wdgzyjx/myxce
1
0
Fork 0
forked from lei.chen/myxce
Code Pull requests Activity

merge job rusage and pending reason support

Browse source
This commit is contained in:
Rongsong Shen 2025-12-16 14:36:28 +08:00
parent 30824ae5ea
commit cb0c7fe9e7
82 changed files with 4280 additions and 2561 deletions
Download patch file Download diff file Expand all files Collapse all files

110
src/controller/job.c
View file

@ -3,6 +3,7 @@
#include <time.h>
#include "cluster.h"
#include "preason.h"
#include "xce/cluster_def.h"
#include "xce/errnox.h"
#include "xce/types.h"
@ -457,7 +458,10 @@ int handle_jobsubmit(cluster_t* c, endpoint_t* ep, xce_request_t* request)
/* Run hooks to tell scheduler */
j = jobmgr_lockjob(&(c->jobmgr), jobid, OBJREF_READ);
if (j == NULL) {
ERROR("%s fail to lock job %lu", __func__, jobid);
char idstring[128];
fmt_jobid(idstring, sizeof(idstring) - 1, jobid);
ERROR("%s fail to lock job %s (%lu)", __func__, idstring, jobid);
} else {
xlist_init(&jlist, 2);
if (xlist_store(&jlist, j) >= 0) {
@ -777,6 +781,9 @@ int update_job_rusages(cluster_t* c, endpoint_t* ep, xce_request_t* request)
TRACE("%s trying to update job(%ld) rusage", __func__, rusages->jobid);
xce_run_hooks(&(c->hooksys), XCE_CONTROLLER_HOOK_JOB, HOOK_BEFORE_EVENT,
HOOK_DETAIL_JOB_RUSAGE_UPDATE, c, rusages);
j = jobmgr_lockjob(&(c->jobmgr), rusages->jobid, OBJREF_UPDATE);
if (j == NULL) {
PERROR("jobmgr_lockjob");
@ -815,5 +822,106 @@ int update_job_rusages(cluster_t* c, endpoint_t* ep, xce_request_t* request)
HOOK_DETAIL_JOB_UPDATE, c, j);
obj_deref(j);
xce_run_hooks(&(c->hooksys), XCE_CONTROLLER_HOOK_JOB, HOOK_AFTER_EVENT,
HOOK_DETAIL_JOB_RUSAGE_UPDATE, c, rusages);
return 0;
}
static int set_preason(cluster_t* c, xce_pendreason_t* reasons)
{
job_t* j;
xce_job_t* job;
TRACE("%s jobid %ld", __func__, reasons->jobid);
j = jobmgr_lockjob(&(c->jobmgr), reasons->jobid, OBJREF_UPDATE);
if (j == NULL) {
PERROR("jobmgr_lockjob");
return -ENOENT;
}
xce_run_hooks(&(c->hooksys), XCE_CONTROLLER_HOOK_JOB, HOOK_BEFORE_EVENT,
HOOK_DETAIL_SETPREASON, c, j);
job = &(j->job);
if (job->pend_reasons != NULL) {
erase_job_preason(job->pend_reasons);
FREEUP(job->pend_reasons);
}
MOVE_PTR(job->pend_reasons, reasons->reasons);
xce_run_hooks(&(c->hooksys), XCE_CONTROLLER_HOOK_JOB, HOOK_AFTER_EVENT,
HOOK_DETAIL_SETPREASON, c, j);
obj_deref(j);
return 0;
}
int do_set_preason(cluster_t* c, endpoint_t* ep, xce_request_t* request)
{
int i;
xce_preason_set_t* reasons = request->body.preasonmsg;
xce_reply_t reply;
init_reply_by_req(&reply, request);
reply.status = 0;
reply.data_which = RequestReply_data_nodata;
xce_reply(ep, c->myhost, &reply);
for (i = 0; i < reasons->nreasons; i++) {
int rc = 0;
if (reasons->reasons[i] == NULL) {
continue;
}
rc = set_preason(c, reasons->reasons[i]);
if (rc < 0) {
PERROR("set_preason");
}
}
return 0;
}
int do_clr_preason(cluster_t* c, endpoint_t* ep, xce_request_t* request)
{
int i;
xce_reply_t reply;
xce_clear_reason_t* clreason = request->body.clrpreasonmsg;
init_reply_by_req(&reply, request);
reply.status = 0;
reply.data_which = RequestReply_data_nodata;
xce_reply(ep, c->myhost, &reply);
for (i = 0; i < clreason->njobs; i++) {
uint64_t jobid = clreason->jobs[i];
job_t* j;
xce_job_t* job;
j = jobmgr_lockjob(&(c->jobmgr), jobid, OBJREF_UPDATE);
if (j == NULL) {
PERROR("jobmgr_lockjob");
continue;
}
xce_run_hooks(&(c->hooksys), XCE_CONTROLLER_HOOK_JOB, HOOK_BEFORE_EVENT,
HOOK_DETAIL_CLRPREASON, c, j);
job = &(j->job);
if (job->pend_reasons != NULL) {
erase_job_preason(job->pend_reasons);
FREEUP(job->pend_reasons);
}
xce_run_hooks(&(c->hooksys), XCE_CONTROLLER_HOOK_JOB, HOOK_BEFORE_EVENT,
HOOK_DETAIL_CLRPREASON, c, j);
obj_deref(j);
}
return 0;
}