1
1
openmpi/orte/mca/plm/process/plm_process_module.c

1007 строки
34 KiB
C
Исходник Обычный вид История

/*
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2008 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2006 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2006 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2007 Los Alamos National Security, LLC. All rights
* reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*
* These symbols are in a file by themselves to provide nice linker
* semantics. Since linkers generally pull in symbols by object
* files, keeping these symbols as the only symbols in this file
* prevents utility programs such as "ompi_info" from having to import
* entire components just to query their version and parameters.
*/
#include "orte_config.h"
#include "orte/constants.h"
#include <stdlib.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
#include <errno.h>
#include <string.h>
#ifdef HAVE_SYS_SELECT_H
#include <sys/select.h>
#endif
#ifdef HAVE_SYS_TIME_H
#include <sys/time.h>
#endif
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif
#ifdef HAVE_SYS_STAT_H
#include <sys/stat.h>
#endif
#ifdef HAVE_SYS_WAIT_H
#include <sys/wait.h>
#endif
#include <fcntl.h>
#include <signal.h>
#ifdef HAVE_PWD_H
#include <pwd.h>
#endif
#include "opal/mca/installdirs/installdirs.h"
#include "opal/mca/base/mca_base_param.h"
#include "opal/util/if.h"
#include "opal/util/os_path.h"
#include "opal/util/path.h"
#include "opal/event/event.h"
#include "opal/util/argv.h"
#include "opal/util/opal_environ.h"
#include "orte/util/show_help.h"
#include "opal/util/trace.h"
#include "opal/util/basename.h"
#include "opal/util/opal_environ.h"
#include "orte/util/session_dir.h"
#include "orte/util/name_fns.h"
#include "orte/runtime/orte_wait.h"
#include "opal/dss/dss.h"
#include "orte/mca/rml/rml.h"
#include "orte/mca/errmgr/errmgr.h"
#include "orte/mca/ras/ras_types.h"
#include "orte/mca/rmaps/rmaps.h"
#include "orte/mca/plm/plm.h"
#include "orte/mca/plm/base/base.h"
#include "orte/mca/plm/base/plm_private.h"
#include "orte/mca/plm/process/plm_process.h"
#define rindex(a,b) strrchr((a),(b)) //daniel
#if OMPI_HAVE_POSIX_THREADS && OMPI_THREADS_HAVE_DIFFERENT_PIDS && OMPI_ENABLE_PROGRESS_THREADS
static int orte_plm_process_launch_threaded(orte_jobid_t jobid);
#endif
/*
* Interface
*/
static int orte_plm_process_init(void);
static int orte_plm_process_set_hnp_name(void);
static int orte_plm_process_launch(orte_job_t*);
static int orte_plm_process_terminate_job(orte_jobid_t);
static int orte_plm_process_terminate_orteds(void);
static int orte_plm_process_signal_job(orte_jobid_t, int32_t);
orte_plm_base_module_t orte_plm_process_module = {
orte_plm_process_init,
orte_plm_base_set_hnp_name,
#if OMPI_HAVE_POSIX_THREADS && OMPI_THREADS_HAVE_DIFFERENT_PIDS && OMPI_ENABLE_PROGRESS_THREADS
orte_plm_process_launch_threaded,
#else
orte_plm_process_launch,
#endif
NULL,
orte_plm_process_terminate_job,
orte_plm_process_terminate_orteds,
orte_plm_process_signal_job,
orte_plm_process_finalize
};
static void set_handler_default(int sig);
enum {
ORTE_PLM_RSH_SHELL_BASH = 0,
ORTE_PLM_RSH_SHELL_ZSH,
ORTE_PLM_RSH_SHELL_TCSH,
ORTE_PLM_RSH_SHELL_CSH,
ORTE_PLM_RSH_SHELL_KSH,
ORTE_PLM_RSH_SHELL_SH,
ORTE_PLM_RSH_SHELL_UNKNOWN
};
typedef int orte_plm_process_shell;
static const char * orte_plm_process_shell_name[] = {
"bash",
"zsh",
"tcsh", /* tcsh has to be first otherwise strstr finds csh */
"csh",
"ksh",
"sh",
"unknown"
};
/* local global storage of timing variables */
static struct timeval joblaunchstart, joblaunchstop;
/* global storage of active jobid being launched */
static orte_jobid_t active_job = ORTE_JOBID_INVALID;
/**
* Init the module
*/
int orte_plm_process_init(void)
{
int rc;
if (ORTE_SUCCESS != (rc = orte_plm_base_comm_start())) {
ORTE_ERROR_LOG(rc);
}
return rc;
}
/**
* Check the Shell variable on the specified node
*/
static int orte_plm_process_probe(orte_node_t * node, orte_plm_process_shell * shell)
{
char ** argv;
int rc, nfds;
int fd[2];
pid_t pid;
/* HANDLE myPipeFd[2];
SECURITY_ATTRIBUTES securityAttr;
STARTUPINFO startupInfo;
PROCESS_INFORMATION processInfo;
*/
fd_set readset;
fd_set errset;
char outbuf[4096];
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: going to check SHELL variable on node %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
node->name));
*shell = ORTE_PLM_RSH_SHELL_UNKNOWN;
/*
* Build argv array
*/
pid = _spawnve( _P_DETACH, argv[0], argv, NULL); //daniel
#if 0
securityAttr.nLength = sizeof(SECURITY_ATTRIBUTES); // Size of struct
securityAttr.lpSecurityDescriptor = NULL; // Default descriptor
securityAttr.bInheritHandle = TRUE; // Inheritable
// Create the pipe
if (CreatePipe(&myPipeFd[0], &myPipeFd[1], &securityAttr, 0)) {
// Create duplicate of write end so that original
if (!DuplicateHandle(
GetCurrentProcess(),
myPipeFd[0], // Original handle
GetCurrentProcess(),
NULL, // don't create new handle
0,
FALSE, // Not inheritable
DUPLICATE_SAME_ACCESS)
) {
CloseHandle(myPipeFd[0]);
CloseHandle(myPipeFd[1]);
opal_output(0, "plm:process: DuplicateHandle failed with errno=%d\n", errno);
return ORTE_ERR_IN_ERRNO;
}
ZeroMemory( &startupInfo, sizeof(startupInfo) );
startupInfo.cb = sizeof(startupInfo);
ZeroMemory( &processInfo, sizeof(processInfo) );
// Now populate startup info for CreateProcess
GetStartupInfo(&startupInfo);
startupInfo.dwFlags = STARTF_USESTDHANDLES;
startupInfo.hStdInput = myPipeFd[0];
startupInfo.hStdOutput = GetStdHandle(STD_OUTPUT_HANDLE);
startupInfo.hStdError = GetStdHandle(STD_ERROR_HANDLE);
// Start the child process.
if( !CreateProcess( argv[0], //module name NULL,
NULL, //(LPSTR)(const char *) argv,
NULL, // Process handle not inheritable
NULL, // Thread handle not inheritable
TRUE, // Set handle inheritance to TRUE;
// each inheritable handle in the calling process is inherited by the new process
0, // No creation flags
NULL, // Use parent's environment block
NULL, // Use parent's starting directory
&startupInfo, // Pointer to STARTUPINFO structure
&processInfo ) // Pointer to PROCESS_INFORMATION structure
)
{
CloseHandle(myPipeFd[1]);
opal_output(0, "plm:process: CreateProcess failed with errno=%d\n", errno); //, GetLastError() );
return ORTE_ERR_IN_ERRNO;
}
}
#endif
/*
if ((pid = fork()) < 0) {
opal_output(0, "plm:process: fork failed with errno=%d\n", errno);
return ORTE_ERR_IN_ERRNO;
}
else if (pid == 0) { // child //processInfo.hProcess
if (dup2(fd[1], 1) < 0) {
opal_output(0, "plm:process: dup2 failed with errno=%d\n", errno);
return ORTE_ERR_IN_ERRNO;
}
execvp(argv[0], argv);
exit(errno);
}
if (close(fd[1])) {
opal_output(0, "plm:process: close failed with errno=%d\n", errno);
return ORTE_ERR_IN_ERRNO;
}
*/
/* Monitor stdout */
FD_ZERO(&readset);
nfds = fd[0]+1;
memset (outbuf, 0, sizeof (outbuf));
rc = ORTE_SUCCESS;;
while (ORTE_SUCCESS == rc) {
int err;
FD_SET (fd[0], &readset);
errset = readset;
err = select(nfds, &readset, NULL, &errset, NULL);
if (err == -1) {
if (errno == EINTR)
continue;
else {
rc = ORTE_ERR_IN_ERRNO;
break;
}
}
if (FD_ISSET(fd[0], &errset) != 0)
rc = ORTE_ERR_FATAL;
/* In case we have something valid to read on stdin */
if (FD_ISSET(fd[0], &readset) != 0) {
ssize_t ret = 1;
char temp[4096];
char * ptr = outbuf;
ssize_t outbufsize = sizeof(outbuf);
memset (temp, 0, sizeof(temp));
while (ret != 0) {
ret = read (fd[0], temp, 256);
if (ret < 0) {
if (errno == EINTR)
continue;
else {
rc = ORTE_ERR_IN_ERRNO;
break;
}
}
else {
if (outbufsize > 0) {
memcpy (ptr, temp, (ret > outbufsize) ? outbufsize : ret);
outbufsize -= ret;
ptr += ret;
if (outbufsize > 0)
*ptr = '\0';
}
}
}
/* After reading complete string (aka read returns 0), we just break */
break;
}
}
/* Search for the substring of known shell-names */
/* for (i = 0; i < (int)(sizeof (orte_plm_process_shell_name)/
sizeof(orte_plm_process_shell_name[0])); i++) {
char *sh_name = NULL;
sh_name = rindex(outbuf, '/');
if ( sh_name != NULL ) {
sh_name++; /* skip '/' */
/* We cannot use "echo -n $SHELL" because -n is not portable. Therefore
* we have to remove the "\n" */
/* if ( sh_name[strlen(sh_name)-1] == '\n' ) {
sh_name[strlen(sh_name)-1] = '\0';
}
if ( 0 == strcmp(sh_name, orte_plm_process_shell_name[i]) ) {
*shell = i;
break;
}
}
}
*/
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: node:%s has SHELL: %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
node->name, orte_plm_process_shell_name[*shell]));
return rc;
}
/**
* Fill the exec_path variable with the directory to the orted
*/
static int orte_plm_process_fill_exec_path( char ** exec_path )
{
struct stat buf;
asprintf(exec_path, "%s/orted", opal_install_dirs.bindir);
if (0 != stat(*exec_path, &buf)) {
char *path = getenv("PATH");
if (NULL == path) {
path = "PATH is empty!";
}
This commit represents a bunch of work on a Mercurial side branch. As such, the commit message back to the master SVN repository is fairly long. = ORTE Job-Level Output Messages = Add two new interfaces that should be used for all new code throughout the ORTE and OMPI layers (we already make the search-and-replace on the existing ORTE / OMPI layers): * orte_output(): (and corresponding friends ORTE_OUTPUT, orte_output_verbose, etc.) This function sends the output directly to the HNP for processing as part of a job-specific output channel. It supports all the same outputs as opal_output() (syslog, file, stdout, stderr), but for stdout/stderr, the output is sent to the HNP for processing and output. More on this below. * orte_show_help(): This function is a drop-in-replacement for opal_show_help(), with two differences in functionality: 1. the rendered text help message output is sent to the HNP for display (rather than outputting directly into the process' stderr stream) 1. the HNP detects duplicate help messages and does not display them (so that you don't see the same error message N times, once from each of your N MPI processes); instead, it counts "new" instances of the help message and displays a message every ~5 seconds when there are new ones ("I got X new copies of the help message...") opal_show_help and opal_output still exist, but they only output in the current process. The intent for the new orte_* functions is that they can apply job-level intelligence to the output. As such, we recommend that all new ORTE and OMPI code use the new orte_* functions, not thei opal_* functions. === New code === For ORTE and OMPI programmers, here's what you need to do differently in new code: * Do not include opal/util/show_help.h or opal/util/output.h. Instead, include orte/util/output.h (this one header file has declarations for both the orte_output() series of functions and orte_show_help()). * Effectively s/opal_output/orte_output/gi throughout your code. Note that orte_output_open() takes a slightly different argument list (as a way to pass data to the filtering stream -- see below), so you if explicitly call opal_output_open(), you'll need to slightly adapt to the new signature of orte_output_open(). * Literally s/opal_show_help/orte_show_help/. The function signature is identical. === Notes === * orte_output'ing to stream 0 will do similar to what opal_output'ing did, so leaving a hard-coded "0" as the first argument is safe. * For systems that do not use ORTE's RML or the HNP, the effect of orte_output_* and orte_show_help will be identical to their opal counterparts (the additional information passed to orte_output_open() will be lost!). Indeed, the orte_* functions simply become trivial wrappers to their opal_* counterparts. Note that we have not tested this; the code is simple but it is quite possible that we mucked something up. = Filter Framework = Messages sent view the new orte_* functions described above and messages output via the IOF on the HNP will now optionally be passed through a new "filter" framework before being output to stdout/stderr. The "filter" OPAL MCA framework is intended to allow preprocessing to messages before they are sent to their final destinations. The first component that was written in the filter framework was to create an XML stream, segregating all the messages into different XML tags, etc. This will allow 3rd party tools to read the stdout/stderr from the HNP and be able to know exactly what each text message is (e.g., a help message, another OMPI infrastructure message, stdout from the user process, stderr from the user process, etc.). Filtering is not active by default. Filter components must be specifically requested, such as: {{{ $ mpirun --mca filter xml ... }}} There can only be one filter component active. = New MCA Parameters = The new functionality described above introduces two new MCA parameters: * '''orte_base_help_aggregate''': Defaults to 1 (true), meaning that help messages will be aggregated, as described above. If set to 0, all help messages will be displayed, even if they are duplicates (i.e., the original behavior). * '''orte_base_show_output_recursions''': An MCA parameter to help debug one of the known issues, described below. It is likely that this MCA parameter will disappear before v1.3 final. = Known Issues = * The XML filter component is not complete. The current output from this component is preliminary and not real XML. A bit more work needs to be done to configure.m4 search for an appropriate XML library/link it in/use it at run time. * There are possible recursion loops in the orte_output() and orte_show_help() functions -- e.g., if RML send calls orte_output() or orte_show_help(). We have some ideas how to fix these, but figured that it was ok to commit before feature freeze with known issues. The code currently contains sub-optimal workarounds so that this will not be a problem, but it would be good to actually solve the problem rather than have hackish workarounds before v1.3 final. This commit was SVN r18434.
2008-05-14 00:00:55 +04:00
orte_show_help("help-plm-process.txt", "no-local-orted",
true, path, opal_install_dirs.bindir);
return ORTE_ERR_NOT_FOUND;
}
return ORTE_SUCCESS;
}
/**
* Callback on daemon exit.
*/
static void orte_plm_process_wait_daemon(pid_t pid, int status, void* cbdata)
{
unsigned long deltat;
if (! WIFEXITED(status) || ! WEXITSTATUS(status) == 0) {
/* tell the user something went wrong */
opal_output(0, "ERROR: A daemon failed to start as expected.");
opal_output(0, "ERROR: There may be more information available from");
opal_output(0, "ERROR: the remote shell (see above).");
if (WIFEXITED(status)) {
opal_output(0, "ERROR: The daemon exited unexpectedly with status %d.",
WEXITSTATUS(status));
} else if (WIFSIGNALED(status)) {
#ifdef WCOREDUMP
if (WCOREDUMP(status)) {
opal_output(0, "The daemon received a signal %d (with core).",
WTERMSIG(status));
} else {
opal_output(0, "The daemon received a signal %d.", WTERMSIG(status));
}
#else
opal_output(0, "The daemon received a signal %d.", WTERMSIG(status));
#endif /* WCOREDUMP */
} else {
opal_output(0, "No extra status information is available: %d.", status);
}
/* report that the daemon has failed so we break out of the daemon
* callback receive and can exit
*/
orte_plm_base_launch_failed(active_job, pid, status, ORTE_JOB_STATE_FAILED_TO_START);
} /* if abnormal exit */
/* release any waiting threads */
OPAL_THREAD_LOCK(&mca_plm_process_component.lock);
if (mca_plm_process_component.num_children-- >=
mca_plm_process_component.num_concurrent ||
mca_plm_process_component.num_children == 0) {
opal_condition_signal(&mca_plm_process_component.cond);
}
if (mca_plm_process_component.timing && mca_plm_process_component.num_children == 0) {
if (0 != gettimeofday(&joblaunchstop, NULL)) {
opal_output(0, "plm_process: could not obtain job launch stop time");
} else {
deltat = (joblaunchstop.tv_sec - joblaunchstart.tv_sec)*1000000 +
(joblaunchstop.tv_usec - joblaunchstart.tv_usec);
opal_output(0, "plm_process: total time to launch job is %lu usec", deltat);
}
}
OPAL_THREAD_UNLOCK(&mca_plm_process_component.lock);
}
/**
* Launch a daemon (bootproxy) on each node. The daemon will be responsible
* for launching the application.
*/
/* When working in this function, ALWAYS jump to "cleanup" if
* you encounter an error so that orterun will be woken up and
* the job can cleanly terminate
*/
int orte_plm_process_launch(orte_job_t *jdata)
{
orte_job_map_t *map = NULL;
int proc_vpid_index;
int local_exec_index;
char *vpid_string = NULL;
char *param;
char **argv = NULL;
char *prefix_dir;
int argc = 0;
int rc;
char *lib_base = NULL, *bin_base = NULL;
bool failed_launch = true;
orte_app_context_t **apps;
orte_node_t **nodes;
orte_std_cntr_t nnode;
orte_job_state_t job_state = ORTE_JOB_NEVER_LAUNCHED;
if (mca_plm_process_component.timing) {
if (0 != gettimeofday(&joblaunchstart, NULL)) {
opal_output(0, "plm_process: could not obtain start time");
joblaunchstart.tv_sec = 0;
joblaunchstart.tv_usec = 0;
}
}
/* create a jobid for this job */
if (ORTE_SUCCESS != (rc = orte_plm_base_create_jobid(&jdata->jobid))) {
ORTE_ERROR_LOG(rc);
return rc;
}
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: launching job %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_JOBID_PRINT(jdata->jobid)));
/* setup the job */
if (ORTE_SUCCESS != (rc = orte_plm_base_setup_job(jdata))) {
ORTE_ERROR_LOG(rc);
goto cleanup;
}
/* set the active jobid */
active_job = jdata->jobid;
/* Get the map for this job */
if (NULL == (map = orte_rmaps.get_job_map(active_job))) {
ORTE_ERROR_LOG(ORTE_ERR_NOT_FOUND);
rc = ORTE_ERR_NOT_FOUND;
goto cleanup;
}
apps = (orte_app_context_t**)jdata->apps->addr;
nodes = (orte_node_t**)map->nodes->addr;
if (0 == map->num_new_daemons) {
/* have all the daemons we need - launch app */
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: no new daemons to launch",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME)));
goto launch_apps;
}
if (orte_debug_daemons_flag &&
mca_plm_process_component.num_concurrent < map->num_new_daemons) {
/**
* If we are in '--debug-daemons' we keep the ssh connection
* alive for the span of the run. If we use this option
* AND we launch on more than "num_concurrent" machines
* then we will deadlock. No connections are terminated
* until the job is complete, no job is started
* since all the orteds are waiting for all the others
* to come online, and the others ore not launched because
* we are waiting on those that have started to terminate
* their ssh tunnels. :(
* As we cannot run in this situation, pretty print the error
* and return an error code.
*/
This commit represents a bunch of work on a Mercurial side branch. As such, the commit message back to the master SVN repository is fairly long. = ORTE Job-Level Output Messages = Add two new interfaces that should be used for all new code throughout the ORTE and OMPI layers (we already make the search-and-replace on the existing ORTE / OMPI layers): * orte_output(): (and corresponding friends ORTE_OUTPUT, orte_output_verbose, etc.) This function sends the output directly to the HNP for processing as part of a job-specific output channel. It supports all the same outputs as opal_output() (syslog, file, stdout, stderr), but for stdout/stderr, the output is sent to the HNP for processing and output. More on this below. * orte_show_help(): This function is a drop-in-replacement for opal_show_help(), with two differences in functionality: 1. the rendered text help message output is sent to the HNP for display (rather than outputting directly into the process' stderr stream) 1. the HNP detects duplicate help messages and does not display them (so that you don't see the same error message N times, once from each of your N MPI processes); instead, it counts "new" instances of the help message and displays a message every ~5 seconds when there are new ones ("I got X new copies of the help message...") opal_show_help and opal_output still exist, but they only output in the current process. The intent for the new orte_* functions is that they can apply job-level intelligence to the output. As such, we recommend that all new ORTE and OMPI code use the new orte_* functions, not thei opal_* functions. === New code === For ORTE and OMPI programmers, here's what you need to do differently in new code: * Do not include opal/util/show_help.h or opal/util/output.h. Instead, include orte/util/output.h (this one header file has declarations for both the orte_output() series of functions and orte_show_help()). * Effectively s/opal_output/orte_output/gi throughout your code. Note that orte_output_open() takes a slightly different argument list (as a way to pass data to the filtering stream -- see below), so you if explicitly call opal_output_open(), you'll need to slightly adapt to the new signature of orte_output_open(). * Literally s/opal_show_help/orte_show_help/. The function signature is identical. === Notes === * orte_output'ing to stream 0 will do similar to what opal_output'ing did, so leaving a hard-coded "0" as the first argument is safe. * For systems that do not use ORTE's RML or the HNP, the effect of orte_output_* and orte_show_help will be identical to their opal counterparts (the additional information passed to orte_output_open() will be lost!). Indeed, the orte_* functions simply become trivial wrappers to their opal_* counterparts. Note that we have not tested this; the code is simple but it is quite possible that we mucked something up. = Filter Framework = Messages sent view the new orte_* functions described above and messages output via the IOF on the HNP will now optionally be passed through a new "filter" framework before being output to stdout/stderr. The "filter" OPAL MCA framework is intended to allow preprocessing to messages before they are sent to their final destinations. The first component that was written in the filter framework was to create an XML stream, segregating all the messages into different XML tags, etc. This will allow 3rd party tools to read the stdout/stderr from the HNP and be able to know exactly what each text message is (e.g., a help message, another OMPI infrastructure message, stdout from the user process, stderr from the user process, etc.). Filtering is not active by default. Filter components must be specifically requested, such as: {{{ $ mpirun --mca filter xml ... }}} There can only be one filter component active. = New MCA Parameters = The new functionality described above introduces two new MCA parameters: * '''orte_base_help_aggregate''': Defaults to 1 (true), meaning that help messages will be aggregated, as described above. If set to 0, all help messages will be displayed, even if they are duplicates (i.e., the original behavior). * '''orte_base_show_output_recursions''': An MCA parameter to help debug one of the known issues, described below. It is likely that this MCA parameter will disappear before v1.3 final. = Known Issues = * The XML filter component is not complete. The current output from this component is preliminary and not real XML. A bit more work needs to be done to configure.m4 search for an appropriate XML library/link it in/use it at run time. * There are possible recursion loops in the orte_output() and orte_show_help() functions -- e.g., if RML send calls orte_output() or orte_show_help(). We have some ideas how to fix these, but figured that it was ok to commit before feature freeze with known issues. The code currently contains sub-optimal workarounds so that this will not be a problem, but it would be good to actually solve the problem rather than have hackish workarounds before v1.3 final. This commit was SVN r18434.
2008-05-14 00:00:55 +04:00
orte_show_help("help-plm-process.txt", "deadlock-params",
true, mca_plm_process_component.num_concurrent, map->num_new_daemons);
rc = ORTE_ERR_FATAL;
goto cleanup;
}
/*
* After a discussion between Ralph & Jeff, we concluded that we
* really are handling the prefix dir option incorrectly. It currently
* is associated with an app_context, yet it really refers to the
* location where OpenRTE/Open MPI is installed on a NODE. Fixing
* this right now would involve significant change to orterun as well
* as elsewhere, so we will intentionally leave this incorrect at this
* point. The error, however, is identical to that seen in all prior
* releases of OpenRTE/Open MPI, so our behavior is no worse than before.
*
* A note to fix this, along with ideas on how to do so, has been filed
* on the project's Trac system under "feature enhancement".
*
* For now, default to the prefix_dir provided in the first app_context.
* Since there always MUST be at least one app_context, we are safe in
* doing this.
*/
prefix_dir = apps[0]->prefix_dir;
/*
* Build argv array
*/
opal_argv_append(&argc, &argv, "<template>");
/* add the daemon command (as specified by user) */
local_exec_index = argc;
opal_argv_append(&argc, &argv, mca_plm_process_component.orted);
/*
* Add the basic arguments to the orted command line, including
* all debug options
*/
orte_plm_base_orted_append_basic_args(&argc, &argv,
"env",
&proc_vpid_index,
false);
if (0 < opal_output_get_verbosity(orte_plm_globals.output)) {
param = opal_argv_join(argv, ' ');
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: final template argv:\n\t%s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
(NULL == param) ? "NULL" : param));
if (NULL != param) free(param);
}
/* Figure out the basenames for the libdir and bindir. This
requires some explanation:
- Use OPAL_LIBDIR and OPAL_BINDIR instead of -D'ing some macros
in this directory's Makefile.am because it makes all the
dependencies work out correctly. These are defined in
opal/install_dirs.h.
- After a discussion on the devel-core mailing list, the
developers decided that we should use the local directory
basenames as the basis for the prefix on the remote note.
This does not handle a few notable cases (e.g., f the
libdir/bindir is not simply a subdir under the prefix, if the
libdir/bindir basename is not the same on the remote node as
it is here in the local node, etc.), but we decided that
--prefix was meant to handle "the common case". If you need
something more complex than this, a) edit your shell startup
files to set PATH/LD_LIBRARY_PATH properly on the remove
node, or b) use some new/to-be-defined options that
explicitly allow setting the bindir/libdir on the remote
node. We decided to implement these options (e.g.,
--remote-bindir and --remote-libdir) to orterun when it
actually becomes a problem for someone (vs. a hypothetical
situation).
Hence, for now, we simply take the basename of this install's
libdir and bindir and use it to append this install's prefix
and use that on the remote node.
*/
lib_base = opal_basename(opal_install_dirs.libdir);
bin_base = opal_basename(opal_install_dirs.bindir);
/* set the job state to indicate we attempted to launch */
job_state = ORTE_JOB_STATE_FAILED_TO_START;
/*
* Iterate through each of the nodes
*/
for(nnode=0; nnode < map->num_nodes; nnode++) {
pid_t pid;
char *exec_path = NULL;
char **exec_argv;
/* if this daemon already exists, don't launch it! */
if (nodes[nnode]->daemon_launched) {
continue;
}
/* if the node's daemon has not been defined, then we
* have an error!
*/
if (NULL == nodes[nnode]->daemon) {
ORTE_ERROR_LOG(ORTE_ERR_FATAL);
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process:launch daemon failed to be defined on node %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
nodes[nnode]->name));
return ORTE_ERR_FATAL;
}
if (0 < opal_output_get_verbosity(orte_plm_globals.output)) {
param = opal_argv_join(argv, ' ');
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: start daemon as:\n\t%s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
(NULL == param) ? "NULL" : param));
if (NULL != param) free(param);
}
{
char** env;
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: launching on node %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
nodes[nnode]->name));
exec_argv = &argv[local_exec_index];
/* If the user provide a prefix then first try to find the application there */
if( NULL != prefix_dir ) {
char* full_path[3];
full_path[0] = opal_os_path( false, prefix_dir, NULL );
full_path[1] = opal_os_path( false, prefix_dir, bin_base, NULL );
full_path[2] = NULL;
exec_path = opal_path_find(exec_argv[0], full_path, F_OK, NULL);
free(full_path[0]); free(full_path[1]);
}
if( NULL == exec_path ) {
/* find the application in the default PATH */
exec_path = opal_path_findv(exec_argv[0], F_OK, environ, NULL);
if( NULL == exec_path ) {
char* full_path[2];
full_path[0] = opal_os_path( false, opal_install_dirs.bindir, NULL );
full_path[1] = NULL;
exec_path = opal_path_find(exec_argv[0], full_path, F_OK, NULL);
free(full_path[0]);
if( NULL == exec_path ) {
rc = orte_plm_process_fill_exec_path (&exec_path);
if (ORTE_SUCCESS != rc) {
return rc;
}
}
}
}
/* If we have a prefix, then modify the PATH and
LD_LIBRARY_PATH environment variables. We're
already in the child process, so it's ok to modify
environ. */
if (NULL != prefix_dir) {
char *oldenv, *newenv;
/* Reset PATH */
newenv = opal_dirname(exec_path);
/*newenv = opal_os_path( false, prefix_dir, bin_base, NULL );*/
oldenv = getenv("PATH");
if (NULL != oldenv) {
char *temp;
asprintf(&temp, "%s;%s", newenv, oldenv );
free( newenv );
newenv = temp;
}
opal_setenv("PATH", newenv, true, &environ);
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: reset PATH: %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
newenv));
free(newenv);
/* Reset LD_LIBRARY_PATH */
newenv = opal_os_path( false, prefix_dir, lib_base, NULL );
oldenv = getenv("LD_LIBRARY_PATH");
if (NULL != oldenv) {
char* temp;
asprintf(&temp, "%s;%s", newenv, oldenv);
free(newenv);
newenv = temp;
}
opal_setenv("LD_LIBRARY_PATH", newenv, true, &environ);
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: reset LD_LIBRARY_PATH: %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
newenv));
free(newenv);
}
/* tell the daemon to setup its own process session/group */
opal_argv_append(&argc, &argv, "--set-sid");
exec_argv = &argv[local_exec_index];
#if 0
/* Finally, chdir($HOME) because we're making the
assumption that this is what will happen on
remote nodes (via submit). This allows a user
to specify a path that is relative to $HOME for
both the cwd and argv[0] and it will work on
all nodes -- including the local host.
Otherwise, it would work on remote nodes and
not the local node. If the user does not start
in $HOME on the remote nodes... well... let's
hope they start in $HOME. :-) */
var = opal_home_directory();
if (NULL != var) {
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: changing to directory %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
var));
/* Ignore errors -- what are we going to do?
(and we ignore errors on the remote nodes
in the fork plm, so this is consistent) */
chdir(var);
}
#endif
/* pass the vpid */
rc = orte_util_convert_vpid_to_string(&vpid_string, nodes[nnode]->daemon->name.vpid);
if (ORTE_SUCCESS != rc) {
opal_output(0, "plm:process: unable to get daemon vpid as string");
goto cleanup;
}
free(argv[proc_vpid_index]);
argv[proc_vpid_index] = strdup(vpid_string);
free(vpid_string);
/* Set signal handlers back to the default. Do this close
to the execve() because the event library may (and likely
will) reset them. If we don't do this, the event
library may have left some set that, at least on some
OS's, don't get reset via fork() or exec(). Hence, the
orted could be unkillable (for example). */
set_handler_default(SIGTERM);
set_handler_default(SIGINT);
//set_handler_default(SIGHUP);
//set_handler_default(SIGPIPE);
set_handler_default(SIGCHLD);
/* setup environment */
env = opal_argv_copy(orte_launch_environ);
/* exec the daemon */
if (0 < opal_output_get_verbosity(orte_plm_globals.output)) {
param = opal_argv_join(exec_argv, ' ');
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: executing:\n\t%s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
(NULL == param) ? "NULL" : param));
if (NULL != param) free(param);
}
pid = _spawnve( _P_NOWAIT, exec_path, exec_argv, env);
if (pid == -1) {
failed_launch = true;
rc = ORTE_ERROR;
goto cleanup;
}
/* indicate this daemon has been launched in case anyone is sitting on that trigger */
nodes[nnode]->daemon->state = ORTE_PROC_STATE_LAUNCHED;
opal_output(0, "plm:process: execv %s hopefully started (pid %d)\n", exec_path, pid);
OPAL_THREAD_LOCK(&mca_plm_process_component.lock);
/* This situation can lead to a deadlock if '--debug-daemons' is set.
* However, the deadlock condition is tested at the begining of this
* function, so we're quite confident it should not happens here.
*/
if (mca_plm_process_component.num_children++ >=
mca_plm_process_component.num_concurrent) {
opal_condition_wait(&mca_plm_process_component.cond, &mca_plm_process_component.lock);
}
OPAL_THREAD_UNLOCK(&mca_plm_process_component.lock);
/* setup callback on sigchild - wait until setup above is complete
* as the callback can occur in the call to orte_wait_cb
*/
orte_wait_cb(pid, orte_plm_process_wait_daemon, NULL);
/* if required - add delay to avoid problems w/ X11 authentication */
if (0 < opal_output_get_verbosity(orte_plm_globals.output)
&& mca_plm_process_component.delay) {
sleep(mca_plm_process_component.delay);
}
}
}
/* wait for daemons to callback */
if (ORTE_SUCCESS != (rc = orte_plm_base_daemon_callback(map->num_new_daemons))) {
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: launch of apps failed for job %s on error %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_JOBID_PRINT(active_job), ORTE_ERROR_NAME(rc)));
goto cleanup;
}
launch_apps:
if (ORTE_SUCCESS != (rc = orte_plm_base_launch_apps(active_job))) {
OPAL_OUTPUT_VERBOSE((1, orte_plm_globals.output,
"%s plm:process: launch of apps failed for job %s on error %s",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_JOBID_PRINT(active_job), ORTE_ERROR_NAME(rc)));
goto cleanup;
}
/* get here if launch went okay */
failed_launch = false;
cleanup:
if (NULL != lib_base) {
free(lib_base);
}
if (NULL != bin_base) {
free(bin_base);
}
if (NULL != argv) {
opal_argv_free(argv);
}
/* check for failed launch - if so, force terminate */
if( failed_launch ) {
orte_plm_base_launch_failed(jdata->jobid, -1, ORTE_ERROR_DEFAULT_EXIT_CODE, job_state);
}
return rc;
}
/**
* Terminate all processes for a given job
*/
int orte_plm_process_terminate_job(orte_jobid_t jobid)
{
int rc;
/* order them to kill their local procs for this job */
if (ORTE_SUCCESS != (rc = orte_plm_base_orted_kill_local_procs(jobid))) {
ORTE_ERROR_LOG(rc);
}
return rc;
}
/**
* Terminate the orteds for a given job
*/
int orte_plm_process_terminate_orteds(void)
{
int rc;
/* now tell them to die! */
if (ORTE_SUCCESS != (rc = orte_plm_base_orted_exit(ORTE_DAEMON_EXIT_WITH_REPLY_CMD))) {
ORTE_ERROR_LOG(rc);
}
return rc;
}
int orte_plm_process_signal_job(orte_jobid_t jobid, int32_t signal)
{
int rc;
/* order them to pass this signal to their local procs */
if (ORTE_SUCCESS != (rc = orte_plm_base_orted_signal_local_procs(jobid, signal))) {
ORTE_ERROR_LOG(rc);
}
return rc;
}
int orte_plm_process_finalize(void)
{
int rc;
/* cleanup any pending recvs */
if (ORTE_SUCCESS != (rc = orte_plm_base_comm_stop())) {
ORTE_ERROR_LOG(rc);
}
return rc;
}
/**
* Handle threading issues.
*/
#if OMPI_HAVE_POSIX_THREADS && OMPI_THREADS_HAVE_DIFFERENT_PIDS && OMPI_ENABLE_PROGRESS_THREADS
struct orte_plm_process_stack_t {
opal_condition_t cond;
opal_mutex_t mutex;
bool complete;
orte_jobid_t jobid;
int rc;
};
typedef struct orte_plm_process_stack_t orte_plm_process_stack_t;
static void orte_plm_process_stack_construct(orte_plm_process_stack_t* stack)
{
OBJ_CONSTRUCT(&stack->mutex, opal_mutex_t);
OBJ_CONSTRUCT(&stack->cond, opal_condition_t);
stack->rc = 0;
stack->complete = false;
}
static void orte_plm_process_stack_destruct(orte_plm_process_stack_t* stack)
{
OBJ_DESTRUCT(&stack->mutex);
OBJ_DESTRUCT(&stack->cond);
}
static OBJ_CLASS_INSTANCE(
orte_plm_process_stack_t,
opal_object_t,
orte_plm_process_stack_construct,
orte_plm_process_stack_destruct);
static void orte_plm_process_launch_cb(int fd, short event, void* args)
{
orte_plm_process_stack_t *stack = (orte_plm_process_stack_t*)args;
OPAL_THREAD_LOCK(&stack->mutex);
stack->rc = orte_plm_process_launch(stack->jobid);
stack->complete = true;
opal_condition_signal(&stack->cond);
OPAL_THREAD_UNLOCK(&stack->mutex);
}
static int orte_plm_process_launch_threaded(orte_jobid_t jobid)
{
struct timeval tv = { 0, 0 };
struct opal_event event;
struct orte_plm_process_stack_t stack;
OBJ_CONSTRUCT(&stack, orte_plm_process_stack_t);
stack.jobid = jobid;
if( opal_event_progress_thread() ) {
stack.rc = orte_plm_process_launch( jobid );
} else {
opal_evtimer_set(&event, orte_plm_process_launch_cb, &stack);
opal_evtimer_add(&event, &tv);
OPAL_THREAD_LOCK(&stack.mutex);
while (stack.complete == false) {
opal_condition_wait(&stack.cond, &stack.mutex);
}
OPAL_THREAD_UNLOCK(&stack.mutex);
}
OBJ_DESTRUCT(&stack);
return stack.rc;
}
#endif
static void set_handler_default(int sig)
{
OPAL_TRACE(1);
/*
struct sigaction act;
act.sa_handler = SIG_DFL;
act.sa_flags = 0;
sigemptyset(&act.sa_mask);
sigaction(sig, &act, (struct sigaction *)0);
*/
}