/*
 * Copyright (C) 1994-2021 Altair Engineering, Inc.
 * For more information, contact Altair at www.altair.com.
 *
 * This file is part of both the OpenPBS software ("OpenPBS")
 * and the PBS Professional ("PBS Pro") software.
 *
 * Open Source License Information:
 *
 * OpenPBS is free software. You can redistribute it and/or modify it under
 * the terms of the GNU Affero General Public License as published by the
 * Free Software Foundation, either version 3 of the License, or (at your
 * option) any later version.
 *
 * OpenPBS 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/>.
 *
 * Commercial License Information:
 *
 * PBS Pro is commercially licensed software that shares a common core with
 * the OpenPBS software.  For a copy of the commercial license terms and
 * conditions, go to: (http://www.pbspro.com/agreement.html) or contact the
 * Altair Legal Department.
 *
 * Altair's dual-license business model allows companies, individuals, and
 * organizations to create proprietary derivative works of OpenPBS and
 * distribute them - whether embedded or bundled with other software -
 * under a commercial license agreement.
 *
 * Use of Altair's trademarks, including but not limited to "PBS™",
 * "OpenPBS®", "PBS Professional®", and "PBS Pro™" and Altair's logos is
 * subject to Altair's trademark licensing policies.
 */

/**
 * contains functions related to PBS scheduler
 */
#include <pbs_config.h> /* the master config generated by configure */

#include <vector>

#include <arpa/inet.h>
#include <ctype.h>
#include <errno.h>
#include <fcntl.h>
#include <grp.h>
#include <limits.h>
#include <netdb.h>
#include <netinet/in.h>
#include <pwd.h>
#include <signal.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/resource.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/types.h>
#include <unistd.h>
#ifdef _POSIX_MEMLOCK
#include <sys/mman.h>
#endif /* _POSIX_MEMLOCK */

#if defined(FD_SET_IN_SYS_SELECT_H)
#include <sys/select.h>
#endif
#include <sys/resource.h>

#include "auth.h"
#include "config.h"
#include "fifo.h"
#include "globals.h"
#include "libpbs.h"
#include "libsec.h"
#include "list_link.h"
#include "log.h"
#include "misc.h"
#include "multi_threading.h"
#include "net_connect.h"
#include "pbs_ecl.h"
#include "pbs_error.h"
#include "pbs_ifl.h"
#include "pbs_share.h"
#include "pbs_version.h"
#include "portability.h"
#include "rm.h"
#include "sched_cmds.h"
#include "server_limits.h"
#include "tpp.h"
#include "libutil.h"

#define START_CLIENTS 2			   /* minimum number of clients */
auto okclients = std::vector<pbs_net_t>(); /* accept connections from */
char *configfile = NULL;		   /* name of file containing
						 client names to be added */

extern char *msg_daemonname;
char **glob_argv;
char usage[] = "[-d home][-L logfile][-p file][-I schedname][-n][-N][-c clientsfile][-t num threads]";
struct sockaddr_in saddr;
sigset_t allsigs;
sigset_t oldsigs;

int sigstoblock[] = {SIGHUP, SIGINT, SIGTERM, SIGUSR1};

/* if we received a sigpipe, this probably means the server went away. */

/* used in segv restart */
time_t segv_start_time;
time_t segv_last_time;

#ifdef NAS /* localmod 030 */
extern int do_soft_cycle_interrupt;
extern int do_hard_cycle_interrupt;
#endif /* localmod 030 */

extern char *msg_startup1;

pthread_mutex_t cleanup_lock;

static void reconnect_server(void);
void sched_svr_init(void);
void open_server_conns();
static int schedule_wrapper(sched_cmd *cmd, int opt_no_restart);
void close_server_conns(void);

typedef int (*schedule_func)(int, const sched_cmd *);

static schedule_func schedule_ptr;

/**
 * @brief
 * 		cleanup after a segv and re-exec.  Trust as little global mem
 * 		as possible... we don't know if it could be corrupt
 *
 * @param[in]	sig	-	signal
 */
void
on_segv(int sig)
{
	int ret_lock;

	/* We want any other threads to block here, we want them alive until abort() is called
	 * as it dumps core for all threads
	 */
	ret_lock = pthread_mutex_lock(&cleanup_lock);
	if (ret_lock != 0)
		pthread_exit(NULL);

	/* we crashed less then 5 minutes ago, lets not restart ourself */
	if ((segv_last_time - segv_start_time) < 300) {
		log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__,
			   "received a sigsegv within 5 minutes of start: aborting.");

		/* Not unlocking mutex on purpose, we need to hold on to it until the process is killed */
		abort();
	}

	log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__,
		   "received segv and restarting");

	if (fork() > 0) {  /* the parent rexec's itself */
		sleep(10); /* allow the child to die */
		execv(glob_argv[0], glob_argv);
		exit(3);
	} else {
		abort(); /* allow to core and exit */
	}
}

/**
 * @brief
 * 		signal function for receiving a sigpipe - set flag so we know not to talk
 * 		to the server any more and leave the cycle as soon as possible
 *
 * @param[in]	sig	-	sigpipe
 */
void
sigfunc_pipe(int sig)
{
	log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, "sigfunc_pipe", "We've received a sigpipe: The server probably died.");
	got_sigpipe = 1;
}

/**
 * @brief	cleanup routine for scheduler exit
 *
 * @param	void
 *
 * @return void
 */
static void
schedexit(void)
{
	/* close any open connections to peers */
	for (auto &pq : conf.peer_queues) {
		if (pq.peer_sd >= 0) {
			/* When peering "local", do not disconnect server */
			if (!pq.remote_server.empty())
				pbs_disconnect(pq.peer_sd);
			pq.peer_sd = -1;
		}
	}

	/* Kill all worker threads */
	if (num_threads > 1) {
		int *thid;

		thid = (int *) pthread_getspecific(th_id_key);

		if (*thid == 0) {
			kill_threads();
			close_server_conns();
			return;
		}
	}

	close_server_conns();
}

/**
 * @brief
 *       Clean up after a signal.
 *
 *  @param[in]	sig	-	signal
 */
void
die(int sig)
{
	int ret_lock;

	ret_lock = pthread_mutex_trylock(&cleanup_lock);
	if (ret_lock != 0)
		pthread_exit(NULL);

	if (sig > 0)
		log_eventf(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__, "caught signal %d", sig);
	else
		log_event(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__, "abnormal termination");

	schedexit();

	{
		int csret;
		if ((csret = CS_close_app()) != CS_SUCCESS) {
			/*had some problem closing the security library*/

			sprintf(log_buffer, "problem closing security library (%d)", csret);
			log_err(-1, "pbs_sched", log_buffer);
		}
	}

	unload_auths();

	log_close(1);
	exit(1);
}

/**
 * @brief
 * 		add a new client to the list of clients.
 *
 * @param[in]	name	-	Client name.
 */
int
addclient(const char *name)
{
	int i;
	struct hostent *host;
	struct in_addr saddr;

	if ((host = gethostbyname(name)) == NULL) {
		sprintf(log_buffer, "host %s not found", name);
		log_err(-1, __func__, log_buffer);
		return -1;
	}

	for (i = 0; host->h_addr_list[i]; i++) {
		memcpy((char *) &saddr, host->h_addr_list[i], host->h_length);
		okclients.push_back(saddr.s_addr);
	}
	return 0;
}

/**
 * @brief
 * 		read_config - read and process the configuration file (see -c option)
 * @par
 *		Currently, the only statement is $clienthost to specify which systems
 *		can contact the scheduler.
 *
 * @param[in]	file	-	configuration file
 *
 * @return	int
 * @retval	0	: Ok
 * @retval	-1	: !nOtOk!
 */
#define CONF_LINE_LEN 120

static int
read_config(char *file)
{
	FILE *conf;
	int i;
	char line[CONF_LINE_LEN];
	char *token;
	struct specialconfig {
		const char *name;
		int (*handler)(const char *);
	} special[] = {
		{"clienthost", addclient},
		{NULL, NULL}};

#if !defined(DEBUG) && !defined(NO_SECURITY_CHECK)
	if (chk_file_sec_user(file, 0, 0, S_IWGRP | S_IWOTH, 1, getuid()))
		return (-1);
#endif

	if ((conf = fopen(file, "r")) == NULL) {
		log_err(errno, __func__, "cannot open config file");
		return (-1);
	}
	while (fgets(line, CONF_LINE_LEN, conf)) {

		if ((line[0] == '#') || (line[0] == '\n'))
			continue;	   /* ignore comment & null line */
		else if (line[0] == '$') { /* special */

			if ((token = strtok(line, " \t")) == NULL)
				token = const_cast<char *>("");
			for (i = 0; special[i].name; i++) {
				if (strcmp(token + 1, special[i].name) == 0)
					break;
			}
			if (special[i].name == NULL) {
				sprintf(log_buffer, "config name %s not known",
					token);
				log_record(PBSEVENT_ERROR,
					   PBS_EVENTCLASS_SERVER, LOG_INFO,
					   msg_daemonname, log_buffer);
				continue;
			}
			token = strtok(NULL, " \t");
			if (*(token + strlen(token) - 1) == '\n')
				*(token + strlen(token) - 1) = '\0';
			if (special[i].handler(token)) {
				fclose(conf);
				return (-1);
			}

		} else {
			log_record(PBSEVENT_ERROR, PBS_EVENTCLASS_SERVER,
				   LOG_INFO, msg_daemonname,
				   "invalid line in config file");
			fclose(conf);
			return (-1);
		}
	}
	fclose(conf);
	return (0);
}
/**
 * @brief
 * 		restart on signal
 *
 * @param[in]	sig	-	signal
 */
void
restart(int sig)
{
	const sched_cmd cmd = {SCH_CONFIGURE, NULL};

	if (sig) {
		log_close(1);
		log_open(logfile, path_log);
		sprintf(log_buffer, "restart on signal %d", sig);
	} else {
		sprintf(log_buffer, "restart command");
	}
	log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__, log_buffer);
	if (configfile) {
		if (read_config(configfile) != 0)
			die(0);
	}
	schedule_ptr(clust_primary_sock, &cmd);
}

#ifdef NAS /* localmod 030 */
/**
 * @brief
 * 		make soft cycle interrupt active
 *
 * @param[in]	sig	-	signal
 */
void
soft_cycle_interrupt(int sig)
{
	do_soft_cycle_interrupt = 1;
}
/**
 * @brief
 * 		make hard cycle interrupt active
 *
 * @param[in]	sig	-	signal
 */
void
hard_cycle_interrupt(int sig)
{
	do_hard_cycle_interrupt = 1;
}
#endif /* localmod 030 */

/**
 * @brief
 * 		lock_out - lock out other daemons from this directory.
 *
 * @param[in]	fds	-	file descriptor
 * @param[in]	op	-	F_WRLCK  or  F_UNLCK
 *
 * @return	1
 */

static void
lock_out(int fds, int op)
{
	struct flock flock;

	(void) lseek(fds, (off_t) 0, SEEK_SET);
	flock.l_type = op;
	flock.l_whence = SEEK_SET;
	flock.l_start = 0;
	flock.l_len = 0; /* whole file */
	if (fcntl(fds, F_SETLK, &flock) < 0) {
		log_err(errno, msg_daemonname, "another scheduler running");
		fprintf(stderr, "pbs_sched: another scheduler running\n");
		exit(1);
	}
}

/**
 * @brief
 * 		are_we_primary - are we on the primary Server host
 *		If either the only configured Server or the Primary in a failover
 *		configuration - return true
 *
 * @return	int
 * @retval	0	: we are the secondary
 * @retval	-1	: cannot be neither
 * @retval	1	: we are the listed primary
 */
static int
are_we_primary()
{
	char server_host[PBS_MAXHOSTNAME + 1];
	char hn1[PBS_MAXHOSTNAME + 1];

	if (pbs_conf.pbs_leaf_name) {
		char *endp;
		snprintf(server_host, sizeof(server_host), "%s", pbs_conf.pbs_leaf_name);
		endp = strchr(server_host, ','); /* find the first name */
		if (endp)
			*endp = '\0';
		endp = strchr(server_host, ':'); /* cut out the port */
		if (endp)
			*endp = '\0';
	} else if ((gethostname(server_host, (sizeof(server_host) - 1)) == -1) ||
		   (get_fullhostname(server_host, server_host, (sizeof(server_host) - 1)) == -1)) {
		log_err(-1, __func__, "Unable to get my host name");
		return -1;
	}

	/* both secondary and primary should be set or neither set */
	if ((pbs_conf.pbs_secondary == NULL) && (pbs_conf.pbs_primary == NULL))
		return 1;
	if ((pbs_conf.pbs_secondary == NULL) || (pbs_conf.pbs_primary == NULL))
		return -1;

	if (get_fullhostname(pbs_conf.pbs_primary, hn1, (sizeof(hn1) - 1)) == -1) {
		log_err(-1, __func__, "Unable to get full host name of primary");
		return -1;
	}

	if (strcmp(hn1, server_host) == 0)
		return 1; /* we are the listed primary */

	if (get_fullhostname(pbs_conf.pbs_secondary, hn1, (sizeof(hn1) - 1)) == -1) {
		log_err(-1, __func__, "Unable to get full host name of secondary");
		return -1;
	}
	if (strcmp(hn1, server_host) == 0)
		return 0; /* we are the secondary */

	return -1; /* cannot be neither */
}

/**
 * @brief close connections to server
 *
 * @return void
 */
void
close_server_conns(void)
{
	int i;

	pbs_disconnect(clust_primary_sock);
	pbs_disconnect(clust_secondary_sock);

	tpp_em_destroy(poll_context);
	poll_context = NULL;

	/* free qrun_list */
	for (i = 0; i < qrun_list_size; i++) {
		if (qrun_list[i].jid != NULL)
			free(qrun_list[i].jid);
	}
	free(qrun_list);
	qrun_list = NULL;

	clust_primary_sock = -1;
	clust_secondary_sock = -1;
}

/**
 * @brief Connect to server. Also add secondary connection to the poll list
 *
 *
 * @return void
 */
void
open_server_conns(void)
{
	sigset_t prevsigs;

	while (1) {
		sigemptyset(&prevsigs);
		/*
	 	 * Connecting to server may potentially fork for reserve port authentication.
	 	 * Call to connect to server must be protected from signals because it can cause
	 	 * scheduler to go into a deadlock on logging mutex.
	 	 */
		if (sigprocmask(SIG_BLOCK, &allsigs, &prevsigs) == -1)
			log_err(errno, __func__, "sigprocmask(SIG_BLOCK)");

		if (clust_primary_sock < 0) {
			clust_primary_sock = pbs_connect(NULL);
			if (clust_primary_sock < 0)
				goto unmask_continue;
		}
		clust_secondary_sock = pbs_connect(NULL);
		if (clust_secondary_sock < 0 || clust_primary_sock == clust_secondary_sock)
			goto unmask_continue;

		if (pbs_register_sched(sc_name, clust_primary_sock, clust_secondary_sock) != 0) {
			log_errf(pbs_errno, __func__, "Couldn't register the scheduler %s with connected server", sc_name);
			goto unmask_continue;
		}

		/* Reached here means everything is success, so we will break out of the loop */
		if (sigprocmask(SIG_SETMASK, &prevsigs, NULL) == -1)
			log_err(errno, __func__, "sigprocmask(SIG_SETMASK)");
		break;

	unmask_continue:
		if (sigprocmask(SIG_SETMASK, &prevsigs, NULL) == -1)
			log_err(errno, __func__, "sigprocmask(SIG_SETMASK)");

		/* wait for 2s for not to burn too much CPU, and then retry connection */
		sleep(2);
		continue;
	}
	log_eventf(PBSEVENT_ADMIN | PBSEVENT_FORCE, PBS_EVENTCLASS_SCHED,
		   LOG_INFO, msg_daemonname, "Connected to the server");

	sched_svr_init();

	if (tpp_em_add_fd(poll_context, clust_secondary_sock, EM_IN | EM_HUP | EM_ERR) < 0) {
		log_errf(errno, __func__, "Couldn't add secondary connection to poll list");
		die(-1);
	}
}

/**
 * @brief Initialises event poll context for the server and also sched commands queue
 *
 * @return void
 */
void
sched_svr_init(void)
{
	if (poll_context == NULL) {
		poll_context = tpp_em_init(1);
		if (poll_context == NULL) {
			log_err(errno, __func__, "Failed to init cmd connections context");
			die(-1);
		}
	}

	qrun_list = static_cast<sched_cmd *>(malloc(2 * sizeof(sched_cmd)));
	if (qrun_list == NULL) {
		log_err(errno, __func__, MEM_ERR_MSG);
		die(0);
	}
}

/**
 * @brief reconnect to all the servers configured
 *
 * @return void
 */
static void
reconnect_server(void)
{
	close_server_conns();
	open_server_conns();
	return;
}

/**
 * @brief read incoming command from given secondary connection
 *        and add it into sched_cmds array
 *
 * @param[in]  sock   - secondary connection to server
 *
 * @return int
 * @retval -2 - failure due to memory operation failed
 * @retval -1 - failure while reading command
 * @return  0 - no cmd, server might have closed connection
 * @return  1 - success, read atleast one command
 */
static int
read_sched_cmd(int sock)
{
	int rc;
	sched_cmd cmd;

	rc = get_sched_cmd(sock, &cmd);
	if (rc != 1)
		return rc;
	else {
		sched_cmd cmd_prio;
		/*
		 * There is possibility that server has sent
		 * priority command after first non-priority command,
		 * while we were in schedule()
		 *
		 * so try read it in non-blocking mode, but don't
		 * return any failure if fails to read, as we have
		 * successfully enqueued first command
		 *
		 * and if we get priority command then just ignore it
		 * since we are not yet in middle of schedule cycle
		 */
		int rc_prio = get_sched_cmd_noblk(sock, &cmd_prio);
		if (rc_prio == -2)
			return 0;
	}

	if (cmd.cmd != SCH_SCHEDULE_RESTART_CYCLE) {
		if (cmd.cmd == SCH_SCHEDULE_AJOB)
			qrun_list[qrun_list_size++] = cmd;
		else {
			if (cmd.cmd >= SCH_SCHEDULE_NULL && cmd.cmd < SCH_CMD_HIGH)
				sched_cmds[cmd.cmd] = 1;
		}
	}

	return rc;
}

/**
 * @brief wait for commands from servers
 *
 * @return void
 */
static void
wait_for_cmds()
{
	int i;
	em_event_t *events;
	int hascmd = 0;
	sigset_t emptyset;

	qrun_list_size = 0;

	while (!hascmd) {
		sigemptyset(&emptyset);
		auto nsocks = tpp_em_pwait(poll_context, &events, -1, &emptyset);
		auto err = errno;

		if (nsocks < 0) {
			if (!(err == EINTR || err == EAGAIN || err == 0)) {
				log_errf(err, __func__, " tpp_em_wait() error, errno=%d", err);
				sleep(1); /* wait for 1s for not to burn too much CPU */
			}
		} else {
			for (i = 0; i < nsocks; i++) {
				int sock = EM_GET_FD(events, i);
				err = read_sched_cmd(sock);
				if (err != 1) {
					/* if memory error ignore, else reconnect server */
					if (err != -2)
						reconnect_server();
				} else
					hascmd = 1;
			}
		}
	}
}

/**
 * @brief	Helper function for send_cycle_end, just sends the message out
 *
 * @param	sec_conn - secondary conn sd to the server
 *
 * @return	bool
 * @retval	true for success
 * @retval	false for failure/error
 */
bool
send_cycle_end_msg(int sec_conn)
{
	static int cycle_end_marker = 0;

	if (diswsi(sec_conn, cycle_end_marker) != DIS_SUCCESS) {
		log_err(errno, __func__, "Not able to send end of cycle");
		return false;
	}

	if (dis_flush(sec_conn) != 0) {
		log_err(errno, __func__, "Not able to send end of cycle");
		return false;
	}

	return true;
}

/**
 *
 * @brief sends end of cycle indication to all the servers configured
 *
 * @param[in] sd - socket descriptor to send end of cycle notification
 *
 * @return void
 */
static void
send_cycle_end(void)
{
	if (!send_cycle_end_msg(clust_secondary_sock))
		goto reconnect;

	if (got_sigpipe)
		goto reconnect;

	return;

reconnect:
	reconnect_server();
	got_sigpipe = 0;
}

int
sched_main(int argc, char *argv[], schedule_func sched_ptr)
{
	int go;
	int c;
	int errflg = 0;
	int lockfds;
	pid_t pid;
	char host[PBS_MAXHOSTNAME + 1];
#ifndef DEBUG
	const char *dbfile = "sched_out";
#endif
	struct sigaction act;
	int opt_no_restart = 0;
	int stalone = 0;
#ifdef _POSIX_MEMLOCK
	int do_mlockall = 0;
#endif /* _POSIX_MEMLOCK */
	int nthreads = -1;
	int num_cores;
	char *endp = NULL;
	pthread_mutexattr_t attr;

	/*the real deal or show version and exit?*/

	schedule_ptr = sched_ptr;

	PRINT_VERSION_AND_EXIT(argc, argv);

	num_cores = sysconf(_SC_NPROCESSORS_ONLN);

	if (pbs_loadconf(0) == 0)
		return (1);

	if (validate_running_user(argv[0]) == 0)
		return (1);

	/* disable attribute verification */
	set_no_attribute_verification();

	/* initialize the thread context */
	if (pbs_client_thread_init_thread_context() != 0) {
		fprintf(stderr, "%s: Unable to initialize thread context\n",
			argv[0]);
		return (1);
	}

	set_log_conf(pbs_conf.pbs_leaf_name, pbs_conf.pbs_mom_node_name,
		     pbs_conf.locallog, pbs_conf.syslogfac,
		     pbs_conf.syslogsvr, pbs_conf.pbs_log_highres_timestamp);

	nthreads = pbs_conf.pbs_sched_threads;

	glob_argv = argv;
	segv_start_time = segv_last_time = time(NULL);

	opterr = 0;
	while ((c = getopt(argc, argv, "lL:NI:d:p:c:nt:")) != EOF) {
		switch (c) {
			case 'l':
#ifdef _POSIX_MEMLOCK
				do_mlockall = 1;
#else
				fprintf(stderr, "-l option - mlockall not supported\n");
#endif /* _POSIX_MEMLOCK */
				break;
			case 'L':
				logfile = optarg;
				break;
			case 'N':
				stalone = 1;
				break;
			case 'I':
				sc_name = optarg;
				break;
			case 'd':
				if (pbs_conf.pbs_home_path != NULL)
					free(pbs_conf.pbs_home_path);
				pbs_conf.pbs_home_path = optarg;
				break;
			case 'p':
#ifndef DEBUG
				dbfile = optarg;
#endif
				break;
			case 'c':
				configfile = optarg;
				break;
			case 'n':
				opt_no_restart = 1;
				break;
			case 't':
				nthreads = strtol(optarg, &endp, 10);
				if (*endp != '\0') {
					fprintf(stderr, "%s: bad num threads value\n", optarg);
					errflg = 1;
				}
				if (nthreads < 1) {
					fprintf(stderr, "%s: bad num threads value (should be in range 1-99999)\n", optarg);
					errflg = 1;
				}
				if (nthreads > num_cores) {
					fprintf(stderr, "%s: cannot be larger than number of cores %d, using number of cores instead\n",
						optarg, num_cores);
					nthreads = num_cores;
				}
				break;
			default:
				errflg = 1;
				break;
		}
	}

	if (sc_name == NULL) {
		sc_name = PBS_DFLT_SCHED_NAME;
		dflt_sched = 1;
	}

	if (errflg) {
		fprintf(stderr, "usage: %s %s\n", argv[0], usage);
		fprintf(stderr, "       %s --version\n", argv[0]);
		exit(1);
	}

	if (dflt_sched) {
		(void) sprintf(log_buffer, "%s/sched_priv", pbs_conf.pbs_home_path);
	} else {
		(void) sprintf(log_buffer, "%s/sched_priv_%s", pbs_conf.pbs_home_path, sc_name);
	}
#if !defined(DEBUG) && !defined(NO_SECURITY_CHECK)
	c = chk_file_sec_user(log_buffer, 1, 0, S_IWGRP | S_IWOTH, 1, getuid());
	c |= chk_file_sec(pbs_conf.pbs_environment, 0, 0, S_IWGRP | S_IWOTH, 0);
	if (c != 0)
		exit(1);
#endif /* not DEBUG and not NO_SECURITY_CHECK */
	if (chdir(log_buffer) == -1) {
		perror("chdir");
		exit(1);
	}
	if (dflt_sched) {
		(void) sprintf(path_log, "%s/sched_logs", pbs_conf.pbs_home_path);
	} else {
		(void) sprintf(path_log, "%s/sched_logs_%s", pbs_conf.pbs_home_path, sc_name);
	}
	if (log_open(logfile, path_log) == -1) {
		fprintf(stderr, "%s: logfile could not be opened\n", argv[0]);
		exit(1);
	}

	/* The following is code to reduce security risks                */
	/* start out with standard umask, system resource limit infinite */

	umask(022);
	if (setup_env(pbs_conf.pbs_environment) == -1)
		exit(1);
	c = getgid();
	(void) setgroups(1, (gid_t *) &c); /* secure suppl. groups */

	set_proc_limits(pbs_conf.pbs_core_limit, 0); /* set_proc_limits can call log_record, so call only after opening log file */

	if (gethostname(host, (sizeof(host) - 1)) == -1) {
		log_err(errno, __func__, "gethostname");
		die(0);
	}

	/*Initialize security library's internal data structures*/
	if (load_auths(AUTH_SERVER)) {
		log_err(-1, "pbs_sched", "Failed to load auth lib");
		die(0);
	}

	{
		int csret;

		/* let Libsec do logging if part of PBS daemon code */
		p_cslog = log_err;

		if ((csret = CS_server_init()) != CS_SUCCESS) {
			sprintf(log_buffer,
				"Problem initializing security library (%d)", csret);
			log_err(-1, "pbs_sched", log_buffer);
			die(0);
		}
	}

	addclient("localhost"); /* who has permission to call MOM */
	addclient(host);
	if (pbs_conf.pbs_server_name)
		addclient(pbs_conf.pbs_server_name);
	if (pbs_conf.pbs_primary && pbs_conf.pbs_secondary) {
		/* Failover is configured when both primary and secondary are set. */
		addclient(pbs_conf.pbs_primary);
		addclient(pbs_conf.pbs_secondary);
	} else if (pbs_conf.pbs_server_host_name) {
		/* Failover is not configured, but PBS_SERVER_HOST_NAME is. */
		addclient(pbs_conf.pbs_server_host_name);
	}
	if (pbs_conf.pbs_leaf_name)
		addclient(pbs_conf.pbs_leaf_name);

	if (configfile) {
		if (read_config(configfile) != 0)
			die(0);
	}

	if ((c = are_we_primary()) == 1) {
		lockfds = open("sched.lock", O_CREAT | O_WRONLY, 0644);
	} else if (c == 0) {
		lockfds = open("sched.lock.secondary", O_CREAT | O_WRONLY, 0644);
	} else {
		log_err(-1, "pbs_sched", "neither primary or secondary server");
		exit(1);
	}
	if (lockfds < 0) {
		log_err(errno, __func__, "open lock file");
		exit(1);
	}

	if (sigemptyset(&allsigs) == -1) {
		perror("sigemptyset");
		exit(1);
	}
	if (sigprocmask(SIG_SETMASK, &allsigs, NULL) == -1) { /* unblock */
		perror("sigprocmask");
		exit(1);
	}
	act.sa_flags = 0;

	/* remember to block these during critical sections so we don't get confused */
	for (auto &sig : sigstoblock) {
		sigaddset(&allsigs, sig);
	}
	act.sa_mask = allsigs;

	act.sa_handler = restart; /* do a restart on SIGHUP */
	sigaction(SIGHUP, &act, NULL);

#ifdef NAS				       /* localmod 030 */
	act.sa_handler = soft_cycle_interrupt; /* do a cycle interrupt on */
					       /* SIGUSR1, subject to     */
					       /* configurable parameters */
	sigaction(SIGUSR1, &act, NULL);
	act.sa_handler = hard_cycle_interrupt; /* do a cycle interrupt on */
					       /* SIGUSR2                 */
	sigaction(SIGUSR2, &act, NULL);
#endif /* localmod 030 */

	act.sa_handler = die; /* bite the biscuit for all following */
	sigaction(SIGINT, &act, NULL);
	sigaction(SIGTERM, &act, NULL);

	act.sa_handler = sigfunc_pipe;
	sigaction(SIGPIPE, &act, NULL);

	if (!opt_no_restart) {
		act.sa_handler = on_segv;
		sigaction(SIGSEGV, &act, NULL);
		sigaction(SIGBUS, &act, NULL);
	}

#ifndef DEBUG
	if (stalone != 1) {
		if ((pid = fork()) == -1) { /* error on fork */
			perror("fork");
			exit(1);
		} else if (pid > 0) /* parent exits */
			exit(0);

		if (setsid() == -1) {
			perror("setsid");
			exit(1);
		}
	}
	lock_out(lockfds, F_WRLCK);
	if (freopen(dbfile, "a", stdout) == NULL) 
		log_errf(-1, __func__, "freopen failed. ERR : %s",strerror(errno));
	setvbuf(stdout, NULL, _IOLBF, 0);
	dup2(fileno(stdout), fileno(stderr));
#else
	if (stalone != 1) {
		(void) sprintf(log_buffer, "Debug build does not fork.");
		log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO,
			   __func__, log_buffer);
	}
	lock_out(lockfds, F_WRLCK);
	setvbuf(stdout, NULL, _IOLBF, 0);
	setvbuf(stderr, NULL, _IOLBF, 0);
#endif
	pid = getpid();
	daemon_protect(0, PBS_DAEMON_PROTECT_ON);
	if (freopen("/dev/null", "r", stdin) == NULL) 
		log_errf(-1, __func__, "freopen failed. ERR : %s",strerror(errno));

	/* write schedulers pid into lockfile */
	if (ftruncate(lockfds, (off_t) 0) == -1) 
		log_errf(-1, __func__, "ftruncate failed. ERR : %s",strerror(errno));
	(void) sprintf(log_buffer, "%ld\n", (long) pid);
	if (write(lockfds, log_buffer, strlen(log_buffer)) == -1) 
		log_errf(-1, __func__, "write failed. ERR : %s",strerror(errno));

#ifdef _POSIX_MEMLOCK
	if (do_mlockall == 1) {
		if (mlockall(MCL_CURRENT | MCL_FUTURE) == -1) {
			log_err(errno, __func__, "mlockall failed");
		}
	}
#endif /* _POSIX_MEMLOCK */

	(void) sprintf(log_buffer, msg_startup1, PBS_VERSION, 0);
	log_event(PBSEVENT_SYSTEM | PBSEVENT_ADMIN | PBSEVENT_FORCE,
		  LOG_NOTICE, PBS_EVENTCLASS_SERVER, msg_daemonname, log_buffer);

	sprintf(log_buffer, "%s startup pid %ld", argv[0], (long) pid);
	log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__, log_buffer);

	/*
	 *  Local initialization stuff
	 */
	/* Set the signal mask temporarily for thread initialization */
	if (sigprocmask(SIG_BLOCK, &allsigs, &oldsigs) == -1)
		log_err(errno, __func__, "sigprocmask(SIG_BLOCK)");
	if (schedinit(nthreads)) {
		(void) sprintf(log_buffer,
			       "local initialization failed, terminating");
		log_record(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO,
			   __func__, log_buffer);
		exit(1);
	}
	if (sigprocmask(SIG_SETMASK, &oldsigs, NULL) == -1)
		log_err(errno, __func__, "sigprocmask(SIG_SETMASK)");

	/* Initialize cleanup lock */
	if (init_mutex_attr_recursive(&attr) != 0)
		die(0);

	pthread_mutex_init(&cleanup_lock, &attr);

	open_server_conns();

	for (go = 1; go;) {
		int i;

		wait_for_cmds();

		/* First walk through the qrun_list as this has high priority followed by the other commands */
		for (i = 0; i < qrun_list_size; i++) {
			if (schedule_wrapper(&qrun_list[i], opt_no_restart) == 1) {
				go = 0;
				break;
			}
			free(qrun_list[i].jid);
			qrun_list[i].jid = NULL;
		}

		for (i = 0; go && (i < SCH_CMD_HIGH); i++) {
			sched_cmd cmd;

			if (sched_cmds[i] == 0)
				continue;

			/* index itself is the command */
			cmd.cmd = i;

			/* jid is always NULL since this list does not contain SCHEDULE_AJOB commands */
			cmd.jid = NULL;

			/* clear the entry of sched_cmds[i] as we are going to process this command now */
			sched_cmds[i] = 0;

			if (schedule_wrapper(&cmd, opt_no_restart) == 1) {
				go = 0;
				break;
			}
		}
	}

	schedexit();

	log_eventf(PBSEVENT_SYSTEM, PBS_EVENTCLASS_SERVER, LOG_INFO, __func__, "%s normal finish pid %ld", argv[0], (long) pid);
	lock_out(lockfds, F_UNLCK);

	unload_auths();
	close_server_conns();
	log_close(1);
	exit(0);
}

/**
 * @brief
 *	schedule_wrapper - Wrapper function for calling schedule
 *
 * @param[in] cmd - pointer to schedulig command
 * @param[in] opt_no_restart - value of opt_no_restart
 *
 * @return	int
 * @retval	0	: continue calling scheduling cycles
 * @retval	1	: exit scheduler
 */
static int
schedule_wrapper(sched_cmd *cmd, int opt_no_restart)
{
	time_t now;

	if (sigprocmask(SIG_BLOCK, &allsigs, &oldsigs) == -1)
		log_err(errno, __func__, "sigprocmask(SIG_BLOCK)");

	/* Keep track of time to use in SIGSEGV handler */
	now = time(NULL);
	if (!opt_no_restart)
		segv_last_time = now;

#ifdef DEBUG
	{
		strftime(log_buffer, sizeof(log_buffer), "%Y-%m-%d %H:%M:%S", localtime(&now));
		DBPRT(("%s Scheduler received command %d\n", log_buffer, cmd->cmd));
	}
#endif

	if (schedule_ptr(clust_primary_sock, cmd)) /* magic happens here */
		return 1;
	else
		send_cycle_end();

	if (sigprocmask(SIG_SETMASK, &oldsigs, NULL) == -1)
		log_err(errno, __func__, "sigprocmask(SIG_SETMASK)");

	return 0;
}