lei.chen/myxce
1
0
Fork 1
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix(harness): xce_test argument validation, RPC error text and leaks (13 bugs)

Browse source 
Stack overflows on user input:
- bug-1702 jobrun -m: host segment strncpy'd into a 256-byte buffer
  unbounded (ASAN stack-buffer-overflow); now cmdutil_split_host_tasks()
  refuses an over-long host.
- bug-1703 jobsub -n: min strncpy'd into a 64-byte buffer; now
  cmdutil_parse_ntasks() parses in place, 1 <= min <= max.
- resupdate's 16-entry vals[] had no bound either; a 17th -n overflowed.

Wrong or silent results:
- bug-1574 resupdate type/model: set_resval() had no string case, left the
  entry's name uninitialized, and crashed the encoder. Moved to
  resval_typeresolve.c as resval_fill(), which fills every type.
- bug-1705 jobsub -x was overwritten (and leaked) by the default exdata;
  cmdutil_default_exdata() only fills an unset one. jobsub's
  execuser/exechome/execcwd copies leaked on every call.
- bug-1707 a trailing -n without -v was dropped silently; resupdate also
  returned 0 after its RPC failed and "succeeded" with nothing to send.
- bug-1708 jobkill -s SIGKILL became signal 0 and was sent.
- bug-1709 jobinfo exited 0 with no -j, a malformed -j, or no hits.
- bug-1710 hello -c 0 sent nothing and reported success.
- bug-1712 a duplicate -v was reported as "no res name found".
- bug-1777 workerlist showed join-time maxmem/maxswap/maxtmp, not the
  effective values (hosts.yaml overrides) the reply already carries in
  resvals_to_sched and the scheduler uses.
- bug-1793 join's maxswap/maxtmp defaults were 1 KB / 2 KB in bytes; now
  1 GiB / 2 GiB, and -m/-s/-t no longer overflow atoi() above 2 GB.

Error reporting and leaks:
- bug-1796 RPC failures printed perror() of a stale errno ("Connection
  refused" / "Invalid argument" / "Inappropriate ioctl for device" for
  the same failure). cmdutil_rpc_error() prints the RPC's own code
  (E_XCE_* included) at all 19 sites.
- bug-1797 qctrl leaked the queue name on the RPC-error path; repeated
  -q/-h leaked the earlier copy.

New src/harness/cmdutil.c (+ cmdutil_unit_test, 14 cases); 9 new
test_test_cli.c cases, 8 RED on the old xce_test; new resval_fill,
worker_effective_attr and join-default tests (join ones RED on the old
join_worker_args.c). Harness suite 6/6 (asan), harness_unit_test 141/0.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Rongsong Shen 2026-09-28 04:17:04 +00:00
parent eee76a75ab
commit da1a663f05
Signed by untrusted user: rshen
GPG key ID: 80533536B0EB651C
14 changed files with 1107 additions and 286 deletions
Download patch file Download diff file Expand all files Collapse all files

472
src/harness/test.c
View file

@ -63,6 +63,7 @@
#include "snapverify.h" /* offline checkpoint-file structural validator (cmd_snapverify) */
#include "configcompare.h" /* cross-member active-config comparator (cmd_configcompare) */
#include "confirm.h" /* destructive-command confirmation gate (xce-test-cli-ux task 2.1) */
#include "cmdutil.h" /* argument parsing + RPC-error reporting helpers */
#include "joblist_display.h" /* worker_status_str/job_state_str/format_job_time
* (xce-test-cli-ux task 2.5 remediation) */
#include "join_worker_args.h" /* join_worker_parse_attrs (xce-test-cli-ux
@ -156,14 +157,6 @@ typedef struct {
int ntasks;
} host_task_t;
typedef struct {
char* PATH;
} sub_envs_t;
typedef struct {
sub_envs_t sub_env_vars;
} exdata_t;
#define MAX_RUN_JOBS (20)
#define MAX_TASK_HOSTS (5)
#define MAX_RUN_TASKS (50)
@ -425,7 +418,8 @@ static const char USAGE_JOIN_WORKER_HEAD[] =
" [-t <maxtmp>] [-d <ndisks>]\n"
" Joins the cluster as a worker (hosttype LINUX64, hostmodel Intel64).\n";
static const char USAGE_JOIN_WORKER_TAIL[] = "example: join -c 8 -m 16384\n";
static const char USAGE_JOIN_WORKER_TAIL[]
= "example: join -c 8 -m 17179869184 (16 GB in bytes)\n";
static const char USAGE_JOBSTART_HEAD[] =
"jstart -j <jobid> [-p <pid>]\n"
@ -2055,9 +2049,19 @@ int cmd_hello(int argc, char* argv[])
while ((opt = getopt_long(argc, argv, optstring, long_opts, NULL)) != -1) {
switch (opt) {
case 'c':
count = atoi(optarg);
case 'c': {
long n;
/* bug-1710: 0, negative or non-numeric sent nothing and still
* reported success */
if (cmdutil_parse_long(optarg, 1, INT_MAX, &n) < 0) {
fprintf(stderr, "hello: -c expects a count >= 1, got '%s'\n",
optarg);
return -EINVAL;
}
count = (int)n;
break;
}
case 'm':
msg = optarg;
break;
@ -2274,8 +2278,7 @@ int cmd_reslist(int argc, char* argv[])
free(call);
if (rc < 0) {
SET_ERROR(-rc);
perror("xce_reslist");
cmdutil_rpc_error("xce_reslist", rc);
return rc;
}
@ -2417,8 +2420,7 @@ int cmd_workerlist(int argc, char* argv[])
free(call);
if (rc < 0) {
SET_ERROR(-rc);
perror("xce_workerlist");
cmdutil_rpc_error("xce_workerlist", rc);
return rc;
}
@ -2438,16 +2440,18 @@ int cmd_workerlist(int argc, char* argv[])
for (i = 0; i < rc; i++) {
cJSON* row = listfmt_json_new_row(array);
workerlist_json_row(row, workers[i]->name,
worker_status_str(workers[i]->status),
workers[i]->attrs->hosttype,
workers[i]->attrs->hostmodel,
workers[i]->attrs->cpufactor,
workers[i]->attrs->maxcpus,
workers[i]->attrs->maxmem,
workers[i]->attrs->maxswap,
workers[i]->attrs->maxtmp,
workers[i]->attrs->ndisks);
/* bug-1777: effective values, as the scheduler sees them */
workerlist_json_row(
row, workers[i]->name, worker_status_str(workers[i]->status),
workers[i]->attrs->hosttype, workers[i]->attrs->hostmodel,
workers[i]->attrs->cpufactor, workers[i]->attrs->maxcpus,
worker_effective_attr(workers[i], "maxmem",
workers[i]->attrs->maxmem),
worker_effective_attr(workers[i], "maxswap",
workers[i]->attrs->maxswap),
worker_effective_attr(workers[i], "maxtmp",
workers[i]->attrs->maxtmp),
workers[i]->attrs->ndisks);
}
listfmt_json_finish(array);
@ -2468,12 +2472,17 @@ int cmd_workerlist(int argc, char* argv[])
char maxswap_buf[64];
char maxtmp_buf[64];
format_bytes_human(workers[i]->attrs->maxmem, maxmem_buf,
sizeof(maxmem_buf));
format_bytes_human(workers[i]->attrs->maxswap, maxswap_buf,
sizeof(maxswap_buf));
format_bytes_human(workers[i]->attrs->maxtmp, maxtmp_buf,
sizeof(maxtmp_buf));
/* bug-1777: the effective value (a hosts.yaml override included),
* which is what the scheduler uses, not the join-time attrs */
format_bytes_human(worker_effective_attr(workers[i], "maxmem",
workers[i]->attrs->maxmem),
maxmem_buf, sizeof(maxmem_buf));
format_bytes_human(worker_effective_attr(workers[i], "maxswap",
workers[i]->attrs->maxswap),
maxswap_buf, sizeof(maxswap_buf));
format_bytes_human(worker_effective_attr(workers[i], "maxtmp",
workers[i]->attrs->maxtmp),
maxtmp_buf, sizeof(maxtmp_buf));
printf("%-20s%-8s%-25s%-25s%12f%12d%12s%12s%12s%12d\n",
workers[i]->name, worker_status_str(workers[i]->status),
@ -2568,7 +2577,7 @@ static int joblist_query(const char* find_jobid, int* contains, int print_table,
* and EXPECTS failure until a primary serves) suppress the per-attempt
* error print so a normal failover wait does not spam the console. */
if (!quiet) {
perror("xce_joblist");
cmdutil_rpc_error("xce_joblist", rc);
}
return rc;
}
@ -3289,6 +3298,9 @@ static int jobinfo_thunk(void* ctxp)
int cmd_jobinfo(int argc, char* argv[])
{
int opt;
int ngiven = 0; /* -j values given */
int nfound = 0; /* lookups that succeeded */
int lastrc = -EINVAL;
struct option long_opts[JOBINFO_NFLAGS + 1];
char optstring[16];
@ -3311,9 +3323,16 @@ int cmd_jobinfo(int argc, char* argv[])
switch (opt) {
case 'j':
ngiven++;
parse_jobid(&jobid, optarg);
if (jobid == 0) {
/* bug-1709: a malformed jobid used to be skipped silently */
fprintf(stderr, "jobinfo: invalid jobid '%s'\n", optarg);
lastrc = -EINVAL;
continue;
}
if (jobid > 0) {
{
jobinfo_call_t* call
= (jobinfo_call_t*)calloc(1, sizeof(*call));
@ -3341,12 +3360,14 @@ int cmd_jobinfo(int argc, char* argv[])
free(call);
if (rc < 0) {
perror("xcecli_jobinfo");
cmdutil_rpc_error("xcecli_jobinfo", rc);
lastrc = rc;
continue;
}
print_jobinfo(job);
free_Job_ptr(&job);
nfound++;
}
break;
@ -3356,7 +3377,14 @@ int cmd_jobinfo(int argc, char* argv[])
"jobinfo -j <jobid> [-j <jobid> ...]");
}
}
return 0;
/* bug-1709: fail when there was nothing to look up or every lookup
* failed, so a script sees it in the exit code */
if (ngiven == 0) {
fprintf(stderr, "jobinfo: no jobid given (use -j <jobid>)\n");
return -EINVAL;
}
return (nfound > 0) ? 0 : lastrc;
}
static int fill_job_alloc(xce_job_alloc_t* jalloc, int nhosts,
@ -3457,19 +3485,28 @@ int cmd_jobrun(int argc, char* argv[])
return -1;
}
if (strchr(optarg, (int)':') != NULL) {
char* p;
char buf[256];
{
char buf[256];
int prc = cmdutil_split_host_tasks(optarg, buf, sizeof(buf),
&count);
p = strchr(optarg, (int)':');
count = atoi(p + 1);
memset(buf, 0x0, sizeof(buf));
strncpy(buf, optarg, (int)(p - optarg));
/* bug-1702: the host segment used to be strncpy'd into buf
* with no length check */
if (prc == -ENAMETOOLONG) {
fprintf(stderr,
"jobrun: host name in -m '%.32s...' is longer "
"than %zu characters\n",
optarg, sizeof(buf) - 1);
return -EINVAL;
}
if (prc < 0) {
fprintf(stderr,
"jobrun: -m expects <host>[:<ntasks>] with "
"ntasks >= 1, got '%s'\n",
optarg);
return -EINVAL;
}
host = strdup(buf);
} else {
host = strdup(optarg);
count = 1;
}
hosts[nhosts].host = host;
@ -3523,8 +3560,7 @@ int cmd_jobrun(int argc, char* argv[])
}
if (rc < 0) {
SET_ERROR(-rc);
perror("xce_jobdispatch");
cmdutil_rpc_error("xce_jobdispatch", rc);
return rc;
}
@ -3536,29 +3572,6 @@ int cmd_jobrun(int argc, char* argv[])
return 0;
}
static int build_job_envs(xce_job_submit_t* submit)
{
exdata_t exdata = { .sub_env_vars = { .PATH = getenv("PATH") } };
s2j_create_json_obj(json_exdata);
s2j_create_json_obj(json_envs);
s2j_json_set_basic_element(json_envs, &(exdata.sub_env_vars), var_string,
PATH);
cJSON_AddItemToObject(json_exdata, "sub_env_vars", json_envs);
submit->submit->start_info->exdata = cJSON_Print(json_exdata);
cJSON_Delete(json_exdata);
return 0;
}
static int build_job_exdata(xce_job_submit_t* submit)
{
build_job_envs(submit);
return 0;
}
static const flagspec_t jobsub_flags[] = {
{ 'u', "user", "user", required_argument, 1, NULL },
{ 'q', "queue", "queue", required_argument, 1, NULL },
@ -3639,27 +3652,19 @@ int cmd_jobsub(int argc, char* argv[])
submit.start_info->cmdline = strdup(optarg);
break;
case 'x':
FREEUP(submit.start_info->exdata);
submit.start_info->exdata = strdup(optarg);
break;
case 'n':
if (strchr(optarg, (int)',') != NULL) {
char buf[64];
char* p = strchr(optarg, (int)',');
max = atoi(p + 1);
memset(buf, 0x0, sizeof(buf));
strncpy(buf, optarg, (int)(p - optarg));
min = atoi(buf);
if (max == 0) {
fprintf(stderr,
"Incorrect setting of max tasks count, set to "
"same as min task value\n");
max = min;
}
} else {
min = atoi(optarg);
max = min;
/* bug-1703: the min part used to be strncpy'd into a 64-byte
* buffer with no length check */
if (cmdutil_parse_ntasks(optarg, &min, &max) < 0) {
fprintf(stderr,
"jobsub: -n expects <min>[,<max>] with 1 <= min <= "
"max, got '%.40s'\n",
optarg);
rc = -EINVAL;
goto cleanup;
}
break;
case 'a':
@ -3677,12 +3682,6 @@ int cmd_jobsub(int argc, char* argv[])
}
}
if (min == 0) {
fprintf(stderr,
"Incorrect setting of min task count, set to 1 as default\n");
min = 1;
}
if (NULLSTR(submit.user)) {
fprintf(stderr, "you need to specify job user with -u\n");
rc = -EINVAL;
@ -3734,7 +3733,12 @@ int cmd_jobsub(int argc, char* argv[])
submit.start_info->execcwd = strdup("/tmp");
}
build_job_exdata(&jobsubmit);
/* bug-1705: only when -x did not give one -- this used to overwrite
* (and leak) -x's exdata unconditionally */
if (cmdutil_default_exdata(&submit.start_info->exdata, getenv("PATH")) < 0) {
rc = -ENOMEM;
goto cleanup;
}
jobsubmit.array_size = arsize;
@ -3781,6 +3785,9 @@ cleanup:
FREEUP(submit.resreq);
FREEUP(submit.start_info->cmdline);
FREEUP(submit.start_info->exdata);
FREEUP(submit.start_info->execuser);
FREEUP(submit.start_info->exechome);
FREEUP(submit.start_info->execcwd);
FREEUP(submit.start_info);
/* Report success only on a clean ack that echoes back the EXACT JobId we
@ -4115,9 +4122,20 @@ int cmd_jobkill(int argc, char* argv[])
while ((opt = getopt_long(argc, argv, optstring, long_opts, NULL))
!= -1) {
switch (opt) {
case 's':
signo = atoi(optarg);
case 's': {
long n;
/* bug-1708: a non-numeric value became signal 0 and the kill
* was still sent and reported */
if (cmdutil_parse_long(optarg, 1, NSIG - 1, &n) < 0) {
fprintf(stderr,
"jobkill: -s expects a signal number 1-%d, got '%s'\n",
NSIG - 1, optarg);
return -EINVAL;
}
signo = (int)n;
break;
}
case 'j':
parse_jobid(&jobid, optarg);
break;
@ -4185,7 +4203,7 @@ int cmd_jobkill(int argc, char* argv[])
}
if (rc < 0) {
perror("xcecli_jobctrl");
cmdutil_rpc_error("xcecli_jobctrl", rc);
return rc;
}
@ -4302,7 +4320,7 @@ int cmd_jobckpt(int argc, char* argv[])
}
if (rc < 0) {
perror("xcecli_jobctrl");
cmdutil_rpc_error("xcecli_jobctrl", rc);
return rc;
}
@ -4385,6 +4403,7 @@ int cmd_qctrl(int argc, char* argv[])
action = QUEUE_CLOSE;
break;
case 'q':
FREEUP(queue);
queue = strdup(optarg);
break;
case 'y':
@ -4437,6 +4456,7 @@ int cmd_qctrl(int argc, char* argv[])
qctrl_call_t* call = (qctrl_call_t*)calloc(1, sizeof(*call));
if (call == NULL) {
free(queue);
return -ENOMEM;
}
@ -4457,7 +4477,8 @@ int cmd_qctrl(int argc, char* argv[])
}
if (rc < 0) {
perror("xcecli_queuectrl");
cmdutil_rpc_error("xcecli_queuectrl", rc);
free(queue); /* bug-1797: leaked on this path */
return rc;
}
@ -4533,6 +4554,7 @@ int cmd_hostctrl(int argc, char* argv[])
action = HOST_CLOSE;
break;
case 'h':
FREEUP(hostname);
hostname = strdup(optarg);
break;
case 'y':
@ -4599,7 +4621,7 @@ int cmd_hostctrl(int argc, char* argv[])
}
if (rc < 0) {
perror("xcecli_hostctrl");
cmdutil_rpc_error("xcecli_hostctrl", rc);
free(hostname);
return rc;
}
@ -4955,8 +4977,7 @@ int cmd_runtasks(int argc, char* argv[])
}
if (rc < 0) {
errno = -rc;
perror("xcecli_runtask");
cmdutil_rpc_error("xcecli_runtask", rc);
} else {
printf("run task ok\n");
}
@ -5108,7 +5129,7 @@ int cmd_setpend(int argc, char* argv[])
}
if (rc < 0) {
perror("xcecli_set_pend_reason");
cmdutil_rpc_error("xcecli_set_pend_reason", rc);
erase_job_preason(&jreasons);
return rc;
}
@ -5174,8 +5195,7 @@ int cmd_join_worker(int argc, char* argv[])
}
if (rc < 0) {
SET_ERROR(-rc);
perror("xce_join");
cmdutil_rpc_error("xce_join", rc);
return rc;
}
@ -5291,8 +5311,7 @@ int cmd_jobstart(int argc, char* argv[])
}
if (rc < 0) {
SET_ERROR(-rc);
perror("xcecli_jobstart");
cmdutil_rpc_error("xcecli_jobstart", rc);
return rc;
}
@ -5394,8 +5413,7 @@ int cmd_jobfinish(int argc, char* argv[])
}
if (rc < 0) {
SET_ERROR(-rc);
perror("xcecli_jobfinish");
cmdutil_rpc_error("xcecli_jobfinish", rc);
return rc;
}
@ -5504,8 +5522,7 @@ int cmd_jobfail(int argc, char* argv[])
}
if (rc < 0) {
SET_ERROR(-rc);
perror("xcecli_jobfail");
cmdutil_rpc_error("xcecli_jobfail", rc);
return rc;
}
@ -5514,95 +5531,6 @@ int cmd_jobfail(int argc, char* argv[])
return 0;
}
/* set_resval() -- thin wrapper (resource-value-verification task 2.1).
*
* Type resolution (name -> Value_u_which, including the unrecognized-name
* numeric-default fallback and `-t` override) now lives in
* resval_typeresolve.c/resval_resolve_type(), shared with the standalone
* unit tests. This function keeps only what still needs REPL state: the
* xce_val_t allocation/parsing per wire type, and `.host` tagging via
* get_setting("myname") (unchanged from before this refactor).
*
* `type_override`: RESVAL_TYPE_UNKNOWN for "no override" (the common case),
* or an explicit Value_u_which tag from cmd_resupdate()'s new `-t` flag.
* Ignored by resval_resolve_type() for any already-known resource name.
*/
static int set_resval(
xce_resval_t* vals, int idx, char* name, char* val, int type_override)
{
int valtype;
int rc = -1;
if (name == NULL) {
return 0;
}
valtype = resval_resolve_type(name, type_override);
if (valtype == RESVAL_TYPE_UNKNOWN) {
/* Only reachable if resval_resolve_type() itself ever declines to
* resolve a non-NULL name -- not possible today (it always falls
* back to Value_u_fvalue or the override), kept as a defensive
* guard against a future change to that function's contract. */
return rc;
}
rc = 1;
vals[idx].val = (xce_val_t*)calloc(1, sizeof(xce_val_t));
switch (valtype) {
case Value_u_fvalue:
vals[idx].name = name;
vals[idx].val->which = valtype;
if (val == NULL) {
vals[idx].val->u.fvalue = 0.0;
} else {
vals[idx].val->u.fvalue = atof(val);
}
break;
case Value_u_ivalue32:
vals[idx].name = name;
vals[idx].val->which = valtype;
if (val == NULL) {
vals[idx].val->u.ivalue32 = 0;
} else {
vals[idx].val->u.ivalue32 = atoi(val);
}
break;
case Value_u_ivalue64:
vals[idx].name = name;
vals[idx].val->which = valtype;
if (val == NULL) {
vals[idx].val->u.ivalue64 = 0;
} else {
vals[idx].val->u.ivalue64 = atol(val);
}
break;
case Value_u_bvalue:
vals[idx].name = name;
vals[idx].val->which = valtype;
if (val == NULL) {
vals[idx].val->u.bvalue = 0;
} else {
if (strcasecmp(val, "t") == 0) {
vals[idx].val->u.bvalue = 1;
} else {
vals[idx].val->u.bvalue = 0;
}
}
break;
}
/* NOTE (pre-existing, unchanged): Value_u_svalue (type/model, and now
* an unrecognized name given `-t string`) has no case here -- the
* switch falls through with vals[idx].name/.val->which left unset.
* This is set_resval()'s existing behavior for type/model today;
* Requirement 1.4 requires preserving it byte-for-byte, so it is
* intentionally not fixed by this task. See CONCERNS in this task's
* status report. */
vals[idx].host = get_setting("myname");
return rc;
}
/*
* resource-value-verification task 2.2: the controller's own update RPC
* reply can never distinguish per-name accept/reject -- update_resval_list()
@ -5668,9 +5596,6 @@ static int resupdate_report_one(char* name, char* host, xce_val_t* sent)
int ever_found = 0;
if (name == NULL) {
/* set_resval()'s Value_u_svalue fallback path leaves .name unset
* (pre-existing gap predating this task -- see set_resval()'s own
* comment above); nothing usable to query or report here. */
return 0;
}
@ -5743,6 +5668,37 @@ static int resupdate_report_one(char* name, char* host, xce_val_t* sent)
return 0;
}
/* Append one -n/-v pair to cmd_resupdate's vals[] (16 entries), taking
* ownership of `name`. A missing value gets the type's default with a
* warning. Returns 0, or -E2BIG when vals[] is full (the old code had no
* bound) or -ENOMEM; on error `name` is freed. */
static int resupdate_add(xce_resval_t* vals, int* nres, char* name,
const char* val, int type_override)
{
int rc;
if (*nres >= 16) {
fprintf(stderr, "resupdate: at most 16 resources per call\n");
free(name);
return -E2BIG;
}
if (val == NULL) {
fprintf(stderr,
"res `%s' has no value specified, default will be used\n",
name);
}
/* bug-1574: resval_fill() sets name/value for string resources
* (type, model) too; set_resval() left them uninitialized */
rc = resval_fill(&vals[*nres], name, val, type_override,
get_setting("myname"));
if (rc < 0) {
free(name);
return rc;
}
(*nres)++;
return 0;
}
static const flagspec_t resupdate_flags[] = {
{ 'n', "resname", "resname", required_argument, 1, NULL },
{ 'v', "value", "value", required_argument, 1, NULL },
@ -5776,8 +5732,6 @@ int cmd_resupdate(int argc, char* argv[])
#define MAX_N_RES (16)
int nres;
xce_resval_t vals[MAX_N_RES];
xce_resval_t* valptrs[MAX_N_RES];
char* subnames[MAX_N_RES];
char* resname = NULL;
char* resval_str = NULL;
int type_override = RESVAL_TYPE_UNKNOWN;
@ -5789,6 +5743,7 @@ int cmd_resupdate(int argc, char* argv[])
optind = 0;
opterr = 0;
nres = 0;
memset(vals, 0, sizeof(vals));
if (flagspec_build_opts(resupdate_flags, RESUPDATE_NFLAGS, long_opts,
optstring, sizeof(optstring))
@ -5798,46 +5753,36 @@ int cmd_resupdate(int argc, char* argv[])
return -EINVAL;
}
/* -t is deferred: it applies to the pair most recently opened by -n
* (design.md's documented order is -n/-v/[-t]), so a pending pair's
* set_resval() call is deferred until the NEXT -n or end-of-args --
* not fired immediately on -v as before -- so a trailing -t still has
* a chance to be seen for that same pair. */
while ((opt = getopt_long(argc, argv, optstring, long_opts, NULL))
!= -1) {
switch (opt) {
case 'n':
if (resname != NULL) {
if (resval_str == NULL) {
fprintf(
stderr,
"res `%s' has no value specified, default will be used\n",
resname);
}
rc = set_resval(vals, nres, resname, resval_str, type_override);
if (rc >= 0) {
valptrs[nres] = &vals[nres];
/* Tracked independently of vals[nres].name: set_resval()
* leaves .name unset on the Value_u_svalue fallback path
* (see its own comment), but `resname` here is always the
* exact name that was actually submitted for this pair. */
subnames[nres] = resname;
nres++;
}
rc = resupdate_add(vals, &nres, resname, resval_str,
type_override);
resname = NULL;
resval_str = NULL;
type_override = RESVAL_TYPE_UNKNOWN;
if (rc < 0) {
goto out;
}
}
resname = strdup(optarg);
break;
case 'v':
if (resname != NULL && resval_str == NULL) {
resval_str = optarg;
} else {
if (resname == NULL) {
fprintf(stderr,
"ignore res value `%s' as no res name found\n",
optarg);
} else if (resval_str != NULL) {
/* bug-1712: this case used to print "no res name found" */
fprintf(stderr,
"ignoring duplicate value `%s' for res `%s' (it "
"already has value `%s')\n",
optarg, resname, resval_str);
} else {
resval_str = optarg;
}
break;
case 't':
@ -5859,31 +5804,38 @@ int cmd_resupdate(int argc, char* argv[])
}
break;
case '?':
return flagspec_reject(
rc = flagspec_reject(
"resupdate", resupdate_flags, RESUPDATE_NFLAGS,
argv[optind - 1],
"resupdate -n <resname> -v <value> [-t <numeric|string>] "
"[-n <resname> -v <value> [-t <numeric|string>] ...]");
FREEUP(resname);
goto out;
}
}
/* Flush the last pending pair. Matches the pre-existing behavior of a
* trailing -n with no -v ever seen (silently dropped, same as before
* this change) -- only flush here when a value was actually supplied. */
if (resname != NULL && resval_str != NULL) {
rc = set_resval(vals, nres, resname, resval_str, type_override);
if (rc >= 0) {
valptrs[nres] = &vals[nres];
subnames[nres] = resname;
nres++;
/* bug-1707: the trailing -n was dropped without a word when it had no
* -v; it now gets the same warning and default as a mid-list one */
if (resname != NULL) {
rc = resupdate_add(vals, &nres, resname, resval_str, type_override);
if (rc < 0) {
goto out;
}
}
if (nres == 0) {
fprintf(stderr,
"resupdate: no resource given (use -n <resname> -v <value>)\n");
rc = -EINVAL;
goto out;
}
{
resupdate_call_t* call = (resupdate_call_t*)calloc(1, sizeof(*call));
if (call == NULL) {
return -ENOMEM;
rc = -ENOMEM;
goto out;
}
call->conn = active_ctx->ctx;
@ -5898,6 +5850,8 @@ int cmd_resupdate(int argc, char* argv[])
rc = xcetest_run_interruptible(resupdate_thunk, call);
if (rc == -EINTR) {
/* the still-running worker thread owns `call`, whose vals[]
* point at the same names/values -- leave them allocated */
errno = EINTR;
perror("resupdate");
return rc;
@ -5909,26 +5863,28 @@ int cmd_resupdate(int argc, char* argv[])
printf("resource update ok!\n");
/* The update RPC's own reply can't carry per-name accept/reject
* (see the comment above resupdate_report_one()) -- verify each
* submitted pair individually via a follow-up query instead. An
* interrupt during one of THOSE follow-up queries stops this
* verification pass immediately and is reported as this command's
* own failure (Requirement 1.3), rather than silently skipping the
* remaining pairs' verification. */
for (i = 0; i < nres; i++) {
int report_rc
= resupdate_report_one(subnames[i], vals[i].host, vals[i].val);
= resupdate_report_one(vals[i].name, vals[i].host, vals[i].val);
if (report_rc == -EINTR) {
return report_rc;
return report_rc; /* same ownership rule as above */
}
}
rc = 0;
} else {
SET_ERROR(-rc);
perror("xcecli_update_resval");
/* the RPC's failure used to be printed and then reported as
* success (return 0) */
cmdutil_rpc_error("xcecli_update_resval", rc);
}
return 0;
out:
for (int i = 0; i < nres; i++) {
free(vals[i].name);
resval_fill_release(&vals[i]);
}
return rc;
#undef MAX_N_RES
}
typedef struct {
@ -5967,8 +5923,7 @@ int cmd_join_sched(int argc, char* argv[])
}
if (rc < 0) {
SET_ERROR(-rc);
perror("xce_join");
cmdutil_rpc_error("xce_join", rc);
return rc;
}
@ -6087,8 +6042,7 @@ int cmd_getconf(int argc, char* argv[])
free(call);
if (rc < 0) {
SET_ERROR(-rc);
perror("xce_clusterconf");
cmdutil_rpc_error("xce_clusterconf", rc);
return rc;
}
@ -6860,7 +6814,7 @@ static int worker_house_keeping(void* arg)
rc = xcecli_heartbeat(ctx->ctx, HEARTBEAT_INTERVAL);
if (rc < 0) {
perror("xcecli_heartbeat");
cmdutil_rpc_error("xcecli_heartbeat", rc);
}
return rc;