boinc/client/app_control.C

884 lines
24 KiB
C

// Berkeley Open Infrastructure for Network Computing
// http://boinc.berkeley.edu
// Copyright (C) 2005 University of California
//
// This is free software; you can redistribute it and/or
// modify it under the terms of the GNU Lesser General Public
// License as published by the Free Software Foundation;
// either version 2.1 of the License, or (at your option) any later version.
//
// This software 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 Lesser General Public License for more details.
//
// To view the GNU Lesser General Public License visit
// http://www.gnu.org/copyleft/lesser.html
// or write to the Free Software Foundation, Inc.,
// 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
// monitoring and process control of running apps
#include "cpp.h"
#ifdef _WIN32
#include "boinc_win.h"
#else
#include <unistd.h>
#include <csignal>
#if HAVE_SYS_IPC_H
#include <sys/ipc.h>
#endif
#if HAVE_SYS_RESOURCE_H
#include <sys/resource.h>
#endif
#if HAVE_SYS_SIGNAL_H
#include <sys/signal.h>
#endif
#if HAVE_SYS_WAIT_H
#include <sys/wait.h>
#endif
#endif
using std::vector;
#include "filesys.h"
#include "error_numbers.h"
#include "util.h"
#include "parse.h"
#include "shmem.h"
#include "client_msgs.h"
#include "client_state.h"
#include "file_names.h"
#include "app.h"
bool ACTIVE_TASK::process_exists() {
switch (task_state) {
case PROCESS_EXECUTING:
case PROCESS_SUSPENDED:
case PROCESS_ABORT_PENDING:
return true;
}
return false;
}
// Send a quit message.
//
int ACTIVE_TASK::request_exit() {
if (!app_client_shm.shm) return 1;
process_control_queue.msg_queue_send(
"<quit/>",
app_client_shm.shm->process_control_request
);
return 0;
}
// send a kill signal.
// This is not caught by the process
//
int ACTIVE_TASK::kill_task() {
#ifdef _WIN32
return !TerminateProcess(pid_handle, -1);
#else
return kill(pid, SIGKILL);
#endif
}
#if !defined(HAVE_WAIT4) && defined(HAVE_WAIT3)
#include <map>
struct proc_info_t {
int status;
rusage r;
proc_info_t() {};
proc_info_t(int s, const rusage &ru);
};
proc_info_t::proc_info_t(int s, const rusage &ru) : status(s), r(ru) {}
pid_t wait4(pid_t pid, int *statusp, int options, struct rusage *rusagep) {
static std::map<pid_t,proc_info_t> proc_info;
pid_t tmp_pid=0;
if (!pid) {
return wait3(statusp,options,rusagep);
} else {
if (proc_info.find(pid) == proc_info.end()) {
do {
tmp_pid=wait3(statusp,options,rusagep);
if ((tmp_pid>0) && (tmp_pid != pid)) {
proc_info[tmp_pid]=proc_info_t(*statusp,*rusagep);
if (!(options && WNOHANG)) {
tmp_pid=0;
}
} else {
return pid;
}
} while (!tmp_pid);
} else {
*statusp=proc_info[pid].status;
*rusagep=proc_info[pid].r;
proc_info.erase(pid);
return pid;
}
}
}
#endif
// We have sent a quit request to the process; see if it's exited.
// This is called when the core client exits,
// or when a project is detached or reset
//
bool ACTIVE_TASK::has_task_exited() {
bool exited = false;
if (!process_exists()) return true;
#ifdef _WIN32
unsigned long exit_code;
if (GetExitCodeProcess(pid_handle, &exit_code)) {
if (exit_code != STILL_ACTIVE) {
exited = true;
}
}
#else
int my_pid, stat;
struct rusage rs;
my_pid = wait4(pid, &stat, WNOHANG, &rs);
if (my_pid == pid) {
exited = true;
}
#endif
if (exited) {
task_state = PROCESS_EXITED;
}
return exited;
}
// preempt this task
// called from the CLIENT_STATE::schedule_cpus()
// if quit_task is true always do this by quitting (we're low on swap space)
//
int ACTIVE_TASK::preempt(bool quit_task) {
int retval;
if (quit_task) {
retval = request_exit();
pending_suspend_via_quit = true;
} else {
retval = suspend();
}
scheduler_state = CPU_SCHED_PREEMPTED;
msg_printf(result->project, MSG_INFO,
"Pausing result %s (%s)",
result->name, (quit_task ? "removed from memory" : "left in memory")
);
return 0;
}
static void limbo_message(ACTIVE_TASK& at) {
msg_printf(at.result->project, MSG_INFO,
"Result %s exited with zero status but no 'finished' file",
at.result->name
);
msg_printf(at.result->project, MSG_INFO,
"If this happens repeatedly you may need to reset the project."
);
}
// deal with a process that has exited, for whatever reason
// (including preemption)
//
#ifdef _WIN32
bool ACTIVE_TASK::handle_exited_app(unsigned long exit_code) {
get_app_status_msg();
get_trickle_up_msg();
result->final_cpu_time = checkpoint_cpu_time;
if (task_state == PROCESS_ABORT_PENDING) {
task_state = PROCESS_ABORTED;
} else {
task_state = PROCESS_EXITED;
if (exit_code) {
char szError[1024];
gstate.report_result_error(
*result,
"%s - exit code %d (0x%x)",
windows_format_error_string(exit_code, szError, sizeof(szError)),
exit_code, exit_code
);
} else {
if (pending_suspend_via_quit) {
pending_suspend_via_quit = false;
task_state = PROCESS_UNINITIALIZED;
close_process_handles();
return true;
}
if (!finish_file_present()) {
scheduler_state = CPU_SCHED_PREEMPTED;
task_state = PROCESS_UNINITIALIZED;
close_process_handles();
limbo_message(*this);
return true;
}
}
result->exit_status = exit_code;
}
if (app_client_shm.shm) {
detach_shmem(shm_handle, app_client_shm.shm);
app_client_shm.shm = NULL;
}
read_stderr_file();
clean_out_dir(slot_dir);
return true;
}
#else
bool ACTIVE_TASK::handle_exited_app(int stat) {
SCOPE_MSG_LOG scope_messages(log_messages, CLIENT_MSG_LOG::DEBUG_TASK);
get_app_status_msg();
get_trickle_up_msg();
result->final_cpu_time = checkpoint_cpu_time;
if (task_state == PROCESS_ABORT_PENDING) {
task_state = PROCESS_ABORTED;
} else {
if (WIFEXITED(stat)) {
task_state = PROCESS_EXITED;
result->exit_status = WEXITSTATUS(stat);
if (result->exit_status) {
gstate.report_result_error(
*result,
"process exited with code %d (0x%x)",
result->exit_status, result->exit_status
);
} else {
// check for cases where an app exits
// without it being done from core client's point of view;
// in these cases, don't clean out slot dir
//
if (pending_suspend_via_quit) {
pending_suspend_via_quit = false;
task_state = PROCESS_UNINITIALIZED;
// destroy shm, since restarting app will re-create it
//
detach_and_destroy_shmem();
return true;
}
if (!finish_file_present()) {
// The process looks like it exited normally
// but there's no "finish file".
// Assume it was externally killed,
// and arrange for it to get restarted.
//
scheduler_state = CPU_SCHED_PREEMPTED;
task_state = PROCESS_UNINITIALIZED;
detach_and_destroy_shmem();
limbo_message(*this);
return true;
}
}
scope_messages.printf(
"ACTIVE_TASK::handle_exited_app(): process exited: status %d\n",
result->exit_status
);
} else if (WIFSIGNALED(stat)) {
int got_signal = WTERMSIG(stat);
// if the process was externally killed, allow it to restart.
//
switch (got_signal) {
case SIGHUP:
case SIGINT:
case SIGQUIT:
case SIGKILL:
case SIGTERM:
case SIGSTOP:
scheduler_state = CPU_SCHED_PREEMPTED;
task_state = PROCESS_UNINITIALIZED;
limbo_message(*this);
return true;
}
result->exit_status = stat;
task_state = PROCESS_WAS_SIGNALED;
signal = got_signal;
gstate.report_result_error(
*result, "process got signal %d", signal
);
scope_messages.printf(
"ACTIVE_TASK::handle_exited_app(): process got signal %d\n",
signal
);
} else {
task_state = PROCESS_EXIT_UNKNOWN;
result->state = PROCESS_EXIT_UNKNOWN;
}
}
read_stderr_file();
clean_out_dir(slot_dir);
return true;
}
#endif
bool ACTIVE_TASK::finish_file_present() {
char path[256];
sprintf(path, "%s%s%s", slot_dir, PATH_SEPARATOR, BOINC_FINISH_CALLED_FILE);
return boinc_file_exists(path);
}
void ACTIVE_TASK_SET::send_trickle_downs() {
unsigned int i;
ACTIVE_TASK* atp;
bool sent;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (!atp->process_exists()) continue;
if (atp->have_trickle_down) {
if (!atp->app_client_shm.shm) continue;
sent = atp->app_client_shm.shm->trickle_down.send_msg("<have_trickle_down/>\n");
if (sent) atp->have_trickle_down = false;
}
}
}
void ACTIVE_TASK_SET::send_heartbeats() {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (!atp->process_exists()) continue;
if (!atp->app_client_shm.shm) continue;
atp->app_client_shm.shm->heartbeat.send_msg("<heartbeat/>\n");
}
}
void ACTIVE_TASK_SET::process_control_poll() {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (!atp->process_exists()) continue;
if (!atp->app_client_shm.shm) continue;
atp->process_control_queue.msg_queue_poll(
atp->app_client_shm.shm->process_control_request
);
}
}
// See if any processes have exited
//
bool ACTIVE_TASK_SET::check_app_exited() {
ACTIVE_TASK* atp;
bool found = false;
SCOPE_MSG_LOG scope_messages(log_messages, CLIENT_MSG_LOG::DEBUG_TASK);
#ifdef _WIN32
unsigned long exit_code;
unsigned int i;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (!atp->process_exists()) continue;
if (GetExitCodeProcess(atp->pid_handle, &exit_code)) {
if (exit_code != STILL_ACTIVE) {
scope_messages.printf("ACTIVE_TASK_SET::check_app_exited(): Process exited with code %d\n", exit_code);
found = true;
atp->handle_exited_app(exit_code);
}
}
}
#else
int pid;
int stat;
struct rusage rs;
if ((pid = wait4(0, &stat, WNOHANG, &rs)) > 0) {
scope_messages.printf("ACTIVE_TASK_SET::check_app_exited(): process %d is done\n", pid);
atp = lookup_pid(pid);
if (!atp) {
msg_printf(NULL, MSG_ERROR, "ACTIVE_TASK_SET::check_app_exited(): pid %d not found\n", pid);
return false;
}
atp->handle_exited_app(stat);
found = true;
}
#endif
if (found) gstate.must_schedule_cpus = true;
return found;
}
// if an app has exceeded its maximum CPU time, abort it
//
bool ACTIVE_TASK::check_max_cpu_exceeded() {
if (current_cpu_time > max_cpu_time) {
msg_printf(result->project, MSG_INFO,
"Aborting result %s: exceeded CPU time limit %f\n",
result->name, max_cpu_time
);
abort_task("Maximum CPU time exceeded");
return true;
}
return false;
}
// if an app has exceeded its maximum disk usage, abort it
//
bool ACTIVE_TASK::check_max_disk_exceeded() {
double disk_usage;
int retval;
// don't do disk check too often
//
retval = current_disk_usage(disk_usage);
if (retval) {
msg_printf(0, MSG_ERROR, "Can't get application disk usage: %d", retval);
} else {
if (disk_usage > max_disk_usage) {
msg_printf(
result->project, MSG_INFO,
"Aborting result %s: exceeded disk limit: %f > %f\n",
result->name, disk_usage, max_disk_usage
);
abort_task("Maximum disk usage exceeded");
return true;
}
}
return false;
}
#if 0
// if an app has exceeded its maximum allowed memory, abort it
//
bool ACTIVE_TASK::check_max_mem_exceeded() {
// TODO: calculate working set size elsewhere
if (working_set_size > max_mem_usage || working_set_size/1048576 > gstate.global_prefs.max_memory_mbytes) {
msg_printf(
result->project, MSG_INFO,
"Aborting result %s: exceeded memory limit %f\n",
result->name,
min(max_mem_usage, gstate.global_prefs.max_memory_mbytes*1048576)
);
abort_task("Maximum memory usage exceeded");
return true;
}
return false;
}
#endif
bool ACTIVE_TASK::check_max_mem_exceeded() {
if (max_mem_usage != 0 && rss_bytes > max_mem_usage) {
msg_printf(
result->project, MSG_INFO,
"result %s: memory usage %f exceeds limit %f\n",
result->name,
rss_bytes,
max_mem_usage
);
//abort_task("Maximum memory usage exceeded");
return true;
}
return false;
}
bool ACTIVE_TASK_SET::vm_limit_exceeded(double vm_limit) {
unsigned int i;
ACTIVE_TASK *atp;
double total_vm_usage = 0;
for (i=0; i<active_tasks.size(); ++i) {
atp = active_tasks[i];
if (!atp->process_exists()) continue;
total_vm_usage += atp->vm_bytes;
}
return (total_vm_usage > vm_limit);
}
// Check if any of the active tasks have exceeded their
// resource limits on disk, CPU time or memory
//
bool ACTIVE_TASK_SET::check_rsc_limits_exceeded() {
unsigned int j;
ACTIVE_TASK *atp;
static double last_disk_check_time = 0;
double now = dtime();
for (j=0;j<active_tasks.size();j++) {
atp = active_tasks[j];
if (atp->task_state != PROCESS_EXECUTING) continue;
if (atp->check_max_cpu_exceeded()) return true;
else if (atp->check_max_mem_exceeded()) return true;
else if (now>last_disk_check_time + gstate.global_prefs.disk_interval) {
last_disk_check_time = now;
if (atp->check_max_disk_exceeded()) return true;
}
}
return false;
}
// If process is running, send it a kill signal
// This is done when app has exceeded CPU, disk, or mem limits
//
int ACTIVE_TASK::abort_task(char* msg) {
if (task_state == PROCESS_EXECUTING || task_state == PROCESS_SUSPENDED) {
task_state = PROCESS_ABORT_PENDING;
kill_task();
} else {
task_state = PROCESS_ABORTED;
}
gstate.report_result_error(*result, msg);
return 0;
}
// check for the stderr file, copy to result record
//
bool ACTIVE_TASK::read_stderr_file() {
char stderr_file[MAX_BLOB_LEN];
char path[256];
int n;
sprintf(path, "%s%s%s", slot_dir, PATH_SEPARATOR, STDERR_FILE);
if (boinc_file_exists(path)) {
FILE* f = fopen(path, "r");
n = fread(stderr_file, 1, sizeof(stderr_file)-1, f);
fclose(f);
if (n < 0) return false;
stderr_file[n] = '\0';
result->stderr_out += "<stderr_txt>\n";
result->stderr_out += stderr_file;
const char* stderr_txt_close = "\n</stderr_txt>\n";
// truncate stderr output to 64KB;
// it's unlikely that more than that will be useful
//
result->stderr_out = result->stderr_out.substr(
0, MAX_BLOB_LEN-1-strlen(stderr_txt_close)
);
result->stderr_out += stderr_txt_close;
return true;
}
return false;
}
// tell a running app to reread project preferences.
// This is called when project prefs change,
// or when a user file has finished downloading.
//
int ACTIVE_TASK::request_reread_prefs() {
int retval;
link_user_files();
retval = write_app_init_file();
if (retval) return retval;
if (!app_client_shm.shm) return 0;
app_client_shm.shm->graphics_request.send_msg(
xml_graphics_modes[MODE_REREAD_PREFS]
);
return 0;
}
// tell all running apps of a project to reread prefs
//
void ACTIVE_TASK_SET::request_reread_prefs(PROJECT* project) {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (atp->result->project != project) continue;
if (!atp->process_exists()) continue;
atp->request_reread_prefs();
}
}
// send quit signal to all tasks in the project
// (or all tasks, if proj==0).
// If they don't exit in 5 seconds,
// send them a kill signal and wait up to 5 more seconds to exit.
// This is called when the core client exits,
// or when a project is detached or reset
//
int ACTIVE_TASK_SET::exit_tasks(PROJECT* proj) {
request_tasks_exit(proj);
// Wait 5 seconds for them to exit normally; if they don't then kill them
//
if (wait_for_exit(5, proj)) {
kill_tasks(proj);
}
wait_for_exit(5, proj);
// get final checkpoint_cpu_times
//
get_msgs();
return 0;
}
// Wait up to wait_time seconds for processes to exit
// If proj is zero, wait for all processes, else that project's
// NOTE: it's bad form to sleep, but it would be complex to avoid it here
//
int ACTIVE_TASK_SET::wait_for_exit(double wait_time, PROJECT* proj) {
bool all_exited;
unsigned int i,n;
ACTIVE_TASK *atp;
for (i=0; i<10; i++) {
all_exited = true;
for (n=0; n<active_tasks.size(); n++) {
atp = active_tasks[n];
if (proj && atp->wup->project != proj) continue;
if (!atp->has_task_exited()) {
all_exited = false;
break;
}
}
if (all_exited) return 0;
boinc_sleep(wait_time/10.0);
}
return ERR_NOT_EXITED;
}
int ACTIVE_TASK_SET::abort_project(PROJECT* project) {
vector<ACTIVE_TASK*>::iterator task_iter;
ACTIVE_TASK* atp;
exit_tasks(project);
task_iter = active_tasks.begin();
while (task_iter != active_tasks.end()) {
atp = *task_iter;
if (atp->result->project == project) {
task_iter = active_tasks.erase(task_iter);
delete atp;
} else {
task_iter++;
}
}
return 0;
}
// Find the ACTIVE_TASK in the current set with the matching PID
//
ACTIVE_TASK* ACTIVE_TASK_SET::lookup_pid(int pid) {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (atp->pid == pid) return atp;
}
return NULL;
}
// Find the ACTIVE_TASK in the current set with the matching result
//
ACTIVE_TASK* ACTIVE_TASK_SET::lookup_result(RESULT* result) {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (atp->result == result) {
return atp;
}
}
return NULL;
}
// suspend all currently running tasks
// called only from CLIENT_STATE::suspend_activities(),
// e.g. because on batteries, time of day, benchmarking, etc.
//
void ACTIVE_TASK_SET::suspend_all(bool leave_apps_in_memory) {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (atp->task_state != PROCESS_EXECUTING) continue;
if (leave_apps_in_memory) {
atp->suspend();
} else {
atp->request_exit();
atp->pending_suspend_via_quit = true;
}
}
}
// resume all currently running tasks
//
void ACTIVE_TASK_SET::unsuspend_all() {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (atp->scheduler_state != CPU_SCHED_SCHEDULED) continue;
if (atp->task_state == PROCESS_UNINITIALIZED) {
if (atp->start(false)) {
msg_printf(
atp->wup->project,
MSG_ERROR,
"ACTIVE_TASK_SET::unsuspend_all(): could not restart active_task"
);
}
} else if (atp->task_state == PROCESS_SUSPENDED) {
atp->unsuspend();
}
}
}
// Check to see if any tasks are running
// called if benchmarking and waiting for suspends to happen
//
bool ACTIVE_TASK_SET::is_task_executing() {
unsigned int i;
ACTIVE_TASK* atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (atp->task_state == PROCESS_EXECUTING) {
return true;
}
}
return false;
}
// Send quit signal to all app processes
// This is called when the core client exits,
// or when a project is detached or reset
//
void ACTIVE_TASK_SET::request_tasks_exit(PROJECT* proj) {
unsigned int i;
ACTIVE_TASK *atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (proj && atp->wup->project != proj) continue;
if (!atp->process_exists()) continue;
atp->request_exit();
}
}
// Send kill signal to all app processes
// Don't wait for them to exit
//
void ACTIVE_TASK_SET::kill_tasks(PROJECT* proj) {
unsigned int i;
ACTIVE_TASK *atp;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (proj && atp->wup->project != proj) continue;
if (!atp->process_exists()) continue;
atp->kill_task();
}
}
// suspend a task
//
int ACTIVE_TASK::suspend() {
if (!app_client_shm.shm) return 0;
process_control_queue.msg_queue_send(
"<suspend/>",
app_client_shm.shm->process_control_request
);
task_state = PROCESS_SUSPENDED;
return 0;
}
// resume a suspended task
//
int ACTIVE_TASK::unsuspend() {
if (!app_client_shm.shm) return 0;
process_control_queue.msg_queue_send(
"<resume/>",
app_client_shm.shm->process_control_request
);
task_state = PROCESS_EXECUTING;
return 0;
}
// See if the app has placed a new message in shared mem
// (with CPU done, frac done etc.)
// If so parse it and return true.
//
bool ACTIVE_TASK::get_app_status_msg() {
char msg_buf[MSG_CHANNEL_SIZE];
bool found = false;
if (!app_client_shm.shm) return false;
if (app_client_shm.shm->app_status.get_msg(msg_buf)) {
fraction_done = current_cpu_time = checkpoint_cpu_time = 0.0;
parse_double(msg_buf, "<fraction_done>", fraction_done);
parse_double(msg_buf, "<current_cpu_time>", current_cpu_time);
parse_double(msg_buf, "<checkpoint_cpu_time>", checkpoint_cpu_time);
parse_double(msg_buf, "<vm_bytes>", vm_bytes);
parse_double(msg_buf, "<rss_bytes>", rss_bytes);
found = true;
}
return found;
}
bool ACTIVE_TASK::get_trickle_up_msg() {
char msg_buf[MSG_CHANNEL_SIZE];
bool found = false;
int retval;
if (!app_client_shm.shm) return false;
if (app_client_shm.shm->trickle_up.get_msg(msg_buf)) {
if (match_tag(msg_buf, "<have_new_trickle_up/>")) {
retval = move_trickle_file();
if (!retval) {
wup->project->sched_rpc_pending = true;
}
}
found = true;
}
return found;
}
// check for msgs from active tasks.
// Return true if any of them has changed its checkpoint_cpu_time
// (since in that case we need to write state file)
//
bool ACTIVE_TASK_SET::get_msgs() {
unsigned int i;
ACTIVE_TASK *atp;
double old_time;
bool action = false;
for (i=0; i<active_tasks.size(); i++) {
atp = active_tasks[i];
if (!atp->process_exists()) continue;
old_time = atp->checkpoint_cpu_time;
if (atp->get_app_status_msg()) {
//atp->estimate_frac_rate_of_change(dtime());
if (old_time != atp->checkpoint_cpu_time) {
action = true;
}
}
atp->get_trickle_up_msg();
}
return action;
}
const char *BOINC_RCSID_10ca137461 = "$Id$";