1
1
openmpi/orte/util/session_dir.c

723 строки
22 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) 2004-2006 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 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.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*
*/
#include "orte_config.h"
#include "orte/constants.h"
#include <stdio.h>
#ifdef HAVE_PWD_H
#include <pwd.h>
#endif
#include <stddef.h>
#include <stdlib.h>
#include <string.h>
#ifdef HAVE_SYS_PARAM_H
#include <sys/param.h>
#endif /* HAVE_SYS_PARAM_H */
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif /* HAVE_SYS_TYPES_H */
#include <sys/stat.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif /* HAVE_UNISTD_H */
#include <errno.h>
#ifdef HAVE_DIRENT_H
#include <dirent.h>
#endif /* HAVE_DIRENT_H */
#ifdef HAVE_PWD_H
#include <pwd.h>
#endif /* HAVE_PWD_H */
#include "opal/util/argv.h"
#include "opal/util/output.h"
#include "opal/util/os_path.h"
#include "opal/util/os_dirpath.h"
#include "opal/util/basename.h"
#include "opal/util/opal_environ.h"
#include "orte/util/proc_info.h"
#include "orte/util/name_fns.h"
#include "orte/util/show_help.h"
#include "orte/mca/errmgr/errmgr.h"
#include "orte/runtime/runtime.h"
#include "orte/runtime/orte_globals.h"
#include "orte/util/session_dir.h"
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*******************************
* Local function Declarations
*******************************/
static int orte_create_dir(char *directory);
static bool orte_dir_check_file(const char *root, const char *path);
static bool orte_dir_check_file_output(const char *root, const char *path);
static char *orte_build_job_session_dir(char *top_dir,
orte_process_name_t *proc,
orte_jobid_t jobid);
#define OMPI_PRINTF_FIX_STRING(a) ((NULL == a) ? "(null)" : a)
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/****************************
* Funcationality
****************************/
/*
* Check and create the directory requested
*/
static int orte_create_dir(char *directory)
{
mode_t my_mode = S_IRWXU; /* I'm looking for full rights */
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
int ret;
/* Sanity check before creating the directory with the proper mode,
* Make sure it doesn't exist already */
if( ORTE_ERR_NOT_FOUND !=
(ret = opal_os_dirpath_access(directory, my_mode)) ) {
/* Failure because opal_os_dirpath_access() indicated that either:
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
* - The directory exists and we can access it (no need to create it again),
* return OPAL_SUCCESS, or
* - don't have access rights, return OPAL_ERROR
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
*/
if (ORTE_SUCCESS != ret) {
ORTE_ERROR_LOG(ret);
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
return(ret);
}
/* Get here if the directory doesn't exist, so create it */
if (ORTE_SUCCESS != (ret = opal_os_dirpath_create(directory, my_mode))) {
ORTE_ERROR_LOG(ret);
}
return ret;
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Construct the fullpath to the session directory
*/
int
orte_session_dir_get_name(char **fulldirpath,
char **return_prefix, /* This will come back as the valid tmp dir */
char **return_frontend,
char *hostid,
char *batchid,
orte_process_name_t *proc) {
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
char *hostname = NULL,
*batchname = NULL,
*sessions = NULL,
*user = NULL,
*prefix = NULL,
*frontend = NULL,
*jobfam = NULL,
*job = NULL,
*vpidstr = NULL;
bool prefix_provided = false;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
int exit_status = ORTE_SUCCESS;
size_t len;
int uid;
struct passwd *pwdent;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/* Ensure that system info is set */
orte_proc_info();
/* get the name of the user */
uid = getuid();
#ifdef HAVE_GETPWUID
pwdent = getpwuid(uid);
#else
pwdent = NULL;
#endif
if (NULL != pwdent) {
user = strdup(pwdent->pw_name);
} else {
if (0 > asprintf(&user, "%d", uid)) {
return ORTE_ERR_OUT_OF_RESOURCE;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* set the 'hostname'
*/
if( NULL != hostid) { /* User specified version */
hostname = strdup(hostid);
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
else { /* check if it is set elsewhere */
if( NULL != orte_process_info.nodename)
hostname = strdup(orte_process_info.nodename);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
else {
/* Couldn't find it, so fail */
ORTE_ERROR_LOG(ORTE_ERR_BAD_PARAM);
exit_status = ORTE_ERR_BAD_PARAM;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
goto cleanup;
}
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* set the 'batchid'
*/
if (NULL != batchid)
batchname = strdup(batchid);
else
batchname = strdup("0");
/*
* get the front part of the session directory
* Will look something like:
* openmpi-sessions-USERNAME@HOSTNAME_BATCHID
*/
if (NULL != orte_process_info.top_session_dir) {
frontend = strdup(orte_process_info.top_session_dir);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
else { /* If not set then construct it */
if (0 > asprintf(&frontend, "openmpi-sessions-%s@%s_%s", user, hostname, batchname)) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
exit_status = ORTE_ERR_OUT_OF_RESOURCE;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
goto cleanup;
}
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Construct the session directory
*/
/* If we were given a valid vpid then we can construct it fully into:
* openmpi-sessions-USERNAME@HOSTNAME_BATCHID/JOB-FAMILY/JOBID/VPID
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
*/
if( NULL != proc) {
if (ORTE_VPID_INVALID != proc->vpid) {
if (0 > asprintf(&jobfam, "%d", ORTE_JOB_FAMILY(proc->jobid))) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
exit_status = ORTE_ERR_OUT_OF_RESOURCE;
goto cleanup;
}
if (0 > asprintf(&job, "%d", ORTE_LOCAL_JOBID(proc->jobid))) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
exit_status = ORTE_ERR_OUT_OF_RESOURCE;
goto cleanup;
}
if (ORTE_SUCCESS != orte_util_convert_vpid_to_string(&vpidstr, proc->vpid)) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
exit_status = ORTE_ERR_OUT_OF_RESOURCE;
goto cleanup;
}
sessions = opal_os_path( false, frontend, jobfam, job, vpidstr, NULL );
if( NULL == sessions ) {
ORTE_ERROR_LOG(ORTE_ERROR);
exit_status = ORTE_ERROR;
goto cleanup;
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
/* If we were given a valid jobid then we can construct it partially into:
* openmpi-sessions-USERNAME@HOSTNAME_BATCHID/JOB-FAMILY/JOBID
*/
else if (ORTE_JOBID_INVALID != proc->jobid) {
if (0 > asprintf(&jobfam, "%d", ORTE_JOB_FAMILY(proc->jobid))) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
exit_status = ORTE_ERR_OUT_OF_RESOURCE;
goto cleanup;
}
if (0 > asprintf(&job, "%d", ORTE_LOCAL_JOBID(proc->jobid))) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
exit_status = ORTE_ERR_OUT_OF_RESOURCE;
goto cleanup;
}
sessions = opal_os_path( false, frontend, jobfam, job, NULL );
if( NULL == sessions ) {
ORTE_ERROR_LOG(ORTE_ERROR);
exit_status = ORTE_ERROR;
goto cleanup;
}
} /* if both are invalid */
else {
sessions = strdup(frontend); /* must dup this to avoid double-free later */
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
} /* If we were not given a proc at all, then we just set it to frontend
*/
else {
sessions = strdup(frontend); /* must dup this to avoid double-free later */
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* If the user specified an invalid prefix, or no prefix at all
* we need to keep looking
*/
if( NULL != fulldirpath && NULL != *fulldirpath) {
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
free(*fulldirpath);
*fulldirpath = NULL;
}
if( NULL != return_prefix && NULL != *return_prefix) { /* use the user specified one, if available */
prefix = strdup(*return_prefix);
prefix_provided = true;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
/* Try to find a proper alternative prefix */
else if (NULL != orte_process_info.tmpdir_base) { /* stored value */
prefix = strdup(orte_process_info.tmpdir_base);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
else { /* General Environment var */
prefix = strdup(opal_tmp_directory());
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
len = strlen(prefix);
/* check for a trailing path separator */
if (OPAL_PATH_SEP[0] == prefix[len-1]) {
prefix[len-1] = '\0';
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/* BEFORE doing anything else, check to see if this prefix is
* allowed by the system
*/
if (NULL != orte_prohibited_session_dirs) {
char **list;
int i, len;
/* break the string into tokens - it should be
* separated by ','
*/
list = opal_argv_split(orte_prohibited_session_dirs, ',');
len = opal_argv_count(list);
/* cycle through the list */
for (i=0; i < len; i++) {
/* check if prefix matches */
if (0 == strncmp(prefix, list[i], strlen(list[i]))) {
/* this is a prohibited location */
orte_show_help("help-orte-runtime.txt",
"orte:session:dir:prohibited",
true, prefix, orte_prohibited_session_dirs);
return ORTE_ERR_FATAL;
}
}
opal_argv_free(list); /* done with this */
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Construct the absolute final path, if requested
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
*/
if (NULL != fulldirpath) {
*fulldirpath = opal_os_path(false, prefix, sessions, NULL);
}
/*
* Return the frontend and prefix, if user requested we do so
*/
if (NULL != return_frontend) {
*return_frontend = strdup(frontend);
}
if (!prefix_provided && NULL != return_prefix) {
*return_prefix = strdup(prefix);
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
cleanup:
if(NULL != hostname)
free(hostname);
if(NULL != batchname)
free(batchname);
if(NULL != sessions)
free(sessions);
if(NULL != user)
free(user);
if (NULL != prefix) free(prefix);
if (NULL != frontend) free(frontend);
if (NULL != jobfam) free(jobfam);
if (NULL != job) free(job);
if (NULL != vpidstr) free(vpidstr);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
return exit_status;
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Construct the session directory and create it if necessary
*/
int orte_session_dir(bool create,
char *prefix, char *hostid,
char *batchid, orte_process_name_t *proc)
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
{
char *fulldirpath = NULL,
*frontend = NULL,
*sav = NULL;
int rc = ORTE_SUCCESS;
char *local_prefix = NULL;
/* use the specified prefix, if one was given */
if (NULL != prefix) {
local_prefix = strdup(prefix);
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Get the session directory full name
*/
if( ORTE_SUCCESS != ( rc = orte_session_dir_get_name(&fulldirpath,
&local_prefix,
&frontend,
hostid,
batchid, proc) ) ) {
if (ORTE_ERR_FATAL == rc) {
/* this indicates we should abort quietly */
rc = ORTE_ERR_SILENT;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
goto cleanup;
}
/* otherwise, bark a little first */
ORTE_ERROR_LOG(rc);
goto cleanup;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Now that we have the full path, go ahead and create it if necessary
*/
if( create ) {
if( ORTE_SUCCESS != (rc = orte_create_dir(fulldirpath) ) ) {
ORTE_ERROR_LOG(rc);
goto cleanup;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
}
/*
* if we are not creating, then just verify that the path is OK
*/
else {
if( ORTE_SUCCESS != (rc = opal_os_dirpath_access(fulldirpath, 0) )) {
/* it is okay for the path not to be found - don't error
* log that case, but do error log others
*/
if (ORTE_ERR_NOT_FOUND != rc) {
ORTE_ERROR_LOG(rc);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
goto cleanup;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
}
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* If we are creating the directory tree, then force overwrite of the
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
* global structure fields
*/
if (create) {
if (NULL != orte_process_info.tmpdir_base) {
free(orte_process_info.tmpdir_base);
orte_process_info.tmpdir_base = NULL;
}
if (NULL != orte_process_info.top_session_dir) {
free(orte_process_info.top_session_dir);
orte_process_info.top_session_dir = NULL;
}
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Update some of the global structures if they are empty
*/
if (NULL == orte_process_info.tmpdir_base) {
orte_process_info.tmpdir_base = strdup(local_prefix);
}
if (NULL == orte_process_info.top_session_dir &&
NULL != frontend) {
orte_process_info.top_session_dir = strdup(frontend);
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Set the process session directory
*/
if (ORTE_VPID_INVALID != proc->vpid) {
if (create) { /* overwrite if creating */
if (NULL != orte_process_info.proc_session_dir) {
free(orte_process_info.proc_session_dir);
orte_process_info.proc_session_dir = NULL;
}
}
if (NULL == orte_process_info.proc_session_dir) {
orte_process_info.proc_session_dir = strdup(fulldirpath);
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/* Strip off last part of directory structure */
sav = opal_dirname(fulldirpath);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
free(fulldirpath);
fulldirpath = sav;
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
sav = NULL;
}
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
/*
* Set the job session directory
*/
if (ORTE_JOBID_INVALID != proc->jobid) {
if (create) { /* overwrite if creating */
if (NULL != orte_process_info.job_session_dir) {
free(orte_process_info.job_session_dir);
orte_process_info.job_session_dir = NULL;
}
}
if (NULL == orte_process_info.job_session_dir) {
orte_process_info.job_session_dir = strdup(fulldirpath);
}
}
if (orte_debug_flag) {
opal_output(0, "procdir: %s",
OMPI_PRINTF_FIX_STRING(orte_process_info.proc_session_dir));
opal_output(0, "jobdir: %s",
OMPI_PRINTF_FIX_STRING(orte_process_info.job_session_dir));
opal_output(0, "top: %s",
OMPI_PRINTF_FIX_STRING(orte_process_info.top_session_dir));
opal_output(0, "tmp: %s",
OMPI_PRINTF_FIX_STRING(orte_process_info.tmpdir_base));
}
cleanup:
if (NULL != local_prefix) {
free(local_prefix);
}
if(NULL != fulldirpath) {
free(fulldirpath);
}
if(NULL != frontend) {
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
free(frontend);
}
return rc;
}
/*
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-29 01:03:31 +04:00
* A job has aborted - so force cleanup of the session directory
*/
int
orte_session_dir_cleanup(orte_jobid_t jobid)
{
int rc = ORTE_SUCCESS;
char *tmp;
char *job_session_dir=NULL;
if (!orte_create_session_dirs) {
/* didn't create them */
return ORTE_SUCCESS;
}
/* need to setup the top_session_dir with the prefix */
tmp = opal_os_path(false,
orte_process_info.tmpdir_base,
orte_process_info.top_session_dir, NULL);
/* we can only blow away session directories for our job family */
job_session_dir = orte_build_job_session_dir(tmp, ORTE_PROC_MY_NAME, jobid);
if (NULL == job_session_dir) {
rc = ORTE_ERR_OUT_OF_RESOURCE;
goto CLEANUP;
}
Bring in the generalized xcast communication system along with the correspondingly revised orted launch. I will send a message out to developers explaining the basic changes. In brief: 1. generalize orte_rml.xcast to become a general broadcast-like messaging system. Messages can now be sent to any tag on the daemons or processes. Note that any message sent via xcast will be delivered to ALL processes in the specified job - you don't get to pick and choose. At a later date, we will introduce an augmented capability that will use the daemons as relays, but will allow you to send to a specified array of process names. 2. extended orte_rml.xcast so it supports more scalable message routing methodologies. At the moment, we support three: (a) direct, which sends the message directly to all recipients; (b) linear, which sends the message to the local daemon on each node, which then relays it to its own local procs; and (b) binomial, which sends the message via a binomial algo across all the daemons, each of which then relays to its own local procs. The crossover points between the algos are adjustable via MCA param, or you can simply demand that a specific algo be used. 3. orteds no longer exhibit two types of behavior: bootproxy or VM. Orteds now always behave like they are part of a virtual machine - they simply launch a job if mpirun tells them to do so. This is another step towards creating an "orteboot" functionality, but also provided a clean system for supporting message relaying. Note one major impact of this commit: multiple daemons on a node cannot be supported any longer! Only a single daemon/node is now allowed. This commit is known to break support for the following environments: POE, Xgrid, Xcpu, Windows. It has been tested on rsh, SLURM, and Bproc. Modifications for TM support have been made but could not be verified due to machine problems at LANL. Modifications for SGE have been made but could not be verified. The developers for the non-verified environments will be separately notified along with suggestions on how to fix the problems. This commit was SVN r15007.
2007-06-12 17:28:54 +04:00
if (ORTE_JOBID_WILDCARD != jobid) {
opal_os_dirpath_destroy(job_session_dir, true, orte_dir_check_file);
Bring in the generalized xcast communication system along with the correspondingly revised orted launch. I will send a message out to developers explaining the basic changes. In brief: 1. generalize orte_rml.xcast to become a general broadcast-like messaging system. Messages can now be sent to any tag on the daemons or processes. Note that any message sent via xcast will be delivered to ALL processes in the specified job - you don't get to pick and choose. At a later date, we will introduce an augmented capability that will use the daemons as relays, but will allow you to send to a specified array of process names. 2. extended orte_rml.xcast so it supports more scalable message routing methodologies. At the moment, we support three: (a) direct, which sends the message directly to all recipients; (b) linear, which sends the message to the local daemon on each node, which then relays it to its own local procs; and (b) binomial, which sends the message via a binomial algo across all the daemons, each of which then relays to its own local procs. The crossover points between the algos are adjustable via MCA param, or you can simply demand that a specific algo be used. 3. orteds no longer exhibit two types of behavior: bootproxy or VM. Orteds now always behave like they are part of a virtual machine - they simply launch a job if mpirun tells them to do so. This is another step towards creating an "orteboot" functionality, but also provided a clean system for supporting message relaying. Note one major impact of this commit: multiple daemons on a node cannot be supported any longer! Only a single daemon/node is now allowed. This commit is known to break support for the following environments: POE, Xgrid, Xcpu, Windows. It has been tested on rsh, SLURM, and Bproc. Modifications for TM support have been made but could not be verified due to machine problems at LANL. Modifications for SGE have been made but could not be verified. The developers for the non-verified environments will be separately notified along with suggestions on how to fix the problems. This commit was SVN r15007.
2007-06-12 17:28:54 +04:00
} else {
/* if we want the session_dir removed for ALL jobids, then
* just recursively blow the whole session away for our job family,
* saving only output files
Bring in the generalized xcast communication system along with the correspondingly revised orted launch. I will send a message out to developers explaining the basic changes. In brief: 1. generalize orte_rml.xcast to become a general broadcast-like messaging system. Messages can now be sent to any tag on the daemons or processes. Note that any message sent via xcast will be delivered to ALL processes in the specified job - you don't get to pick and choose. At a later date, we will introduce an augmented capability that will use the daemons as relays, but will allow you to send to a specified array of process names. 2. extended orte_rml.xcast so it supports more scalable message routing methodologies. At the moment, we support three: (a) direct, which sends the message directly to all recipients; (b) linear, which sends the message to the local daemon on each node, which then relays it to its own local procs; and (b) binomial, which sends the message via a binomial algo across all the daemons, each of which then relays to its own local procs. The crossover points between the algos are adjustable via MCA param, or you can simply demand that a specific algo be used. 3. orteds no longer exhibit two types of behavior: bootproxy or VM. Orteds now always behave like they are part of a virtual machine - they simply launch a job if mpirun tells them to do so. This is another step towards creating an "orteboot" functionality, but also provided a clean system for supporting message relaying. Note one major impact of this commit: multiple daemons on a node cannot be supported any longer! Only a single daemon/node is now allowed. This commit is known to break support for the following environments: POE, Xgrid, Xcpu, Windows. It has been tested on rsh, SLURM, and Bproc. Modifications for TM support have been made but could not be verified due to machine problems at LANL. Modifications for SGE have been made but could not be verified. The developers for the non-verified environments will be separately notified along with suggestions on how to fix the problems. This commit was SVN r15007.
2007-06-12 17:28:54 +04:00
*/
opal_os_dirpath_destroy(job_session_dir, true, orte_dir_check_file_output);
}
/* now attempt to eliminate the top level directory itself - this
* will fail if anything is present, but ensures we cleanup if
* we are the last one out
*/
opal_os_dirpath_destroy(tmp, false, orte_dir_check_file);
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
if (NULL != job_session_dir && opal_os_dirpath_is_empty(job_session_dir)) {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: found job session dir empty - deleting");
}
rmdir(job_session_dir);
} else {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: job session dir not empty - leaving");
}
goto CLEANUP;
}
if (opal_os_dirpath_is_empty(tmp)) {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: found top session dir empty - deleting");
}
rmdir(tmp);
} else {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: top session dir not empty - leaving");
}
}
CLEANUP:
free(tmp);
Bring in the generalized xcast communication system along with the correspondingly revised orted launch. I will send a message out to developers explaining the basic changes. In brief: 1. generalize orte_rml.xcast to become a general broadcast-like messaging system. Messages can now be sent to any tag on the daemons or processes. Note that any message sent via xcast will be delivered to ALL processes in the specified job - you don't get to pick and choose. At a later date, we will introduce an augmented capability that will use the daemons as relays, but will allow you to send to a specified array of process names. 2. extended orte_rml.xcast so it supports more scalable message routing methodologies. At the moment, we support three: (a) direct, which sends the message directly to all recipients; (b) linear, which sends the message to the local daemon on each node, which then relays it to its own local procs; and (b) binomial, which sends the message via a binomial algo across all the daemons, each of which then relays to its own local procs. The crossover points between the algos are adjustable via MCA param, or you can simply demand that a specific algo be used. 3. orteds no longer exhibit two types of behavior: bootproxy or VM. Orteds now always behave like they are part of a virtual machine - they simply launch a job if mpirun tells them to do so. This is another step towards creating an "orteboot" functionality, but also provided a clean system for supporting message relaying. Note one major impact of this commit: multiple daemons on a node cannot be supported any longer! Only a single daemon/node is now allowed. This commit is known to break support for the following environments: POE, Xgrid, Xcpu, Windows. It has been tested on rsh, SLURM, and Bproc. Modifications for TM support have been made but could not be verified due to machine problems at LANL. Modifications for SGE have been made but could not be verified. The developers for the non-verified environments will be separately notified along with suggestions on how to fix the problems. This commit was SVN r15007.
2007-06-12 17:28:54 +04:00
if (NULL != job_session_dir) free(job_session_dir);
return rc;
}
int
orte_session_dir_finalize(orte_process_name_t *proc)
{
int rc;
char *tmp;
char *job_session_dir, *vpid, *proc_session_dir;
if (!orte_create_session_dirs) {
/* didn't create them */
return ORTE_SUCCESS;
}
/* need to setup the top_session_dir with the prefix */
tmp = opal_os_path(false,
orte_process_info.tmpdir_base,
orte_process_info.top_session_dir, NULL);
/* define the proc and job session directories for this process */
if (ORTE_SUCCESS != (rc = orte_util_convert_vpid_to_string(&vpid, proc->vpid))) {
ORTE_ERROR_LOG(rc);
free(tmp);
return rc;
}
job_session_dir = orte_build_job_session_dir(tmp, proc, proc->jobid);
if( NULL == job_session_dir) {
free(tmp);
free(vpid);
return ORTE_ERR_OUT_OF_RESOURCE;
}
proc_session_dir = opal_os_path( false, job_session_dir, vpid, NULL );
if( NULL == proc_session_dir ) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
free(tmp);
free(vpid);
free(job_session_dir);
return ORTE_ERR_OUT_OF_RESOURCE;
}
opal_os_dirpath_destroy(proc_session_dir,
false, orte_dir_check_file);
opal_os_dirpath_destroy(job_session_dir,
false, orte_dir_check_file);
opal_os_dirpath_destroy(tmp,
false, orte_dir_check_file);
if (opal_os_dirpath_is_empty(proc_session_dir)) {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: found proc session dir empty - deleting");
}
rmdir(proc_session_dir);
} else {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: proc session dir not empty - leaving");
}
goto CLEANUP;
}
if (opal_os_dirpath_is_empty(job_session_dir)) {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: found job session dir empty - deleting");
}
rmdir(job_session_dir);
} else {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: job session dir not empty - leaving");
}
goto CLEANUP;
}
if (opal_os_dirpath_is_empty(tmp)) {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: found top session dir empty - deleting");
}
rmdir(tmp);
} else {
if (orte_debug_flag) {
opal_output(0, "sess_dir_finalize: top session dir not empty - leaving");
}
}
CLEANUP:
free(tmp);
free(vpid);
free(job_session_dir);
free(proc_session_dir);
return ORTE_SUCCESS;
}
static bool
orte_dir_check_file(const char *root, const char *path)
{
/*
* Keep:
* - files starting with "output-"
* - files that indicate abort
*/
if( (0 == strncmp(path, "output-", strlen("output-"))) ||
(0 == strcmp(path, "abort"))) {
return false;
}
return true;
}
static bool
orte_dir_check_file_output(const char *root, const char *path)
{
/*
* Keep:
* - files starting with "output-"
*/
if( 0 == strncmp(path, "output-", strlen("output-"))) {
return false;
}
return true;
}
static char *orte_build_job_session_dir(char *top_dir,
orte_process_name_t *proc,
orte_jobid_t jobid)
{
char *jobfam = NULL;
char *job_session_dir;
if (0 > asprintf(&jobfam, "%d", ORTE_JOB_FAMILY(proc->jobid))) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
return NULL;
}
if (ORTE_JOBID_WILDCARD != jobid) {
char *job = NULL;
if (0 > asprintf(&job, "%d", ORTE_LOCAL_JOBID(jobid))) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
job_session_dir = NULL;
goto out;
}
job_session_dir = opal_os_path(false, top_dir, jobfam, job, NULL);
free(job);
if (NULL == job_session_dir) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
}
} else {
job_session_dir = opal_os_path(false, top_dir, jobfam, NULL);
if( NULL == job_session_dir) {
ORTE_ERROR_LOG(ORTE_ERR_OUT_OF_RESOURCE);
}
}
out:
free(jobfam);
return job_session_dir;
}