


        pthread_mutex_lock(&thread_pool.wait_mutex);

        do {
                pid = waitpid(0, &status, WNOHANG);
                if (pid <= 0) break;

                mytf.pid = pid;
                tf = fr_hash_table_finddata(thread_pool.waiters, &mytf);
                if (!tf) continue;

                tf->status = status;
                tf->exited = 1;
        } while (fr_hash_table_num_elements(thread_pool.waiters) > 0);

        pthread_mutex_unlock(&thread_pool.wait_mutex);
}
#else
#define reap_children()
#endif /* WNOHANG */

/*
 *      Add a request to the list of waiting requests.
 *      This function gets called ONLY from the main handler thread...
 *
 *      This function should never fail.
 */
static int request_enqueue(REQUEST *request, RAD_REQUEST_FUNP fun)
{
        request_queue_t *entry;

        pthread_mutex_lock(&thread_pool.queue_mutex);

        thread_pool.request_count++;

        if (thread_pool.num_queued >= thread_pool.max_queue_size) {
                pthread_mutex_unlock(&thread_pool.queue_mutex);

                /*
                 *      Mark the request as done.
                 */
                radlog(L_ERR, "!!! ERROR !!! The server is blocked: discarding new request %d", request->number);
                request->child_state = REQUEST_DONE;
                return 0;
        }

        entry = rad_malloc(sizeof(*entry));
        entry->request = request;
        entry->fun = fun;

        /*
         *      Push the request onto the appropriate fifo for that
         */
        if (!fr_fifo_push(thread_pool.fifo[request->priority],
                            entry)) {
                pthread_mutex_unlock(&thread_pool.queue_mutex);
                radlog(L_ERR, "!!! ERROR !!! Failed inserting request %d into the queue", request->number);
                request->child_state = REQUEST_DONE;
                return 0;
        }

        thread_pool.num_queued++;

        pthread_mutex_unlock(&thread_pool.queue_mutex);

        /*
         *      There's one more request in the queue.
         *
         *      Note that we're not touching the queue any more, so
         *      the semaphore post is outside of the mutex.  This also
         *      means that when the thread wakes up and tries to lock
         *      the mutex, it will be unlocked, and there won't be
         *      contention.
         */
        sem_post(&thread_pool.semaphore);

        return 1;
}

/*
 *      Remove a request from the queue.
 */
static int request_dequeue(REQUEST **request, RAD_REQUEST_FUNP *fun)
{
        RAD_LISTEN_TYPE i, start;
        request_queue_t *entry;

        reap_children();

        pthread_mutex_lock(&thread_pool.queue_mutex);

        /*
         *      Clear old requests from all queues.
         *
         *      We only do one pass over the queue, in order to
         *      amortize the work across the child threads.  Since we
         *      do N checks for one request de-queued, the old
         *      requests will be quickly cleared.
         */
        for (i = 0; i < RAD_LISTEN_MAX; i++) {
                entry = fr_fifo_peek(thread_pool.fifo[i]);
                if (!entry ||
                    (entry->request->master_state != REQUEST_STOP_PROCESSING)) {
                        continue;
}
                /*
                 *      This entry was marked to be stopped.  Acknowledge it.
                 */
                entry = fr_fifo_pop(thread_pool.fifo[i]);
                rad_assert(entry != NULL);
                entry->request->child_state = REQUEST_DONE;
                thread_pool.num_queued--;
                free(entry);
                entry = NULL;
        }

        start = 0;
 retry:
        /*
         *      Pop results from the top of the queue
         */
        for (i = start; i < RAD_LISTEN_MAX; i++) {
                entry = fr_fifo_pop(thread_pool.fifo[i]);
                if (entry) {
                        start = i;
                        break;
                }
        }

        if (!entry) {
                pthread_mutex_unlock(&thread_pool.queue_mutex);
                *request = NULL;
                *fun = NULL;
                return 0;
        }

        rad_assert(thread_pool.num_queued > 0);
        thread_pool.num_queued--;
        *request = entry->request;
        *fun = entry->fun;
        free(entry);
        entry = NULL;

        rad_assert(*request != NULL);
        rad_assert((*request)->magic == REQUEST_MAGIC);
        rad_assert(*fun != NULL);

        /*
         *      If the request has sat in the queue for too long,
         *      kill it.
         *
         *      The main clean-up code can't delete the request from
         *      the queue, and therefore won't clean it up until we
         *      have acknowledged it as "done".
         */
        if ((*request)->master_state == REQUEST_STOP_PROCESSING) {
                (*request)->child_state = REQUEST_DONE;
                goto retry;
        }

        /*
         *      The thread is currently processing a request.
         */
        thread_pool.active_threads++;

        pthread_mutex_unlock(&thread_pool.queue_mutex);

        return 1;
}


/*
 *      The main thread handler for requests.
 *
 *      Wait on the semaphore until we have it, and process the request.
 */
static void *request_handler_thread(void *arg)
{
        RAD_REQUEST_FUNP  fun;
        THREAD_HANDLE     *self = (THREAD_HANDLE *) arg;

        /*
         *      Loop forever, until told to exit.
         */
        do {
                /*
                 *      Wait to be signalled.
                 */
                DEBUG2("Thread %d waiting to be assigned a request",
                       self->thread_num);
        re_wait:
                if (sem_wait(&thread_pool.semaphore) != 0) {
                        /*
                         *      Interrupted system call.  Go back to
                         *      waiting, but DON'T print out any more
                         *      text.
                         */
                        if (errno == EINTR) {
                                DEBUG2("Re-wait %d", self->thread_num);
                                goto re_wait;
                        }
                        radlog(L_ERR, "Thread %d failed waiting for semaphore: %s: Exiting\n",
                               self->thread_num, strerror(errno));
                        break;
                }

                DEBUG2("Thread %d got semaphore", self->thread_num);

#ifdef HAVE_OPENSSL_ERR_H
                /*
                 *      Clear the error queue for the current thread.
                 */
                ERR_clear_error ();
#endif

                /*
                 *      Try to grab a request from the queue.
                 *
                 *      It may be empty, in which case we fail
                 *      gracefully.
                 */
                if (!request_dequeue(&self->request, &fun)) continue;

                self->request->child_pid = self->pthread_id;
                self->request_count++;

                DEBUG2("Thread %d handling request %d, (%d handled so far)",
                       self->thread_num, self->request->number,
                       self->request_count);

                radius_handle_request(self->request, fun);

                /*
                 *      Update the active threads.
                 */
                pthread_mutex_lock(&thread_pool.queue_mutex);
                rad_assert(thread_pool.active_threads > 0);
                thread_pool.active_threads--;
                pthread_mutex_unlock(&thread_pool.queue_mutex);
        } while (self->status != THREAD_CANCELLED);

        DEBUG2("Thread %d exiting...", self->thread_num);

#ifdef HAVE_OPENSSL_ERR_H
        /*
         *      If we linked with OpenSSL, the application
         *      must remove the thread's error queue before
         *      exiting to prevent memory leaks.
         */
        ERR_remove_state(0);
#endif

        /*
         *  Do this as the LAST thing before exiting.
         */
        self->request = NULL;
        self->status = THREAD_EXITED;

        return NULL;
}

/*
 *      Take a THREAD_HANDLE, delete it from the thread pool and
 *      free its resources.
 *
 *      This function is called ONLY from the main server thread,
 *      ONLY after the thread has exited.
 */
static void delete_thread(THREAD_HANDLE *handle)
{
        THREAD_HANDLE *prev;
        THREAD_HANDLE *next;

        rad_assert(handle->request == NULL);

        DEBUG2("Deleting thread %d", handle->thread_num);

        prev = handle->prev;
        next = handle->next;
        rad_assert(thread_pool.total_threads > 0);
        thread_pool.total_threads--;

        /*
         *      Remove the handle from the list.
         */
        if (prev == NULL) {
                rad_assert(thread_pool.head == handle);
                thread_pool.head = next;
        } else {
                prev->next = next;
        }

        if (next == NULL) {
                rad_assert(thread_pool.tail == handle);
                thread_pool.tail = prev;
        } else {
                next->prev = prev;
        }

        /*
         *      Free the handle, now that it's no longer referencable.
         */
        free(handle);
}


/*
 *      Spawn a new thread, and place it in the thread pool.
 *
 *      The thread is started initially in the blocked state, waiting
 *      for the semaphore.
 */
static THREAD_HANDLE *spawn_thread(time_t now)
{
        int rcode;
        THREAD_HANDLE *handle;
        pthread_attr_t attr;

        /*
         *      Ensure that we don't spawn too many threads.
         */
        if (thread_pool.total_threads >= thread_pool.max_threads) {
                DEBUG2("Thread spawn failed.  Maximum number of threads (%d) already running.", thread_pool.max_threads);
                return NULL;
        }

        /*
         *      Allocate a new thread handle.
         */
        handle = (THREAD_HANDLE *) rad_malloc(sizeof(THREAD_HANDLE));
        memset(handle, 0, sizeof(THREAD_HANDLE));
        handle->prev = NULL;
        handle->next = NULL;
        handle->thread_num = thread_pool.max_thread_num++;
        handle->request_count = 0;
        handle->status = THREAD_RUNNING;
        handle->timestamp = time(NULL);

        /*
         *      Initialize the thread's attributes to detached.
         *
         *      We could call pthread_detach() later, but if the thread
         *      exits between the create & detach calls, it will need to
         *      be joined, which will never happen.
         */
        pthread_attr_init(&attr);
        pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_DETACHED);

        /*
         *      Create the thread detached, so that it cleans up it's
         *      own memory when it exits.
         *
         *      Note that the function returns non-zero on error, NOT
         *      -1.  The return code is the error, and errno isn't set.
         */
        rcode = pthread_create(&handle->pthread_id, &attr,
                        request_handler_thread, handle);
        if (rcode != 0) {
                radlog(L_ERR, "Thread create failed: %s",
                       strerror(rcode));
                return NULL;
        }
        pthread_attr_destroy(&attr);

        /*
         *      One more thread to go into the list.
         */
        thread_pool.total_threads++;
        DEBUG2("Thread spawned new child %d. Total threads in pool: %d",
                        handle->thread_num, thread_pool.total_threads);

        /*
         *      Add the thread handle to the tail of the thread pool list.
         */
        if (thread_pool.tail) {
                thread_pool.tail->next = handle;
                handle->prev = thread_pool.tail;
                thread_pool.tail = handle;
        } else {
                rad_assert(thread_pool.head == NULL);
                thread_pool.head = thread_pool.tail = handle;
        }

        /*
         *      Update the time we last spawned a thread.
         */
        thread_pool.time_last_spawned = now;

        /*
         *      And return the new handle to the caller.
         */
        return handle;
}

/*
 *      Temporary function to prevent server from executing a SIGHUP
 *      until all threads are finished handling requests.  This returns
 *      the number of active threads to 'radiusd.c'.
 */
int total_active_threads(void)
{
        /*
         *      We don't acquire the mutex, so this is just an estimate.
         *      We can't return with the lock held, so there's no point
         *      in getting the guaranteed correct value; by the time
         *      the caller sees it, it can be wrong again.
         */
        return thread_pool.active_threads;
}


#ifdef WNOHANG
static uint32_t pid_hash(const void *data)
{
        const thread_fork_t *tf = data;

        return fr_hash(&tf->pid, sizeof(tf->pid));
}

static int pid_cmp(const void *one, const void *two)
{
        const thread_fork_t *a = one;
        const thread_fork_t *b = two;

        return (a->pid - b->pid);
}
#endif

/*
 *      Allocate the thread pool, and seed it with an initial number
 *      of threads.
 *
 *      FIXME: What to do on a SIGHUP???
 */
int thread_pool_init(CONF_SECTION *cs, int spawn_flag)
{
        int             i, rcode;
        CONF_SECTION    *pool_cf;
        time_t          now;

        now = time(NULL);

        /*
         *      We're not spawning new threads, don't do
         *      anything.
         */
        if (!spawn_flag) return 0;

        /*
         *      After a SIGHUP, we don't over-write the previous values.
         */
        if (!pool_initialized) {
                /*
                 *      Initialize the thread pool to some reasonable values.
                 */
                memset(&thread_pool, 0, sizeof(THREAD_POOL));
                thread_pool.head = NULL;
                thread_pool.tail = NULL;
                thread_pool.total_threads = 0;
                thread_pool.max_thread_num = 1;
                thread_pool.cleanup_delay = 5;
                thread_pool.spawn_flag = spawn_flag;

#ifdef WNOHANG
                if ((pthread_mutex_init(&thread_pool.wait_mutex,NULL) != 0)) {
                        radlog(L_ERR, "FATAL: Failed to initialize wait mutex: %s",
                               strerror(errno));
                        return -1;
                }

                /*
                 *      Create the hash table of child PID's
                 */
                thread_pool.waiters = fr_hash_table_create(pid_hash,
                                                             pid_cmp,
                                                             free);
                if (!thread_pool.waiters) {
                        radlog(L_ERR, "FATAL: Failed to set up wait hash");
                        return -1;
                }
#endif
        }

        pool_cf = cf_subsection_find_next(cs, NULL, "thread");
        if (!pool_cf) {
                radlog(L_ERR, "FATAL: Attempting to start in multi-threaded mode with no thread configuration in radiusd.conf");
                return -1;
        }

        if (cf_section_parse(pool_cf, NULL, thread_config) < 0) {
                return -1;
        }

        /*
         *      Catch corner cases.
         */
        if (thread_pool.min_spare_threads < 1)
                thread_pool.min_spare_threads = 1;
        if (thread_pool.max_spare_threads < 1)
                thread_pool.max_spare_threads = 1;
        if (thread_pool.max_spare_threads < thread_pool.min_spare_threads)
                thread_pool.max_spare_threads = thread_pool.min_spare_threads;

        /*
         *      The pool has already been initialized.  Don't spawn
         *      new threads, and don't forget about forked children,
         */
        if (pool_initialized) {
                return 0;
        }

        /*
         *      Initialize the queue of requests.
         */
        memset(&thread_pool.semaphore, 0, sizeof(thread_pool.semaphore));
        rcode = sem_init(&thread_pool.semaphore, 0, SEMAPHORE_LOCKED);
        if (rcode != 0) {
                radlog(L_ERR, "FATAL: Failed to initialize semaphore: %s",
                       strerror(errno));
                return -1;
        }

        rcode = pthread_mutex_init(&thread_pool.queue_mutex,NULL);
        if (rcode != 0) {
                radlog(L_ERR, "FATAL: Failed to initialize queue mutex: %s",
                       strerror(errno));
                return -1;
        }

        /*
         *      Allocate multiple fifos.
         */
        for (i = 0; i < RAD_LISTEN_MAX; i++) {
                thread_pool.fifo[i] = fr_fifo_create(65536, NULL);
                if (!thread_pool.fifo[i]) {
                        radlog(L_ERR, "FATAL: Failed to set up request fifo");
                        return -1;
                }
        }

#ifdef HAVE_OPENSSL_CRYPTO_H
        /*
         *      If we're linking with OpenSSL too, then we need
         *      to set up the mutexes and enable the thread callbacks.
         */
        if (!setup_ssl_mutexes()) {
                radlog(L_ERR, "FATAL: Failed to set up SSL mutexes");
                return -1;
        }
#endif


        /*
         *      Create a number of waiting threads.
         *
         *      If we fail while creating them, do something intelligent.
         */
        for (i = 0; i < thread_pool.start_threads; i++) {
                if (spawn_thread(now) == NULL) {
                        return -1;
                }
        }

        DEBUG2("Thread pool initialized");
        pool_initialized = TRUE;
        return 0;
}


/*
 *      Assign a new request to a free thread.
 *
 *      If there isn't a free thread, then try to create a new one,
 *      up to the configured limits.
 */
int thread_pool_addrequest(REQUEST *request, RAD_REQUEST_FUNP fun)
{
        time_t now = request->timestamp;

        /*
         *      We've been told not to spawn threads, so don't.
         */
        if (!thread_pool.spawn_flag) {
                radius_handle_request(request, fun);

#ifdef WNOHANG
                /*
                 *      Requests that care about child process exit
                 *      codes have already either called
                 *      rad_waitpid(), or they've given up.
                 */
                wait(NULL);
#endif
                return 1;
        }

        /*
         *      Add the new request to the queue.
         */
        if (!request_enqueue(request, fun)) return 0;

        /*
         *      If we haven't checked the number of child threads
         *      in a while, OR if the thread pool appears to be full,
         *      go manage it.
         */
        if ((last_cleaned < now) ||
            (thread_pool.active_threads == thread_pool.total_threads)) {
                thread_pool_manage(now);
        }

        return 1;
}

/*
 *      Check the min_spare_threads and max_spare_threads.
 *
 *      If there are too many or too few threads waiting, then we
 *      either create some more, or delete some.
 */
static void thread_pool_manage(time_t now)
{
        int spare;
        int i, total;
        THREAD_HANDLE *handle, *next;
        int active_threads;

        /*
         *      We don't need a mutex lock here, as we're reading
         *      active_threads, and not modifying it.  We want a close
         *      approximation of the number of active threads, and this
         *      is good enough.
         */
        active_threads = thread_pool.active_threads;
        spare = thread_pool.total_threads - active_threads;
        if (debug_flag) {
                static int old_total = -1;
                static int old_active = -1;

                if ((old_total != thread_pool.total_threads) ||
                                (old_active != active_threads)) {
                        DEBUG2("Threads: total/active/spare threads = %d/%d/%d",
                                        thread_pool.total_threads, active_threads, spare);
                        old_total = thread_pool.total_threads;
                        old_active = active_threads;
                }
        }

        /*
         *      If there are too few spare threads.  Go create some more.
         */
        if (spare < thread_pool.min_spare_threads) {
                total = thread_pool.min_spare_threads - spare;

                DEBUG2("Threads: Spawning %d spares", total);

                /*
                 *      Create a number of spare threads.
                 */
                for (i = 0; i < total; i++) {
                        handle = spawn_thread(now);
                        if (handle == NULL) {
                                return;
                        }
                }

                return;         /* there aren't too many spare threads */
        }

        /*
         *      Only delete spare threads if we haven't already done
         *      so this second.
         */
        if (now == last_cleaned) {
                return;
        }
        last_cleaned = now;

        /*
         *      Loop over the thread pool, deleting exited threads.
         */
        for (handle = thread_pool.head; handle; handle = next) {
                next = handle->next;

                /*
                 *      Maybe we've asked the thread to exit, and it
                 *      has agreed.
                 */
                if (handle->status == THREAD_EXITED) {
                        delete_thread(handle);
                }
        }

        /*
         *      Only delete the spare threads if sufficient time has
         *      passed since we last created one.  This helps to minimize
         *      the amount of create/delete cycles.
         */
        if ((now - thread_pool.time_last_spawned) < thread_pool.cleanup_delay) {
                return;
        }

        /*
         *      If there are too many spare threads, delete one.
         *
         *      Note that we only delete ONE at a time, instead of
         *      wiping out many.  This allows the excess servers to
         *      be slowly reaped, just in case the load spike comes again.
         */
        if (spare > thread_pool.max_spare_threads) {

                spare -= thread_pool.max_spare_threads;

                DEBUG2("Threads: deleting 1 spare out of %d spares", spare);

                /*
                 *      Walk through the thread pool, deleting the
                 *      first idle thread we come across.
                 */
                for (handle = thread_pool.head; (handle != NULL) && (spare > 0) ; handle = next) {
                        next = handle->next;

                        /*
                         *      If the thread is not handling a
                         *      request, but still live, then tell it
                         *      to exit.
                         *
                         *      It will eventually wake up, and realize
                         *      it's been told to commit suicide.
                         */
                        if ((handle->request == NULL) &&
                            (handle->status == THREAD_RUNNING)) {
                                handle->status = THREAD_CANCELLED;
                                /*
                                 *      Post an extra semaphore, as a
                                 *      signal to wake up, and exit.
                                 */
                                sem_post(&thread_pool.semaphore);
                                spare--;
                                break;
                        }
                }
        }

        /*
         *      If the thread has handled too many requests, then make it
         *      exit.
         */
        if (thread_pool.max_requests_per_thread > 0) {
                for (handle = thread_pool.head; handle; handle = next) {
                        next = handle->next;

                        /*
                         *      Not handling a request, but otherwise
                         *      live, we can kill it.
                         */
                        if ((handle->request == NULL) &&
                            (handle->status == THREAD_RUNNING) &&
                            (handle->request_count > thread_pool.max_requests_per_thread)) {
                                handle->status = THREAD_CANCELLED;
                                sem_post(&thread_pool.semaphore);
                        }
                }
        }

        /*
         *      Otherwise everything's kosher.  There are not too few,
         *      or too many spare threads.  Exit happily.
         */
        return;
}


#ifdef WNOHANG
/*
 *      Thread wrapper for fork().
 */
pid_t rad_fork(void)
{
        pid_t child_pid;

        if (!pool_initialized) return fork();

        reap_children();        /* be nice to non-wait thingies */

        if (fr_hash_table_num_elements(thread_pool.waiters) >= 1024) {
                return -1;
        }

        /*
         *      Fork & save the PID for later reaping.
         */
        child_pid = fork();
        if (child_pid > 0) {
                int rcode;
                thread_fork_t *tf;

                tf = rad_malloc(sizeof(*tf));
                memset(tf, 0, sizeof(*tf));

                tf->pid = child_pid;

                pthread_mutex_lock(&thread_pool.wait_mutex);
                rcode = fr_hash_table_insert(thread_pool.waiters, tf);
                pthread_mutex_unlock(&thread_pool.wait_mutex);

                if (!rcode) {
                        radlog(L_ERR, "Failed to store PID, creating what will be a zombie process %d",
                               (int) child_pid);
                }
        }

        /*
         *      Return whatever we were told.
         */
        return child_pid;
}


/*
 *      Wait 10 seconds at most for a child to exit, then give up.
 */
pid_t rad_waitpid(pid_t pid, int *status)
{
        int i;
        thread_fork_t mytf, *tf;

        if (!pool_initialized) return waitpid(pid, status, 0);

        if (pid <= 0) return -1;

        mytf.pid = pid;

        pthread_mutex_lock(&thread_pool.wait_mutex);
        tf = fr_hash_table_finddata(thread_pool.waiters, &mytf);
        pthread_mutex_unlock(&thread_pool.wait_mutex);

        if (!tf) return -1;

        for (i = 0; i < 100; i++) {
                reap_children();

                if (tf->exited) {
                        *status = tf->status;

                        pthread_mutex_lock(&thread_pool.wait_mutex);
                        fr_hash_table_delete(thread_pool.waiters, &mytf);
                        pthread_mutex_unlock(&thread_pool.wait_mutex);
                        return pid;
                }
                usleep(100000); /* sleep for 1/10 of a second */
        }

        /*
         *      10 seconds have passed, give up on the child.
         */
        pthread_mutex_lock(&thread_pool.wait_mutex);
        fr_hash_table_delete(thread_pool.waiters, &mytf);
        pthread_mutex_unlock(&thread_pool.wait_mutex);

        return 0;
}
#else
/*
 *      No rad_fork or rad_waitpid
 */
#endif

void thread_pool_lock(void)
{
        pthread_mutex_lock(&thread_pool.queue_mutex);
}

void thread_pool_unlock(void)
{
        pthread_mutex_unlock(&thread_pool.queue_mutex);
}

void thread_pool_queue_stats(int *array)
{
        int i;

        if (pool_initialized) {
                for (i = 0; i < RAD_LISTEN_MAX; i++) {
                        array[i] = fr_fifo_num_elements(thread_pool.fifo[i]);
                }
        } else {
                for (i = 0; i < RAD_LISTEN_MAX; i++) {
                        array[i] = 0;
                }
        }
}
#endif /* HAVE_PTHREAD_H */