/*
* Copyright (C) 1994-2021 Altair Engineering, Inc.
* For more information, contact Altair at www.altair.com.
*
* This file is part of both the OpenPBS software ("OpenPBS")
* and the PBS Professional ("PBS Pro") software.
*
* Open Source License Information:
*
* OpenPBS is free software. You can redistribute it and/or modify it under
* the terms of the GNU Affero General Public License as published by the
* Free Software Foundation, either version 3 of the License, or (at your
* option) any later version.
*
* OpenPBS is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public
* License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see .
*
* Commercial License Information:
*
* PBS Pro is commercially licensed software that shares a common core with
* the OpenPBS software. For a copy of the commercial license terms and
* conditions, go to: (http://www.pbspro.com/agreement.html) or contact the
* Altair Legal Department.
*
* Altair's dual-license business model allows companies, individuals, and
* organizations to create proprietary derivative works of OpenPBS and
* distribute them - whether embedded or bundled with other software -
* under a commercial license agreement.
*
* Use of Altair's trademarks, including but not limited to "PBS™",
* "OpenPBS®", "PBS Professional®", and "PBS Pro™" and Altair's logos is
* subject to Altair's trademark licensing policies.
*/
#ifndef _PBS_JOB_H
#define _PBS_JOB_H
#ifdef __cplusplus
extern "C" {
#endif
#include "list_link.h"
#include "attribute.h"
#include "range.h"
#include "Long.h"
/*
* job.h - structure definations for job objects
*
* Include Files Required:
*
* "list_link.h"
* "attribute.h"
* "server_limits.h"
* "reservation.h"
*/
#ifndef _SERVER_LIMITS_H
#include "server_limits.h"
#endif
#include "work_task.h"
#ifdef PBS_MOM /* For the var_table used in env funcs */
/* struct var_table = used to hold environment variables for the job */
struct var_table {
char **v_envp;
int v_ensize;
int v_used;
};
#endif
/*
* Dependent Job Structures
*
* This set of structures are used by the server to track job
* dependency.
*
* The depend (parent) structure is used to record the type of
* dependency. It also heads the list of depend_job related via this type.
* For a type of "sycnto", the number of jobs expected, registered and
* ready are also recorded.
*/
struct depend {
pbs_list_link dp_link; /* link to next dependency, if any */
short dp_type; /* type of dependency (all) */
short dp_numexp; /* num jobs expected (on or syncct only) */
short dp_numreg; /* num jobs registered (syncct only) */
short dp_released; /* This job released to run (syncwith) */
short dp_numrun; /* num jobs supposed to run */
pbs_list_head dp_jobs; /* list of related jobs (all) */
};
/*
* The depend_job structure is used to record the name and location
* of each job which is involved with the dependency
*/
struct depend_job {
pbs_list_link dc_link;
short dc_state; /* released / ready to run (syncct) */
long dc_cost; /* cost of this child (syncct) */
char dc_child[PBS_MAXSVRJOBID + 1]; /* child (dependent) job */
char dc_svr[PBS_MAXSERVERNAME + 1]; /* server owning job */
};
/*
* Warning: the relation between the numbers assigned to after* and before*
* is critical.
*/
#define JOB_DEPEND_TYPE_AFTERSTART 0
#define JOB_DEPEND_TYPE_AFTEROK 1
#define JOB_DEPEND_TYPE_AFTERNOTOK 2
#define JOB_DEPEND_TYPE_AFTERANY 3
#define JOB_DEPEND_TYPE_BEFORESTART 4
#define JOB_DEPEND_TYPE_BEFOREOK 5
#define JOB_DEPEND_TYPE_BEFORENOTOK 6
#define JOB_DEPEND_TYPE_BEFOREANY 7
#define JOB_DEPEND_TYPE_ON 8
#define JOB_DEPEND_TYPE_RUNONE 9
#define JOB_DEPEND_NUMBER_TYPES 11
#define JOB_DEPEND_OP_REGISTER 1
#define JOB_DEPEND_OP_RELEASE 2
#define JOB_DEPEND_OP_READY 3
#define JOB_DEPEND_OP_DELETE 4
#define JOB_DEPEND_OP_UNREG 5
/*
* The badplace structure is used to keep track of destinations
* which have been tried by a route queue and given a "reject"
* status back, see svr_movejob.c.
*/
typedef struct badplace {
pbs_list_link bp_link;
char bp_dest[PBS_MAXROUTEDEST + 1];
} badplace;
/*
* The grpcache structure defined here is used by MOM to maintain the
* home directory, uid and gid of the user name under
* which the job is running.
* The information is keep here rather than make repeated hits on the
* password and group files.
*/
struct grpcache {
uid_t gc_uid; /* uid job will execute under */
gid_t gc_gid; /* gid job will execute under */
gid_t gc_rgid; /* login gid of user uid */
char gc_homedir[1]; /* more space allocated as part of this */
/* structure following here */
};
/*
* Job attributes/resources are maintained in one of two ways.
* Most of the attributes are maintained in a decoded or parsed form.
* This allows quick access to the attribute and resource values
* when making decisions about the job (scheduling, routing, ...),
*
* Any attribute or resource which is not recognized on this server
* are kept in an "attrlist", a linked list of the "external"
* form (attr_extern, see attribute.h). These are maintained because
* the job may be passed on to another server (route or qmove) that
* does recognize them.
* See the job structure entry ji_attrlist and the attrlist structure.
*/
/*
* The following job_atr enum provide an index into the array of
* decoded job attributes, for quick access.
* Most of the attributes here are "public", but some are Read Only,
* Private, or even Internal data items; maintained here because of
* their variable size.
*
* "JOB_ATR_LAST" must be the last value as its number is used to
* define the size of the array.
*/
enum job_atr {
#include "job_attr_enum.h"
#include "site_job_attr_enum.h"
JOB_ATR_UNKN, /* the special "unknown" type */
JOB_ATR_LAST /* This MUST be LAST */
};
/* the following enum defines the type of checkpoint to be done */
/* none, based on cputime or walltime. Used only by Mom */
enum PBS_Chkpt_By {
PBS_CHECKPOINT_NONE, /* no checkpoint */
PBS_CHECKPOINT_CPUT, /* checkpoint by cputime interval */
PBS_CHECKPOINT_WALLT /* checkpoint by walltime interval */
};
typedef struct string_and_number_t {
char *str;
int num;
} string_and_number_t;
typedef struct resc_limit { /* per node limits for Mom */
int rl_ncpus; /* number of cpus */
int rl_ssi; /* ssinodes (for irix cpusets */
long long rl_mem; /* working set size (real mem) */
long long rl_vmem; /* total mem space (virtual) */
int rl_naccels; /* number of accelerators */
long long rl_accel_mem; /* accelerator mem (real mem) */
pbs_list_head rl_other_res; /* list of all other resources found in execvnode and sched select*/
unsigned int rl_res_count; /* total count of resources */
char *chunkstr; /* chunk represented */
int chunkstr_sz; /* size of chunkstr */
char *chunkspec; /* the spec in select string representing the chunk */
string_and_number_t host_chunk[2]; /* chunks representing exec_host/exec_host2 */
} resc_limit_t;
/*
* The "definations" for the job attributes are in the following array,
* it is also indexed by the JOB_ATR_... enums.
*/
extern attribute_def job_attr_def[];
extern void *job_attr_idx;
#ifndef PBS_MOM
typedef enum histjob_type {
T_FIN_JOB, /* Job finished execution or terminated */
T_MOV_JOB, /* Job moved to different destination */
T_MOM_DOWN /* Non-rerunnable Job failed because of
MOM failure.*/
} histjob_type;
#endif /* SERVER only! */
#ifdef PBS_MOM
#include "tm_.h"
/*
* host_vlist - an array of these is hung off of the hnodent for this host,
* i.e. if the hnodent index equals that in ji_nodeid;
* The array contains one entry for each vnode allocated for this job
* from this host.
*
* WARNING: This array only exists for cpuset machines !
*
* The mom hooks code (src/resmom/mom_hook_func.c:run_hook())
* depends on this structure. If newer resource members
* are added (besides hv_ncpus, hv_mem), then please update
* mom hooks code, as well as the appropriate RFE.
*
*/
typedef struct host_vlist {
char hv_vname[PBS_MAXNODENAME + 1]; /* vnode name */
int hv_ncpus; /* ncpus assigned */
size_t hv_mem; /* mem assigned */
} host_vlist_t;
/*
** Track nodes with an array of structures which each
** point to a list of events
*/
typedef struct hnodent {
tm_host_id hn_node; /* host (node) identifier (index) */
char *hn_host; /* hostname of node */
int hn_port; /* port of Mom */
int hn_stream; /* stream to MOM on node */
time_t hn_eof_ts; /* timestamp of when the stream went down */
int hn_sister; /* save error for KILL_JOB event */
int hn_nprocs; /* num procs allocated to this node */
int hn_vlnum; /* num entries in vlist */
host_vlist_t *hn_vlist; /* list of vnodes allocated */
resc_limit_t hn_nrlimit; /* resc limits per node */
void *hn_setup; /* save any setup info here */
pbs_list_head hn_events; /* pointer to list of events */
} hnodent;
typedef struct vmpiprocs {
tm_node_id vn_node; /* user's vnode identifier */
hnodent *vn_host; /* parent (host) nodeent entry */
char *vn_hname; /* host name for MPI, if null value */
/* use vn_host->hn_host */
/* host name for MPI */
char *vn_vname; /* vnode name */
int vn_cpus; /* number of cpus allocated to proc */
int vn_mpiprocs; /* number for mpiprocs */
int vn_threads; /* number for OMP_NUM_THREADS */
long long vn_mem; /* working set size (real mem) */
long long vn_vmem; /* total mem space (virtual) */
int vn_naccels; /* number of accelerators */
int vn_need_accel; /* should we reserve the accelerators */
char *vn_accel_model; /* model of the desired accelerator */
long long vn_accel_mem; /* amt of accelerator memory wanted */
} vmpiprocs;
/* the following enum defines if a node resource is to be reported by Mom */
enum PBS_NodeRes_Status {
PBS_NODERES_ACTIVE, /* resource reported from a non-released node */
PBS_NODERES_DELETE /* resource reported from a released node */
};
/*
* Mother Superior gets to hold an array of information from each
* of the other nodes for resource usage.
*/
typedef struct noderes {
char *nodehost; /* corresponding node name */
long nr_cput; /* cpu time */
long nr_mem; /* memory */
long nr_cpupercent; /* cpu percent */
attribute nr_used; /* node resources used */
enum PBS_NodeRes_Status nr_status;
} noderes;
/* State for a sister */
#define SISTER_OKAY 0
#define SISTER_KILLDONE 1000
#define SISTER_BADPOLL 1001
#define SISTER_EOF 1099
/* job flags for ji_flags (mom only) */
#define MOM_CHKPT_ACTIVE 0x0001 /* checkpoint in progress */
#define MOM_CHKPT_POST 0x0002 /* post checkpoint call returned */
#define MOM_SISTER_ERR 0x0004 /* a sisterhood operation failed */
#define MOM_NO_PROC 0x0008 /* no procs found for job */
#define MOM_RESTART_ACTIVE 0x0010 /* restart in progress */
#define PBS_MAX_POLL_DOWNTIME 300 /* 5 minutes by default */
#endif /* MOM */
/*
* specific structures for Job Array attributes
*/
/* subjob index table */
typedef struct ajinfo {
int tkm_ct; /* count of original entries in table */
int tkm_start; /* start of range (x in x-y:z) */
int tkm_end; /* end of range (y in x-y:z) */
int tkm_step; /* stepping factor for range (z in x-y:z) */
int tkm_flags; /* special flags for array job */
int tkm_subjsct[PBS_NUMJOBSTATE]; /* count of subjobs in various states */
int tkm_dsubjsct; /* count of deleted subjobs */
range *trm_quelist; /* pointer to range list */
} ajinfo_t;
/*
* Discard Job Structure, see Server's discard_job function
* Used to record which Mom has responded to when we need to tell them
* to discard a running job because of problems with a Mom
*/
struct jbdscrd {
struct machine_info *jdcd_mom; /* ptr to Mom */
int jdcd_state; /* 0 - waiting on her */
};
#define JDCD_WAITING 0 /* still waiting to hear from this Mom */
#define JDCD_REPLIED 1 /* this Mom has replied to the discard job */
#define JDCD_DOWN -1 /* this Mom is down */
/* Special array job flags in tkm_flags */
#define TKMFLG_NO_DELETE 0x01 /* delete subjobs in progess */
#define TKMFLG_CHK_ARRAY 0x02 /* chk_array_doneness() already in call stack */
/* Structure for block job reply processing */
struct block_job_reply {
char jobid[PBS_MAXSVRJOBID + 1];
char client[PBS_MAXHOSTNAME + 1];
int port;
int exitstat;
time_t reply_time; /* The timestamp at which the block job tried it's first attempt to reply */
char *msg; /* Abort message to be send to client */
int fd;
};
#define BLOCK_JOB_REPLY_TIMEOUT 60
/*
* THE JOB
*
* This structure is used by the server to maintain internal
* quick access to the state and status of each job.
* There is one instance of this structure per job known by the server.
*
* This information must be PRESERVED and is done so by updating the
* job file in the jobs subdirectory which corresponds to this job.
*
* ji_state is the state of the job. It is kept up front to provide for a
* "quick" update of the job state with minimum rewritting of the job file.
* Which is why the sub-struct ji_qs exists, that is the part which is
* written on the "quick" save. Update the symbolic constants JSVERSION_**
* if any changes to the format of the "quick-save" area are made.
*
* The unparsed string set forms of the attributes (including resources)
* are maintained in the struct attrlist as discussed above.
*/
#define JSVERSION_18 800 /* 18 denotes the PBS version and it covers the job structure from >= 13.x to <= 18.x */
#define JSVERSION_19 1900 /* 1900 denotes the 19.x.x version */
#define JSVERSION 2200 /* denotes 22.x and newer */
#define ji_taskid ji_extended.ji_ext.ji_taskidx
#define ji_nodeid ji_extended.ji_ext.ji_nodeidx
enum bg_hook_request {
BG_NONE,
BG_IS_DISCARD_JOB,
BG_PBS_BATCH_DeleteJob,
BG_PBSE_SISCOMM,
BG_IM_DELETE_JOB_REPLY,
BG_IM_DELETE_JOB,
BG_IM_DELETE_JOB2,
BG_CHECKPOINT_ABORT
};
struct job {
/*
* Note: these members, upto ji_qs, are not saved to disk
* IMPORTANT: if adding to this are, see create_subjob()
* in array_func.c; add the copy of the required elements
*/
pbs_list_link ji_alljobs; /* links to all jobs in server */
pbs_list_link ji_jobque; /* SVR: links to jobs in same queue, MOM: links to polled jobs */
pbs_list_link ji_unlicjobs; /* links to unlicensed jobs */
int ji_momhandle; /* open connection handle to MOM */
int ji_mom_prot; /* PROT_TCP or PROT_TPP */
struct batch_request *ji_rerun_preq; /* outstanding rerun request */
#ifdef PBS_MOM
void *ji_pending_ruu; /* pending last update */
struct batch_request *ji_preq; /* outstanding request */
struct grpcache *ji_grpcache; /* cache of user's groups */
enum PBS_Chkpt_By ji_chkpttype; /* checkpoint type */
time_t ji_chkpttime; /* periodic checkpoint time */
time_t ji_chkptnext; /* next checkpoint time */
time_t ji_sampletim; /* last usage sample time, irix only */
time_t ji_polltime; /* last poll from mom superior */
time_t ji_actalarm; /* time of site callout alarm */
time_t ji_joinalarm; /* time of job's sister join job alarm, also, time obit sent, all */
time_t ji_overlmt_timestamp; /*time the job exceeded limit*/
int ji_jsmpipe; /* pipe from child starter process */
int ji_mjspipe; /* pipe to child starter for ack */
int ji_jsmpipe2; /* pipe for child starter process to send special requests to parent mom */
int ji_mjspipe2; /* pipe for parent mom to ack special request from child starter process */
int ji_child2parent_job_update_pipe; /* read pipe to receive special request from child starter process */
int ji_parent2child_job_update_pipe; /* write pipe for parent mom to send info to child starter process */
int ji_parent2child_job_update_status_pipe; /* write pipe for parent mom to send job update status to child starter process */
int ji_parent2child_moms_status_pipe; /* write pipe for parent mom to send sister moms status to child starter process */
int ji_updated; /* set to 1 if job's node assignment was updated */
time_t ji_walltime_stamp; /* time stamp for accumulating walltime */
struct work_task *ji_bg_hook_task;
struct work_task *ji_report_task;
#ifdef WIN32
HANDLE ji_momsubt; /* process HANDLE to mom subtask */
#else /* not WIN32 */
pid_t ji_momsubt; /* pid of mom subtask */
#endif /* WIN32 */
struct var_table ji_env; /* environment for the job */
/* ptr to post processing func */
void (*ji_mompost)(struct job *, int);
tm_event_t ji_postevent; /* event waiting on mompost */
int ji_numnodes; /* number of nodes (at least 1) */
int ji_numrescs; /* number of entries in ji_resources*/
int ji_numvnod; /* number of virtual nodes */
int ji_num_assn_vnodes; /* number of virtual nodes (full count) */
tm_event_t ji_obit; /* event for end-of-job */
hnodent *ji_hosts; /* ptr to job host management stuff */
vmpiprocs *ji_vnods; /* ptr to job vnode management stuff */
noderes *ji_resources; /* ptr to array of node resources */
vmpiprocs *ji_assn_vnodes; /* ptr to actual assigned vnodes (for hooks) */
pbs_list_head ji_tasks; /* list of task structs */
pbs_list_head ji_failed_node_list; /* list of mom nodes which fail to join job */
pbs_list_head ji_node_list; /* list of functional mom nodes with vnodes assigned to the job */
tm_node_id ji_nodekill; /* set to nodeid requesting job die */
int ji_flags; /* mom only flags */
void *ji_setup; /* save setup info */
#ifdef WIN32
HANDLE ji_hJob; /* handle for job */
struct passwd *ji_user; /* user info */
#endif /* WIN32 */
int ji_stdout; /* socket for stdout */
int ji_stderr; /* socket for stderr */
int ji_ports[2]; /* ports for stdout/err */
enum bg_hook_request ji_hook_running_bg_on; /* set when hook starts in the background*/
int ji_msconnected; /* 0 - not connected, 1 - connected */
pbs_list_head ji_multinodejobs; /* links to recovered multinode jobs */
#else /* END Mom ONLY - start Server ONLY */
struct batch_request *ji_pmt_preq; /* outstanding preempt job request for deleting jobs */
int ji_discarding; /* discarding job */
struct batch_request *ji_prunreq; /* outstanding runjob request */
pbs_list_head ji_svrtask; /* links to svr work_task list */
struct pbs_queue *ji_qhdr; /* current queue header */
struct resc_resv *ji_myResv; /* !=0 job belongs to a reservation, see also, attribute JOB_ATR_myResv */
int ji_lastdest; /* last destin tried by route */
int ji_retryok; /* ok to retry, some reject was temp */
int ji_terminated; /* job terminated by deljob batch req */
int ji_deletehistory; /* job history should not be saved */
pbs_list_head ji_rejectdest; /* list of rejected destinations */
struct job *ji_parentaj; /* subjob: parent Array Job */
ajinfo_t *ji_ajinfo; /* ArrayJob: information about subjobs and its state counts */
struct jbdscrd *ji_discard; /* see discard_job() */
int ji_jdcd_waiting; /* set if waiting on a mom for a response to discard job request */
char *ji_acctrec; /* holder for accounting info */
char *ji_clterrmsg; /* error message to return to client */
/*
* This variable is used to temporarily hold the script for a new job
* in memory instead of immediately saving it to the database in the
* req_jobscript function. The script is eventually saved into the
* database along with saving the job structure as part of req_commit
* under one single transaction. After this the memory is freed.
*/
char *ji_script;
/*
* This flag is to indicate if queued entity limit attribute usage
* is decremented when the job is run
*/
int ji_etlimit_decr_queued;
struct preempt_ordering *preempt_order;
int preempt_order_index;
struct work_task *ji_prov_startjob_task;
#endif /* END SERVER ONLY */
/*
* fixed size internal data - maintained via "quick save"
* some of the items are copies of attributes, if so this
* internal version takes precendent
*
* This area CANNOT contain any pointers!
*/
#ifndef PBS_MOM
char qs_hash[DIGEST_LENGTH];
#endif
struct jobfix {
int ji_jsversion; /* job structure version - JSVERSION */
int ji_svrflags; /* server flags */
time_t ji_stime; /* time job started execution */
time_t ji_obittime; /* time job has ended execution */
char ji_jobid[PBS_MAXSVRJOBID + 1]; /* job identifier */
char ji_fileprefix[PBS_JOBBASE + 1]; /* no longer used */
char ji_queue[PBS_MAXQUEUENAME + 1]; /* name of current queue */
char ji_destin[PBS_MAXROUTEDEST + 1]; /* dest from qmove/route, MomS for execution */
int ji_un_type; /* type of ji_un union */
union { /* depends on type of queue currently in */
struct { /* if in execution queue .. */
pbs_net_t ji_momaddr; /* host addr of Server */
unsigned int ji_momport; /* port # */
int ji_exitstat; /* job exit status from MOM */
} ji_exect;
struct {
time_t ji_quetime; /* time entered queue */
time_t ji_rteretry; /* route retry time */
} ji_routet;
struct {
int ji_fromsock; /* socket job coming over */
pbs_net_t ji_fromaddr; /* host job coming from */
unsigned int ji_scriptsz; /* script size */
} ji_newt;
struct {
pbs_net_t ji_svraddr; /* host addr of Server */
int ji_exitstat; /* job exit status from MOM */
uid_t ji_exuid; /* execution uid */
gid_t ji_exgid; /* execution gid */
} ji_momt;
} ji_un;
} ji_qs;
/*
* Extended job save area
*
* This area CANNOT contain any pointers!
*/
union jobextend {
char fill[256]; /* fill to keep same size */
struct {
char ji_jid[8]; /* extended job save data for ALPS */
int ji_credtype; /* credential type */
#ifdef PBS_MOM
tm_host_id ji_nodeidx; /* my node id */
tm_task_id ji_taskidx; /* generate task id's for job */
int ji_stdout;
int ji_stderr;
#if MOM_ALPS
long ji_reservation;
/* ALPS reservation identifier */
unsigned long long ji_pagg;
/* ALPS process aggregate ID */
#endif /* MOM_ALPS */
#endif /* PBS_MOM */
} ji_ext;
} ji_extended;
/*
* The following array holds the decode format of the attributes.
* Its presence is for rapid acces to the attributes.
*/
attribute ji_wattr[JOB_ATR_LAST]; /* decoded attributes */
short newobj; /* newly created job? */
};
typedef struct job job;
#ifdef PBS_MOM
/*
** Tasks are sessions belonging to a job, running on one of the
** nodes assigned to the job.
*/
typedef struct pbs_task {
job *ti_job; /* pointer to owning job */
unsigned long ti_cput; /* track cput by task */
pbs_list_link ti_jobtask; /* links to tasks for this job */
int *ti_tmfd; /* DIS file descriptors to tasks */
int ti_tmnum; /* next avail entry in ti_tmfd */
int ti_tmmax; /* size of ti_tmfd */
int ti_protover; /* protocol version number */
int ti_flags; /* task internal flags */
#ifdef WIN32
HANDLE ti_hProc; /* keep proc handle */
#endif
tm_event_t ti_register; /* event if task registers */
pbs_list_head ti_obits; /* list of obit events */
pbs_list_head ti_info; /* list of named info */
struct taskfix {
char ti_parentjobid[PBS_MAXSVRJOBID + 1];
tm_node_id ti_parentnode; /* parent vnode */
tm_node_id ti_myvnode; /* my vnode */
tm_task_id ti_parenttask; /* parent task */
tm_task_id ti_task; /* task's taskid */
int ti_status; /* status of task */
pid_t ti_sid; /* session id */
int ti_exitstat; /* exit status */
union {
int ti_hold[16]; /* reserved space */
} ti_u;
} ti_qs;
} pbs_task;
/*
** A linked list of eventent structures is maintained for all events
** for which we are waiting for another MOM to report back.
*/
typedef struct eventent {
int ee_command; /* command event is for */
int ee_fd; /* TM stream */
int ee_retry; /* event message retry attempt number */
tm_event_t ee_client; /* client event number */
tm_event_t ee_event; /* MOM event number */
tm_task_id ee_taskid; /* which task id */
char **ee_argv; /* save args for spawn */
char **ee_envp; /* save env for spawn */
pbs_list_link ee_next; /* link to next one */
} eventent;
/*
** The information needed for a task manager obit request
** is indicated with OBIT_TYPE_TMEVENT. The information needed
** for a batch request is indicated with OBIT_TYPE_BREVENT.
*/
#define OBIT_TYPE_TMEVENT 0
#define OBIT_TYPE_BREVENT 1
/*
** A task can have events which are triggered when it exits.
** These are tracked by obitent structures linked to the task.
*/
typedef struct obitent {
int oe_type; /* what kind of obit */
union oe_u {
struct oe_tm {
int oe_fd; /* TM reply fd */
tm_node_id oe_node; /* where does notification go */
tm_event_t oe_event; /* event number */
tm_task_id oe_taskid; /* which task id */
} oe_tm;
struct batch_request *oe_preq;
} oe_u;
pbs_list_link oe_next; /* link to next one */
} obitent;
/*
** A task can have a list of named infomation which it makes
** available to other tasks in the job.
*/
typedef struct infoent {
char *ie_name; /* published name */
void *ie_info; /* the glop */
size_t ie_len; /* how much glop */
pbs_list_link ie_next; /* link to next one */
} infoent;
#define TI_FLAGS_INIT 1 /* task has called tm_init */
#define TI_FLAGS_CHKPT 2 /* task has checkpointed */
#define TI_FLAGS_ORPHAN 4 /* MOM not parent of task */
#define TI_FLAGS_SAVECKP 8 /* save value of CHKPT flag during checkpoint op */
#define TI_STATE_EMBRYO 0
#define TI_STATE_RUNNING 1
#define TI_STATE_EXITED 2 /* ti_exitstat valid */
#define TI_STATE_DEAD 3
/*
** Here is the set of commands for InterMOM (IM) requests.
*/
#define IM_ALL_OKAY 0
#define IM_JOIN_JOB 1
#define IM_KILL_JOB 2
#define IM_SPAWN_TASK 3
#define IM_GET_TASKS 4
#define IM_SIGNAL_TASK 5
#define IM_OBIT_TASK 6
#define IM_POLL_JOB 7
#define IM_GET_INFO 8
#define IM_GET_RESC 9
#define IM_ABORT_JOB 10
#define IM_GET_TID 11 /* no longer used */
#define IM_SUSPEND 12
#define IM_RESUME 13
#define IM_CHECKPOINT 14
#define IM_CHECKPOINT_ABORT 15
#define IM_RESTART 16
#define IM_DELETE_JOB 17
#define IM_REQUEUE 18
#define IM_DELETE_JOB_REPLY 19
#define IM_SETUP_JOB 20
#define IM_DELETE_JOB2 21 /* sent by sister mom to delete job early */
#define IM_SEND_RESC 22
#define IM_UPDATE_JOB 23
#define IM_EXEC_PROLOGUE 24
#define IM_CRED 25
#define IM_PMIX 26
#define IM_RECONNECT_TO_MS 27
#define IM_JOIN_RECOV_JOB 28
#define IM_ERROR 99
#define IM_ERROR2 100
eventent *
event_alloc(job *pjob,
int command,
int fd,
hnodent *pnode,
tm_event_t event,
tm_task_id taskid);
pbs_task *momtask_create(job *pjob);
pbs_task *
task_find(job *pjob,
tm_task_id taskid);
#endif /* MOM */
/*
* server flags (in ji_svrflags)
*/
#define JOB_SVFLG_HERE 0x01 /* SERVER: job created here */
/* MOM: set for Mother Superior */
#define JOB_SVFLG_HASWAIT 0x02 /* job has timed task entry for wait time */
#define JOB_SVFLG_HASRUN 0x04 /* job has been run before (being rerun) */
#define JOB_SVFLG_HOTSTART 0x08 /* job was running, if hot init, restart */
#define JOB_SVFLG_CHKPT 0x10 /* job has checkpoint file for restart */
#define JOB_SVFLG_SCRIPT 0x20 /* job has a Script file */
#define JOB_SVFLG_OVERLMT1 0x40 /* job over limit first time, MOM only */
#define JOB_SVFLG_OVERLMT2 0x80 /* job over limit second time, MOM only */
#define JOB_SVFLG_ChkptMig 0x100 /* job has migratable checkpoint */
#define JOB_SVFLG_Suspend 0x200 /* job suspended (signal suspend) */
#define JOB_SVFLG_StagedIn 0x400 /* job has files that have been staged in */
#define JOB_SVFLG_HASHOLD 0x800 /* job has a hold request sent to MoM */
#define JOB_SVFLG_HasNodes 0x1000 /* job has nodes allocated to it */
#define JOB_SVFLG_RescAssn 0x2000 /* job resources accumulated in server/que */
#define JOB_SVFLG_SPSwitch 0x2000 /* SP switch loaded for job, SP MOM only */
#define JOB_SVFLG_Actsuspd 0x4000 /* job suspend because workstation active */
#define JOB_SVFLG_cpuperc 0x8000 /* cpupercent violation logged, MOM only */
#define JOB_SVFLG_ArrayJob 0x10000 /* Job is an Array Job */
#define JOB_SVFLG_SubJob 0x20000 /* Job is a subjob of an Array */
#define JOB_SVFLG_StgoFal 0x40000 /* Stageout failed, del jobdir, MOM only */
#define JOB_SVFLG_TERMJOB 0x80000 /* terminate in progress by TERM, MOM only */
/* 0x100000 is UNUSED, previously called JOB_SVFLG_StgoDel for stageout succ */
/* If you intend to use it, make sure jobs to be recovered do not have
* 0x100000 bit set. Refer SPM229744
*/
#define JOB_SVFLG_AdmSuspd 0x200000 /* Job is suspended for maintenance */
#define MAIL_NONE (int) 'n'
#define MAIL_ABORT (int) 'a'
#define MAIL_BEGIN (int) 'b'
#define MAIL_END (int) 'e'
#define MAIL_OTHER (int) 'o'
#define MAIL_STAGEIN (int) 's'
#define MAIL_CONFIRM (int) 'c' /*scheduler requested reservation be confirmed*/
#define MAIL_SUBJOB (int) 'j'
#define MAIL_NORMAL 0
#define MAIL_FORCE 1
#define JOB_FILE_COPY ".JC" /* tmp copy while updating */
#define JOB_FILE_SUFFIX ".JB" /* job control file */
#define JOB_CRED_SUFFIX ".CR" /* job credential file */
#define JOB_EXPORT_SUFFIX ".XP" /* job export security context */
#define JOB_SCRIPT_SUFFIX ".SC" /* job script file */
#define JOB_STDOUT_SUFFIX ".OU" /* job standard out */
#define JOB_STDERR_SUFFIX ".ER" /* job standard error */
#define JOB_CKPT_SUFFIX ".CK" /* job checkpoint file */
#define JOB_TASKDIR_SUFFIX ".TK" /* job task directory */
#define JOB_BAD_SUFFIX ".BD" /* save bad job file */
#define JOB_DEL_SUFFIX ".RM" /* file pending to be removed */
/*
* Job states are defined by POSIX as:
*/
#define JOB_STATE_TRANSIT 0
#define JOB_STATE_QUEUED 1
#define JOB_STATE_HELD 2
#define JOB_STATE_WAITING 3
#define JOB_STATE_RUNNING 4
#define JOB_STATE_EXITING 5
#define JOB_STATE_EXPIRED 6
#define JOB_STATE_BEGUN 7
/* Job states defined for history jobs and OGF-BES model */
#define JOB_STATE_MOVED 8
#define JOB_STATE_FINISHED 9
#define JOB_STATE_LTR_UNKNOWN '0'
#define JOB_STATE_LTR_BEGUN 'B'
#define JOB_STATE_LTR_EXITING 'E'
#define JOB_STATE_LTR_FINISHED 'F'
#define JOB_STATE_LTR_HELD 'H'
#define JOB_STATE_LTR_MOVED 'M'
#define JOB_STATE_LTR_QUEUED 'Q'
#define JOB_STATE_LTR_RUNNING 'R'
#define JOB_STATE_LTR_SUSPENDED 'S'
#define JOB_STATE_LTR_TRANSIT 'T'
#define JOB_STATE_LTR_USUSPENDED 'U'
#define JOB_STATE_LTR_WAITING 'W'
#define JOB_STATE_LTR_EXPIRED 'X'
/*
* job sub-states are defined by PBS (more detailed) as:
*/
#define JOB_SUBSTATE_UNKNOWN -1
#define JOB_SUBSTATE_TRANSIN 00 /* Transit in, wait for commit, commit not yet called */
#define JOB_SUBSTATE_TRANSICM 01 /* Transit in, job is being commited */
#define JOB_SUBSTATE_TRNOUT 02 /* transiting job outbound */
#define JOB_SUBSTATE_TRNOUTCM 03 /* transiting outbound, rdy to commit */
#define JOB_SUBSTATE_QUEUED 10 /* job queued and ready for selection */
#define JOB_SUBSTATE_PRESTAGEIN 11 /* job queued, has files to stage in */
#define JOB_SUBSTATE_SYNCRES 13 /* job waiting on sync start ready */
#define JOB_SUBSTATE_STAGEIN 14 /* job staging in files then wait */
#define JOB_SUBSTATE_STAGEGO 15 /* job staging in files and then run */
#define JOB_SUBSTATE_STAGECMP 16 /* job stage in complete */
#define JOB_SUBSTATE_HELD 20 /* job held - user or operator */
#define JOB_SUBSTATE_SYNCHOLD 21 /* job held - waiting on sync regist */
#define JOB_SUBSTATE_DEPNHOLD 22 /* job held - waiting on dependency */
#define JOB_SUBSTATE_WAITING 30 /* job waiting on execution time */
#define JOB_SUBSTATE_STAGEFAIL 37 /* job held - file stage in failed */
#define JOB_SUBSTATE_PRERUN 41 /* job set to MOM to run */
#define JOB_SUBSTATE_RUNNING 42 /* job running */
#define JOB_SUBSTATE_SUSPEND 43 /* job suspended by client */
#define JOB_SUBSTATE_SCHSUSP 45 /* job supsended by scheduler */
#define JOB_SUBSTATE_EXITING 50 /* Start of job exiting processing */
#define JOB_SUBSTATE_STAGEOUT 51 /* job staging out (other) files */
#define JOB_SUBSTATE_STAGEDEL 52 /* job deleting staged out files */
#define JOB_SUBSTATE_EXITED 53 /* job exit processing completed */
#define JOB_SUBSTATE_ABORT 54 /* job is being aborted by server */
#define JOB_SUBSTATE_KILLSIS 56 /* (MOM) job kill IM to sisters */
#define JOB_SUBSTATE_RUNEPILOG 57 /* (MOM) job epilogue running */
#define JOB_SUBSTATE_OBIT 58 /* (MOM) job obit notice sent */
#define JOB_SUBSTATE_TERM 59 /* Job is in site termination stage */
#define JOB_SUBSTATE_DELJOB 153 /* (MOM) Job del_job_wait to sisters */
#define JOB_SUBSTATE_RERUN 60 /* job is rerun, recover output stage */
#define JOB_SUBSTATE_RERUN1 61 /* job is rerun, stageout phase */
#define JOB_SUBSTATE_RERUN2 62 /* job is rerun, delete files stage */
#define JOB_SUBSTATE_RERUN3 63 /* job is rerun, mom delete job */
#define JOB_SUBSTATE_EXPIRED 69 /* subjob (of an array) is gone */
#define JOB_SUBSTATE_BEGUN 70 /* Array job has begun */
#define JOB_SUBSTATE_PROVISION 71 /* job is waiting for provisioning tocomplete */
#define JOB_SUBSTATE_WAITING_JOIN_JOB 72 /* job waiting on IM_JOIN_JOB completion */
/*
* Job sub-states defined in PBS to support history jobs and OGF-BES model:
*/
#define JOB_SUBSTATE_TERMINATED 91
#define JOB_SUBSTATE_FINISHED 92
#define JOB_SUBSTATE_FAILED 93
#define JOB_SUBSTATE_MOVED 94
/* decriminator for ji_un union type */
#define JOB_UNION_TYPE_NEW 0
#define JOB_UNION_TYPE_EXEC 1
#define JOB_UNION_TYPE_ROUTE 2
#define JOB_UNION_TYPE_MOM 3
/* job hold (internal) types */
#define HOLD_n 0
#define HOLD_u 1
#define HOLD_o 2
#define HOLD_s 4
#define HOLD_bad_password 8
/* Array Job related Defines */
/* See is_job_array() in array_func.c */
#define IS_ARRAY_NO 0 /* Not an array job nor subjob */
#define IS_ARRAY_ArrayJob 1 /* Is an Array Job */
#define IS_ARRAY_Single 2 /* A single Sub Job */
#define IS_ARRAY_Range 3 /* A range of Subjobs */
#define PBS_FILE_ARRAY_INDEX_TAG "^array_index^"
/* Special Job Exit Values, Set by the job starter (child of MOM) */
/* see server/req_jobobit() & mom/start_exec.c */
#define JOB_EXEC_OK 0 /* job exec successfull */
#define JOB_EXEC_FAIL1 -1 /* Job exec failed, before files, no retry */
#define JOB_EXEC_FAIL2 -2 /* Job exec failed, after files, no retry */
#define JOB_EXEC_RETRY -3 /* Job execution failed, do retry */
#define JOB_EXEC_INITABT -4 /* Job aborted on MOM initialization */
#define JOB_EXEC_INITRST -5 /* Job aborted on MOM init, chkpt, no migrate */
#define JOB_EXEC_INITRMG -6 /* Job aborted on MOM init, chkpt, ok migrate */
#define JOB_EXEC_BADRESRT -7 /* Job restart failed */
#define JOB_EXEC_FAILUID -10 /* invalid uid/gid for job */
#define JOB_EXEC_RERUN -11 /* Job rerun */
#define JOB_EXEC_CHKP -12 /* Job was checkpointed and killed */
#define JOB_EXEC_FAIL_PASSWORD -13 /* Job failed due to a bad password */
#define JOB_EXEC_RERUN_SIS_FAIL -14 /* Job rerun */
#define JOB_EXEC_QUERST -15 /* requeue job for restart from checkpoint */
#define JOB_EXEC_FAILHOOK_RERUN -16 /* job exec failed due to a hook rejection, requeue job for later retry (usually returned by the "early" hooks" */
#define JOB_EXEC_FAILHOOK_DELETE -17 /* job exec failed due to a hook rejection, delete the job at end */
#define JOB_EXEC_HOOK_RERUN -18 /* a hook requested for job to be requeued */
#define JOB_EXEC_HOOK_DELETE -19 /* a hook requested for job to be deleted */
#define JOB_EXEC_RERUN_MS_FAIL -20 /* Mother superior connection failed */
#define JOB_EXEC_FAIL_SECURITY -21 /* Security breach in PBS directory */
#define JOB_EXEC_HOOKERROR -22 /* job exec failed due to
* unexpected exception or
* hook execution timed out
*/
#define JOB_EXEC_FAIL_KRB5 -23 /* Error no kerberos credentials supplied */
#define JOB_EXEC_UPDATE_ALPS_RESV_ID 1 /* Update ALPS reservation ID to parent mom as soon
* as it is available.
* This is neither a success nor a failure exit code,
* so we are using a positive value
*/
#define JOB_EXEC_KILL_NCPUS_BURST -24 /* job exec failed due to exceeding ncpus (burst) */
#define JOB_EXEC_KILL_NCPUS_SUM -25 /* job exec failed due to exceeding ncpus (sum) */
#define JOB_EXEC_KILL_VMEM -26 /* job exec failed due to exceeding vmem */
#define JOB_EXEC_KILL_MEM -27 /* job exec failed due to exceeding mem */
#define JOB_EXEC_KILL_CPUT -28 /* job exec failed due to exceeding cput */
#define JOB_EXEC_KILL_WALLTIME -29 /* job exec failed due to exceeding walltime */
#define JOB_EXEC_JOINJOB -30 /* Job exec failed due to join job error */
/*
* Fake "random" number added onto the end of the staging
* and execution directory when sandbox=private
* used in jobdirname()
*/
#define FAKE_RANDOM "x8z"
/* The default project assigned to jobs when project attribute is unset */
#define PBS_DEFAULT_PROJECT "_pbs_project_default"
extern void add_dest(job *);
extern int depend_on_que(attribute *, void *, int);
extern int depend_on_exec(job *);
extern int depend_runone_remove_dependency(job *);
extern int depend_runone_hold_all(job *);
extern int depend_runone_release_all(job *);
extern int depend_on_term(job *);
extern struct depend *find_depend(int type, attribute *pattr);
extern struct depend_job *find_dependjob(struct depend *pdep, char *name);
extern int send_depend_req(job *pjob, struct depend_job *pparent, int type, int op, int schedhint, void (*postfunc)(struct work_task *));
extern void post_runone(struct work_task *pwt);
extern job *find_job(char *);
extern char *get_variable(job *, char *);
extern void check_block(job *, char *);
extern char *lookup_variable(void *, int, char *);
extern void issue_track(job *);
extern void issue_delete(job *);
extern int job_abt(job *, char *);
extern job *job_alloc(void);
extern void job_free(job *);
extern int modify_job_attr(job *, svrattrl *, int, int *);
extern char *prefix_std_file(job *, int);
extern void cat_default_std(job *, int, char *, char **);
extern int set_objexid(void *, int, attribute *);
extern bool update_deljob_rply(struct batch_request *, char *, int);
#if 0
extern int site_check_user_map(job *, char *);
#endif
extern int site_check_user_map(void *, int, char *);
extern int site_allow_u(char *user, char *host);
extern void svr_dequejob(job *);
extern int svr_enquejob(job *, char *);
extern void svr_evaljobstate(job *, char *, int *, int);
extern int svr_setjobstate(job *, char, int);
extern int state_char2int(char);
extern char state_int2char(int);
extern int uniq_nameANDfile(char *, char *, char *);
extern long determine_accruetype(job *);
extern int update_eligible_time(long, job *);
#define TOLERATE_NODE_FAILURES_ALL "all"
#define TOLERATE_NODE_FAILURES_JOB_START "job_start"
#define TOLERATE_NODE_FAILURES_NONE "none"
extern int do_tolerate_node_failures(job *);
int check_job_state(const job *pjob, char state);
int check_job_substate(const job *pjob, int substate);
char get_job_state(const job *pjob);
int get_job_state_num(const job *pjob);
long get_job_substate(const job *pjob);
char *get_jattr_str(const job *pjob, int attr_idx);
struct array_strings *get_jattr_arst(const job *pjob, int attr_idx);
pbs_list_head get_jattr_list(const job *pjob, int attr_idx);
long get_jattr_long(const job *pjob, int attr_idx);
long long get_jattr_ll(const job *pjob, int attr_idx);
svrattrl *get_jattr_usr_encoded(const job *pjob, int attr_idx);
svrattrl *get_jattr_priv_encoded(const job *pjob, int attr_idx);
void set_job_state(job *pjob, char val);
void set_job_substate(job *pjob, long val);
int set_jattr_str_slim(job *pjob, int attr_idx, char *val, char *rscn);
int set_jattr_l_slim(job *pjob, int attr_idx, long val, enum batch_op op);
int set_jattr_ll_slim(job *pjob, int attr_idx, long long val, enum batch_op op);
int set_jattr_b_slim(job *pjob, int attr_idx, long val, enum batch_op op);
int set_jattr_c_slim(job *pjob, int attr_idx, char val, enum batch_op op);
int set_jattr_generic(job *pjob, int attr_idx, char *val, char *rscn, enum batch_op op);
int is_jattr_set(const job *pjob, int attr_idx);
void free_jattr(job *pjob, int attr_idx);
void mark_jattr_not_set(job *pjob, int attr_idx);
void mark_jattr_set(job *pjob, int attr_idx);
attribute *get_jattr(const job *pjob, int attr_idx);
/*
* The filesystem related recovery/save routines are renamed
* with the suffix "_fs", and the database versions of them
* are suffixed "_db". This distinguishes between the two
* version. The "_fs" version will continue to be used by
* migration routine "svr_migrate_data" and by "mom". Rest of
* the server code will typically use only the "_db" routines.
* Since mom uses only the "_fs" versions, define the "_fs"
* versions to names with the suffix, so that the mom code
* remain unchanges and continues to use the "_fs" versions.
*/
#ifdef PBS_MOM
extern job *job_recov_fs(char *);
extern int job_save_fs(job *);
#define job_save job_save_fs
#define job_recov job_recov_fs
#else
extern job *job_recov_db(char *, job *pjob);
extern int job_save_db(job *);
#define job_save job_save_db
#define job_recov job_recov_db
extern char *get_job_credid(char *);
#endif
#ifdef _BATCH_REQUEST_H
extern job *chk_job_request(char *, struct batch_request *, int *, int *);
extern int net_move(job *, struct batch_request *);
extern int svr_chk_owner(struct batch_request *, job *);
extern int svr_movejob(job *, char *, struct batch_request *);
extern struct batch_request *cpy_stage(struct batch_request *, job *, enum job_atr, int);
#ifdef _RESERVATION_H
extern int svr_chk_ownerResv(struct batch_request *, resc_resv *);
#endif /* _RESERVATION_H */
#endif /* _BATCH_REQUEST_H */
#ifdef _QUEUE_H
extern int svr_chkque(job *, pbs_queue *, char *, char *, int mtype);
extern int default_router(job *, pbs_queue *, long);
extern int site_alt_router(job *, pbs_queue *, long);
extern int site_acl_check(job *, pbs_queue *);
#endif /* _QUEUE_H */
#ifdef _WORK_TASK_H
extern int issue_signal(job *, char *, void (*)(struct work_task *), void *);
extern void on_job_exit(struct work_task *);
#endif /* _WORK_TASK_H */
#ifdef _PBS_IFL_H
extern int update_resources_list(job *, char *, int, char *, enum batch_op op, int, int);
#endif
extern int Mystart_end_dur_wall(void *, int);
extern int get_wall(job *);
extern int get_softwall(job *);
extern int get_used_wall(job *);
extern int get_used_cput(job *);
extern int get_cput(job *);
extern void remove_deleted_resvs(void);
extern void degrade_corrupted_confirmed_resvs(void);
extern int pbsd_init_job(job *pjob, int type);
extern void del_job_related_file(job *pjob, char *fsuffix);
#ifdef PBS_MOM
extern void del_job_dirs(job *pjob, char *taskdir);
extern void del_chkpt_files(job *pjob);
#endif
extern void get_jobowner(char *, char *);
extern struct batch_request *cpy_stage(struct batch_request *, job *, enum job_atr, int);
extern struct batch_request *cpy_stdfile(struct batch_request *, job *, enum job_atr);
extern int has_stage(job *);
void svr_evalsetjobstate(job *jobp);
#ifdef __cplusplus
}
#endif
#endif /* _PBS_JOB_H */