job-manager: add plugin to assist posting manual job events

Problem: Sometimes a job becomes stuck in a given state, requiring
intervention by forcing an event to be posted to the job manager.
Currently, this requires building a custom jobtap plugin, which is
annoying and not sysadmin friendly.

Add a builtin jobtap plugin that will simply post a custom job
event defined in the RPC.

Author: grondo

src/modules/job-manager/Makefile.am

@@ -74,7 +74,8 @@ libjob_manager_la_SOURCES = \
 	plugins/begin-time.c \
 	plugins/update-duration.c \
 	plugins/validate-duration.c \
-	plugins/history.c
+	plugins/history.c \
+	plugins/post-event.c
 
 fluxinclude_HEADERS = \
 	jobtap.h

src/modules/job-manager/jobtap.c

@@ -51,6 +51,7 @@ extern int begin_time_plugin_init (flux_plugin_t *p);
 extern int validate_duration_plugin_init (flux_plugin_t *p);
 extern int update_duration_plugin_init (flux_plugin_t *p);
 extern int history_plugin_init (flux_plugin_t *p);
+extern int post_event_init (flux_plugin_t *p);
 
 struct jobtap_builtin {
     const char *name;
@@ -72,6 +73,7 @@ static struct jobtap_builtin jobtap_builtins [] = {
     { ".validate-duration", &validate_duration_plugin_init },
     { ".update-duration", &update_duration_plugin_init },
     { ".history", &history_plugin_init },
+    { ".post-event", &post_event_init },
     { 0 },
 };
 

src/modules/job-manager/plugins/post-event.c

@@ -0,0 +1,59 @@
+/************************************************************\
+ * Copyright 2024 Lawrence Livermore National Security, LLC
+ * (c.f. AUTHORS, NOTICE.LLNS, COPYING)
+ *
+ * This file is part of the Flux resource manager framework.
+ * For details, see https://github.com/flux-framework.
+ *
+ * SPDX-License-Identifier: LGPL-3.0
+\************************************************************/
+
+/* post-event.c - post manual events to job eventlog
+ */
+
+#if HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <jansson.h>
+#include <flux/core.h>
+#include <flux/jobtap.h>
+
+static void post_event_cb (flux_t *h,
+                           flux_msg_handler_t *mh,
+                           const flux_msg_t *msg,
+                           void *arg)
+{
+    flux_plugin_t *p = arg;
+    flux_jobid_t id;
+    const char *name;
+    json_t *context = NULL;
+
+    if (flux_msg_unpack (msg,
+                         "{s:I s:s s?o}",
+                         "id", &id,
+                         "name", &name,
+                         "context", &context) < 0)
+        goto error;
+    if (context) {
+        if (flux_jobtap_event_post_pack (p, id, name, "O", context) < 0)
+            goto error;
+    }
+    else if (flux_jobtap_event_post_pack (p, id, name, NULL) < 0)
+            goto error;
+    if (flux_respond (h, msg, NULL) < 0)
+        flux_log_error (h, "error responding to job-manager.post-event");
+error:
+    if (flux_respond_error (h, msg, errno, NULL) < 0)
+        flux_log_error (h, "error responding to job-manager.post-event");
+}
+
+
+int post_event_init (flux_plugin_t *p)
+{
+    if (flux_jobtap_service_register_ex (p, "post", 0, post_event_cb, p) < 0)
+        return -1;
+    return 0;
+}
+
+// vi:ts=4 sw=4 expandtab