X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=client.c;h=7915268fe0efa3b50ae80fd254ffdca3d4a816ea;hp=358903bb1f29bff18df48b67c189e38c7d05149f;hb=ad705bcb7e79a7cdb9891db17b4c40b13b6c30c3;hpb=3f3a4542887ba038231144be08b2c36d7cc83424

diff --git a/client.c b/client.c
index 358903bb..7915268f 100644
--- a/client.c
+++ b/client.c
@@ -21,19 +21,23 @@
 #include "hash.h"
 
 struct client_eta {
-	struct jobs_eta eta;
 	unsigned int pending;
+	struct jobs_eta eta;
 };
 
 struct fio_client {
 	struct flist_head list;
 	struct flist_head hash_list;
 	struct flist_head arg_list;
-	struct sockaddr_in addr;
-	struct sockaddr_un addr_un;
+	union {
+		struct sockaddr_in addr;
+		struct sockaddr_in6 addr6;
+		struct sockaddr_un addr_un;
+	};
 	char *hostname;
 	int port;
 	int fd;
+	unsigned int refs;
 
 	char *name;
 
@@ -41,12 +45,24 @@ struct fio_client {
 
 	int skip_newline;
 	int is_sock;
+	int disk_stats_shown;
+	unsigned int jobs;
+	unsigned int nr_stat;
+	int error;
+	int ipv6;
+	int sent_job;
+	int did_stat;
 
 	struct flist_head eta_list;
 	struct client_eta *eta_in_flight;
 
+	struct flist_head cmd_list;
+
 	uint16_t argc;
 	char **argv;
+
+	char **ini_file;
+	unsigned int nr_ini_file;
 };
 
 static struct timeval eta_tv;
@@ -55,8 +71,9 @@ enum {
 	Client_created		= 0,
 	Client_connected	= 1,
 	Client_started		= 2,
-	Client_stopped		= 3,
-	Client_exited		= 4,
+	Client_running		= 3,
+	Client_stopped		= 4,
+	Client_exited		= 5,
 };
 
 static FLIST_HEAD(client_list);
@@ -64,6 +81,12 @@ static FLIST_HEAD(eta_list);
 
 static FLIST_HEAD(arg_list);
 
+static struct thread_stat client_ts;
+static struct group_run_stats client_gs;
+static int sum_stat_clients = 0;
+static int sum_stat_nr;
+static int do_output_all_clients;
+
 #define FIO_CLIENT_HASH_BITS	7
 #define FIO_CLIENT_HASH_SZ	(1 << FIO_CLIENT_HASH_BITS)
 #define FIO_CLIENT_HASH_MASK	(FIO_CLIENT_HASH_SZ - 1)
@@ -103,8 +126,10 @@ static struct fio_client *find_client_by_fd(int fd)
 	flist_for_each(entry, &client_hash[bucket]) {
 		client = flist_entry(entry, struct fio_client, hash_list);
 
-		if (client->fd == fd)
+		if (client->fd == fd) {
+			client->refs++;
 			return client;
+		}
 	}
 
 	return NULL;
@@ -112,6 +137,11 @@ static struct fio_client *find_client_by_fd(int fd)
 
 static void remove_client(struct fio_client *client)
 {
+	assert(client->refs);
+
+	if (--client->refs)
+		return;
+
 	dprint(FD_NET, "client: removed <%s>\n", client->hostname);
 	flist_del(&client->list);
 
@@ -127,11 +157,23 @@ static void remove_client(struct fio_client *client)
 		free(client->argv);
 	if (client->name)
 		free(client->name);
+	while (client->nr_ini_file)
+		free(client->ini_file[--client->nr_ini_file]);
+	if (client->ini_file)
+		free(client->ini_file);
+
+	if (!client->did_stat)
+		sum_stat_clients -= client->nr_stat;
 
 	free(client);
 	nr_clients--;
 }
 
+static void put_client(struct fio_client *client)
+{
+	remove_client(client);
+}
+
 static void __fio_client_add_cmd_option(struct fio_client *client,
 					const char *opt)
 {
@@ -163,6 +205,19 @@ void fio_client_add_cmd_option(void *cookie, const char *opt)
 	}
 }
 
+void fio_client_add_ini_file(void *cookie, const char *ini_file)
+{
+	struct fio_client *client = cookie;
+	size_t new_size;
+
+	dprint(FD_NET, "client <%s>: add ini %s\n", client->hostname, ini_file);
+
+	new_size = (client->nr_ini_file + 1) * sizeof(char *);
+	client->ini_file = realloc(client->ini_file, new_size);
+	client->ini_file[client->nr_ini_file] = strdup(ini_file);
+	client->nr_ini_file++;
+}
+
 int fio_client_add(const char *hostname, void **cookie)
 {
 	struct fio_client *existing = *cookie;
@@ -188,13 +243,17 @@ int fio_client_add(const char *hostname, void **cookie)
 	INIT_FLIST_HEAD(&client->hash_list);
 	INIT_FLIST_HEAD(&client->arg_list);
 	INIT_FLIST_HEAD(&client->eta_list);
+	INIT_FLIST_HEAD(&client->cmd_list);
 
 	if (fio_server_parse_string(hostname, &client->hostname,
 					&client->is_sock, &client->port,
-					&client->addr.sin_addr))
+					&client->addr.sin_addr,
+					&client->addr6.sin6_addr,
+					&client->ipv6))
 		return -1;
 
 	client->fd = -1;
+	client->refs = 1;
 
 	__fio_client_add_cmd_option(client, "fio");
 
@@ -207,18 +266,31 @@ int fio_client_add(const char *hostname, void **cookie)
 
 static int fio_client_connect_ip(struct fio_client *client)
 {
-	int fd;
-
-	client->addr.sin_family = AF_INET;
-	client->addr.sin_port = htons(client->port);
+	struct sockaddr *addr;
+	socklen_t socklen;
+	int fd, domain;
+
+	if (client->ipv6) {
+		client->addr6.sin6_family = AF_INET6;
+		client->addr6.sin6_port = htons(client->port);
+		domain = AF_INET6;
+		addr = (struct sockaddr *) &client->addr6;
+		socklen = sizeof(client->addr6);
+	} else {
+		client->addr.sin_family = AF_INET;
+		client->addr.sin_port = htons(client->port);
+		domain = AF_INET;
+		addr = (struct sockaddr *) &client->addr;
+		socklen = sizeof(client->addr);
+	}
 
-	fd = socket(AF_INET, SOCK_STREAM, 0);
+	fd = socket(domain, SOCK_STREAM, 0);
 	if (fd < 0) {
 		log_err("fio: socket: %s\n", strerror(errno));
 		return -1;
 	}
 
-	if (connect(fd, (struct sockaddr *) &client->addr, sizeof(client->addr)) < 0) {
+	if (connect(fd, addr, socklen) < 0) {
 		log_err("fio: connect: %s\n", strerror(errno));
 		log_err("fio: failed to connect to %s:%u\n", client->hostname,
 								client->port);
@@ -232,7 +304,7 @@ static int fio_client_connect_ip(struct fio_client *client)
 static int fio_client_connect_sock(struct fio_client *client)
 {
 	struct sockaddr_un *addr = &client->addr_un;
-	fio_socklen_t len;
+	socklen_t len;
 	int fd;
 
 	memset(addr, 0, sizeof(*addr));
@@ -266,6 +338,8 @@ static int fio_client_connect(struct fio_client *client)
 	else
 		fd = fio_client_connect_ip(client);
 
+	dprint(FD_NET, "client: %s connected %d\n", client->hostname, fd);
+
 	if (fd < 0)
 		return 1;
 
@@ -285,7 +359,7 @@ void fio_clients_terminate(void)
 	flist_for_each(entry, &client_list) {
 		client = flist_entry(entry, struct fio_client, list);
 
-		fio_net_send_simple_cmd(client->fd, FIO_NET_CMD_QUIT, 0);
+		fio_net_send_simple_cmd(client->fd, FIO_NET_CMD_QUIT, 0, NULL);
 	}
 }
 
@@ -295,6 +369,11 @@ static void sig_int(int sig)
 	fio_clients_terminate();
 }
 
+static void sig_show_status(int sig)
+{
+	show_running_run_stats();
+}
+
 static void client_signal_handler(void)
 {
 	struct sigaction act;
@@ -308,14 +387,26 @@ static void client_signal_handler(void)
 	act.sa_handler = sig_int;
 	act.sa_flags = SA_RESTART;
 	sigaction(SIGTERM, &act, NULL);
+
+/* Windows uses SIGBREAK as a quit signal from other applications */
+#ifdef WIN32
+	memset(&act, 0, sizeof(act));
+	act.sa_handler = sig_int;
+	act.sa_flags = SA_RESTART;
+	sigaction(SIGBREAK, &act, NULL);
+#endif
+
+	memset(&act, 0, sizeof(act));
+	act.sa_handler = sig_show_status;
+	act.sa_flags = SA_RESTART;
+	sigaction(SIGUSR1, &act, NULL);
 }
 
 static void probe_client(struct fio_client *client)
 {
 	dprint(FD_NET, "client: send probe\n");
 
-	fio_net_send_simple_cmd(client->fd, FIO_NET_CMD_PROBE, 0);
-	handle_client(client);
+	fio_net_send_simple_cmd(client->fd, FIO_NET_CMD_PROBE, 0, &client->cmd_list);
 }
 
 static int send_client_cmd_line(struct fio_client *client)
@@ -371,6 +462,11 @@ int fio_clients_connect(void)
 	struct flist_head *entry, *tmp;
 	int ret;
 
+#ifdef WIN32
+	WSADATA wsd;
+	WSAStartup(MAKEWORD(2,2), &wsd);
+#endif
+
 	dprint(FD_NET, "client: connect all\n");
 
 	client_signal_handler();
@@ -443,6 +539,7 @@ static int fio_client_send_ini(struct fio_client *client, const char *filename)
 		return 1;
 	}
 
+	client->sent_job = 1;
 	ret = fio_net_send_cmd(client->fd, FIO_NET_CMD_JOB, buf, sb.st_size, 0);
 	free(buf);
 	close(fd);
@@ -457,8 +554,21 @@ int fio_clients_send_ini(const char *filename)
 	flist_for_each_safe(entry, tmp, &client_list) {
 		client = flist_entry(entry, struct fio_client, list);
 
-		if (fio_client_send_ini(client, filename))
+		if (client->nr_ini_file) {
+			int i;
+
+			for (i = 0; i < client->nr_ini_file; i++) {
+				const char *ini = client->ini_file[i];
+
+				if (fio_client_send_ini(client, ini)) {
+					remove_client(client);
+					break;
+				}
+			}
+		} else if (!filename || fio_client_send_ini(client, filename))
 			remove_client(client);
+
+		client->sent_job = 1;
 	}
 
 	return !nr_clients;
@@ -485,8 +595,9 @@ static void convert_ts(struct thread_stat *dst, struct thread_stat *src)
 	dst->groupid	= le32_to_cpu(src->groupid);
 	dst->pid	= le32_to_cpu(src->pid);
 	dst->members	= le32_to_cpu(src->members);
+	dst->unified_rw_rep	= le32_to_cpu(src->unified_rw_rep);
 
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
 		convert_io_stat(&dst->clat_stat[i], &src->clat_stat[i]);
 		convert_io_stat(&dst->slat_stat[i], &src->slat_stat[i]);
 		convert_io_stat(&dst->lat_stat[i], &src->lat_stat[i]);
@@ -518,11 +629,11 @@ static void convert_ts(struct thread_stat *dst, struct thread_stat *src)
 		dst->io_u_lat_m[i]	= le32_to_cpu(src->io_u_lat_m[i]);
 	}
 
-	for (i = 0; i < 2; i++)
+	for (i = 0; i < DDIR_RWDIR_CNT; i++)
 		for (j = 0; j < FIO_IO_U_PLAT_NR; j++)
 			dst->io_u_plat[i][j] = le32_to_cpu(src->io_u_plat[i][j]);
 
-	for (i = 0; i < 3; i++) {
+	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
 		dst->total_io_u[i]	= le64_to_cpu(src->total_io_u[i]);
 		dst->short_io_u[i]	= le64_to_cpu(src->short_io_u[i]);
 	}
@@ -530,7 +641,7 @@ static void convert_ts(struct thread_stat *dst, struct thread_stat *src)
 	dst->total_submit	= le64_to_cpu(src->total_submit);
 	dst->total_complete	= le64_to_cpu(src->total_complete);
 
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
 		dst->io_bytes[i]	= le64_to_cpu(src->io_bytes[i]);
 		dst->runtime[i]		= le64_to_cpu(src->runtime[i]);
 	}
@@ -540,13 +651,14 @@ static void convert_ts(struct thread_stat *dst, struct thread_stat *src)
 	dst->total_err_count	= le64_to_cpu(src->total_err_count);
 	dst->first_error	= le32_to_cpu(src->first_error);
 	dst->kb_base		= le32_to_cpu(src->kb_base);
+	dst->unit_base		= le32_to_cpu(src->unit_base);
 }
 
 static void convert_gs(struct group_run_stats *dst, struct group_run_stats *src)
 {
 	int i;
 
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
 		dst->max_run[i]		= le64_to_cpu(src->max_run[i]);
 		dst->min_run[i]		= le64_to_cpu(src->min_run[i]);
 		dst->max_bw[i]		= le64_to_cpu(src->max_bw[i]);
@@ -556,10 +668,12 @@ static void convert_gs(struct group_run_stats *dst, struct group_run_stats *src)
 	}
 
 	dst->kb_base	= le32_to_cpu(src->kb_base);
+	dst->unit_base	= le32_to_cpu(src->unit_base);
 	dst->groupid	= le32_to_cpu(src->groupid);
+	dst->unified_rw_rep	= le32_to_cpu(src->unified_rw_rep);
 }
 
-static void handle_ts(struct fio_net_cmd *cmd)
+static void handle_ts(struct fio_client *client, struct fio_net_cmd *cmd)
 {
 	struct cmd_ts_pdu *p = (struct cmd_ts_pdu *) cmd->payload;
 
@@ -567,6 +681,22 @@ static void handle_ts(struct fio_net_cmd *cmd)
 	convert_gs(&p->rs, &p->rs);
 
 	show_thread_status(&p->ts, &p->rs);
+	client->did_stat = 1;
+
+	if (!do_output_all_clients)
+		return;
+
+	sum_thread_stats(&client_ts, &p->ts, sum_stat_nr);
+	sum_group_stats(&client_gs, &p->rs);
+
+	client_ts.members++;
+	client_ts.groupid = p->ts.groupid;
+	client_ts.unified_rw_rep = p->ts.unified_rw_rep;
+
+	if (++sum_stat_nr == sum_stat_clients) {
+		strcpy(client_ts.name, "All clients");
+		show_thread_status(&client_ts, &client_gs);
+	}
 }
 
 static void handle_gs(struct fio_net_cmd *cmd)
@@ -577,6 +707,54 @@ static void handle_gs(struct fio_net_cmd *cmd)
 	show_group_stats(gs);
 }
 
+static void convert_agg(struct disk_util_agg *agg)
+{
+	int i;
+
+	for (i = 0; i < 2; i++) {
+		agg->ios[i]	= le32_to_cpu(agg->ios[i]);
+		agg->merges[i]	= le32_to_cpu(agg->merges[i]);
+		agg->sectors[i]	= le64_to_cpu(agg->sectors[i]);
+		agg->ticks[i]	= le32_to_cpu(agg->ticks[i]);
+	}
+
+	agg->io_ticks		= le32_to_cpu(agg->io_ticks);
+	agg->time_in_queue	= le32_to_cpu(agg->time_in_queue);
+	agg->slavecount		= le32_to_cpu(agg->slavecount);
+	agg->max_util.u.f	= fio_uint64_to_double(__le64_to_cpu(agg->max_util.u.i));
+}
+
+static void convert_dus(struct disk_util_stat *dus)
+{
+	int i;
+
+	for (i = 0; i < 2; i++) {
+		dus->ios[i]	= le32_to_cpu(dus->ios[i]);
+		dus->merges[i]	= le32_to_cpu(dus->merges[i]);
+		dus->sectors[i]	= le64_to_cpu(dus->sectors[i]);
+		dus->ticks[i]	= le32_to_cpu(dus->ticks[i]);
+	}
+
+	dus->io_ticks		= le32_to_cpu(dus->io_ticks);
+	dus->time_in_queue	= le32_to_cpu(dus->time_in_queue);
+	dus->msec		= le64_to_cpu(dus->msec);
+}
+
+static void handle_du(struct fio_client *client, struct fio_net_cmd *cmd)
+{
+	struct cmd_du_pdu *du = (struct cmd_du_pdu *) cmd->payload;
+
+	convert_dus(&du->dus);
+	convert_agg(&du->agg);
+
+	if (!client->disk_stats_shown) {
+		client->disk_stats_shown = 1;
+		log_info("\nDisk stats (read/write):\n");
+	}
+
+	print_disk_util(&du->dus, &du->agg, output_format == FIO_OUTPUT_TERSE);
+}
+
 static void convert_jobs_eta(struct jobs_eta *je)
 {
 	int i;
@@ -590,13 +768,15 @@ static void convert_jobs_eta(struct jobs_eta *je)
 	je->m_iops		= le32_to_cpu(je->m_iops);
 	je->t_iops		= le32_to_cpu(je->t_iops);
 
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
 		je->rate[i]	= le32_to_cpu(je->rate[i]);
 		je->iops[i]	= le32_to_cpu(je->iops[i]);
 	}
 
 	je->elapsed_sec		= le64_to_cpu(je->elapsed_sec);
 	je->eta_sec		= le64_to_cpu(je->eta_sec);
+	je->is_pow2		= le32_to_cpu(je->is_pow2);
+	je->unit_base   = le32_to_cpu(je->unit_base);
 }
 
 static void sum_jobs_eta(struct jobs_eta *dst, struct jobs_eta *je)
@@ -612,7 +792,7 @@ static void sum_jobs_eta(struct jobs_eta *dst, struct jobs_eta *je)
 	dst->m_iops		+= je->m_iops;
 	dst->t_iops		+= je->t_iops;
 
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
 		dst->rate[i]	+= je->rate[i];
 		dst->iops[i]	+= je->iops[i];
 	}
@@ -631,10 +811,34 @@ static void dec_jobs_eta(struct client_eta *eta)
 	}
 }
 
+static void remove_reply_cmd(struct fio_client *client, struct fio_net_cmd *cmd)
+{
+	struct fio_net_int_cmd *icmd = NULL;
+	struct flist_head *entry;
+
+	flist_for_each(entry, &client->cmd_list) {
+		icmd = flist_entry(entry, struct fio_net_int_cmd, list);
+
+		if (cmd->tag == (uintptr_t) icmd)
+			break;
+
+		icmd = NULL;
+	}
+
+	if (!icmd) {
+		log_err("fio: client: unable to find matching tag\n");
+		return;
+	}
+
+	flist_del(&icmd->list);
+	cmd->tag = icmd->saved_tag;
+	free(icmd);
+}
+
 static void handle_eta(struct fio_client *client, struct fio_net_cmd *cmd)
 {
 	struct jobs_eta *je = (struct jobs_eta *) cmd->payload;
-	struct client_eta *eta = (struct client_eta *) cmd->tag;
+	struct client_eta *eta = (struct client_eta *) (uintptr_t) cmd->tag;
 
 	dprint(FD_NET, "client: got eta tag %p, %d\n", eta, eta->pending);
 
@@ -652,6 +856,7 @@ static void handle_probe(struct fio_client *client, struct fio_net_cmd *cmd)
 {
 	struct cmd_probe_pdu *probe = (struct cmd_probe_pdu *) cmd->payload;
 	const char *os, *arch;
+	char bit[16];
 
 	os = fio_get_os_string(probe->os);
 	if (!os)
@@ -661,14 +866,41 @@ static void handle_probe(struct fio_client *client, struct fio_net_cmd *cmd)
 	if (!arch)
 		os = "unknown";
 
-	log_info("hostname=%s, be=%u, os=%s, arch=%s, fio=%u.%u.%u\n",
-		probe->hostname, probe->bigendian, os, arch, probe->fio_major,
-		probe->fio_minor, probe->fio_patch);
+	sprintf(bit, "%d-bit", probe->bpp * 8);
+
+	log_info("hostname=%s, be=%u, %s, os=%s, arch=%s, fio=%s\n",
+		probe->hostname, probe->bigendian, bit, os, arch,
+		probe->fio_version);
 
 	if (!client->name)
 		client->name = strdup((char *) probe->hostname);
 }
 
+static void handle_start(struct fio_client *client, struct fio_net_cmd *cmd)
+{
+	struct cmd_start_pdu *pdu = (struct cmd_start_pdu *) cmd->payload;
+
+	client->state = Client_started;
+	client->jobs = le32_to_cpu(pdu->jobs);
+	client->nr_stat = le32_to_cpu(pdu->stat_outputs);
+
+	if (sum_stat_clients > 1)
+		do_output_all_clients = 1;
+
+	sum_stat_clients += client->nr_stat;
+}
+
+static void handle_stop(struct fio_client *client, struct fio_net_cmd *cmd)
+{
+	struct cmd_end_pdu *pdu = (struct cmd_end_pdu *) cmd->payload;
+
+	client->state = Client_stopped;
+	client->error = le32_to_cpu(pdu->error);
+
+	if (client->error)
+		log_info("client <%s>: exited with error %d\n", client->hostname, client->error);
+}
+
 static int handle_client(struct fio_client *client)
 {
 	struct fio_net_cmd *cmd;
@@ -679,8 +911,8 @@ static int handle_client(struct fio_client *client)
 	if (!cmd)
 		return 0;
 
-	dprint(FD_NET, "client: got cmd op %d from %s\n",
-					cmd->opcode, client->hostname);
+	dprint(FD_NET, "client: got cmd op %s from %s\n",
+				fio_server_op(cmd->opcode), client->hostname);
 
 	switch (cmd->opcode) {
 	case FIO_NET_CMD_QUIT:
@@ -702,8 +934,12 @@ static int handle_client(struct fio_client *client)
 		free(cmd);
 		break;
 		}
+	case FIO_NET_CMD_DU:
+		handle_du(client, cmd);
+		free(cmd);
+		break;
 	case FIO_NET_CMD_TS:
-		handle_ts(cmd);
+		handle_ts(client, cmd);
 		free(cmd);
 		break;
 	case FIO_NET_CMD_GS:
@@ -711,23 +947,29 @@ static int handle_client(struct fio_client *client)
 		free(cmd);
 		break;
 	case FIO_NET_CMD_ETA:
+		remove_reply_cmd(client, cmd);
 		handle_eta(client, cmd);
 		free(cmd);
 		break;
 	case FIO_NET_CMD_PROBE:
+		remove_reply_cmd(client, cmd);
 		handle_probe(client, cmd);
 		free(cmd);
 		break;
+	case FIO_NET_CMD_RUN:
+		client->state = Client_running;
+		free(cmd);
+		break;
 	case FIO_NET_CMD_START:
-		client->state = Client_started;
+		handle_start(client, cmd);
 		free(cmd);
 		break;
 	case FIO_NET_CMD_STOP:
-		client->state = Client_stopped;
+		handle_stop(client, cmd);
 		free(cmd);
 		break;
 	default:
-		log_err("fio: unknown client op: %d\n", cmd->opcode);
+		log_err("fio: unknown client op: %s\n", fio_server_op(cmd->opcode));
 		free(cmd);
 		break;
 	}
@@ -755,12 +997,14 @@ static void request_client_etas(void)
 			skipped++;
 			continue;
 		}
+		if (client->state != Client_running)
+			continue;
 
 		assert(!client->eta_in_flight);
 		flist_add_tail(&client->eta_list, &eta_list);
 		client->eta_in_flight = eta;
 		fio_net_send_simple_cmd(client->fd, FIO_NET_CMD_SEND_ETA,
-						(uint64_t) eta);
+					(uintptr_t) eta, &client->cmd_list);
 	}
 
 	while (skipped--)
@@ -769,36 +1013,101 @@ static void request_client_etas(void)
 	dprint(FD_NET, "client: requested eta tag %p\n", eta);
 }
 
-int fio_handle_clients(void)
+static int client_check_cmd_timeout(struct fio_client *client,
+				    struct timeval *now)
+{
+	struct fio_net_int_cmd *cmd;
+	struct flist_head *entry, *tmp;
+	int ret = 0;
+
+	flist_for_each_safe(entry, tmp, &client->cmd_list) {
+		cmd = flist_entry(entry, struct fio_net_int_cmd, list);
+
+		if (mtime_since(&cmd->tv, now) < FIO_NET_CLIENT_TIMEOUT)
+			continue;
+
+		log_err("fio: client %s, timeout on cmd %s\n", client->hostname,
+						fio_server_op(cmd->cmd.opcode));
+		flist_del(&cmd->list);
+		free(cmd);
+		ret = 1;
+	}
+
+	return flist_empty(&client->cmd_list) && ret;
+}
+
+static int fio_client_timed_out(void)
 {
 	struct fio_client *client;
-	struct flist_head *entry;
+	struct flist_head *entry, *tmp;
+	struct timeval tv;
+	int ret = 0;
+
+	fio_gettime(&tv, NULL);
+
+	flist_for_each_safe(entry, tmp, &client_list) {
+		client = flist_entry(entry, struct fio_client, list);
+
+		if (flist_empty(&client->cmd_list))
+			continue;
+
+		if (!client_check_cmd_timeout(client, &tv))
+			continue;
+
+		log_err("fio: client %s timed out\n", client->hostname);
+		remove_client(client);
+		ret = 1;
+	}
+
+	return ret;
+}
+
+int fio_handle_clients(void)
+{
 	struct pollfd *pfds;
-	int i, ret = 0;
+	int i, ret = 0, retval = 0;
 
-	gettimeofday(&eta_tv, NULL);
+	fio_gettime(&eta_tv, NULL);
 
 	pfds = malloc(nr_clients * sizeof(struct pollfd));
 
+	init_thread_stat(&client_ts);
+	init_group_run_stat(&client_gs);
+
 	while (!exit_backend && nr_clients) {
+		struct flist_head *entry, *tmp;
+		struct fio_client *client;
+
 		i = 0;
-		flist_for_each(entry, &client_list) {
+		flist_for_each_safe(entry, tmp, &client_list) {
 			client = flist_entry(entry, struct fio_client, list);
 
+			if (!client->sent_job &&
+			    flist_empty(&client->cmd_list)) {
+				remove_client(client);
+				continue;
+			}
+
 			pfds[i].fd = client->fd;
 			pfds[i].events = POLLIN;
 			i++;
 		}
 
+		if (!nr_clients)
+			break;
+
 		assert(i == nr_clients);
 
 		do {
 			struct timeval tv;
 
-			gettimeofday(&tv, NULL);
+			fio_gettime(&tv, NULL);
 			if (mtime_since(&eta_tv, &tv) >= 900) {
 				request_client_etas();
 				memcpy(&eta_tv, &tv, sizeof(tv));
+
+				if (fio_client_timed_out())
+					break;
 			}
 
 			ret = poll(pfds, nr_clients, 100);
@@ -824,10 +1133,13 @@ int fio_handle_clients(void)
 				log_info("client: host=%s disconnected\n",
 						client->hostname);
 				remove_client(client);
-			}
+				retval = 1;
+			} else if (client->error)
+				retval = 1;
+			put_client(client);
 		}
 	}
 
 	free(pfds);
-	return 0;
+	return retval;
 }