core: coldplug possible nop_job

chenglin130 · Jan Synacek · commit 8ee652948b02 · 2020-04-30T11:43:14.000+02:00
When a unit in a state INACTIVE or DEACTIVATING, JobType JOB_TRY_RESTART or JOB_TRY_RELOAD will be collapsed to JOB_NOP. And use u->nop_job instead of u->job. If a JOB_NOP job is going on with a waiting state, a parallel daemon-reload just install it during deserialization. Without a coldplug, the job will not be in m->run_queue, which results in a hung try-restart or try-reload process. Reproduce: 1. run systemctl try-restart test.servcie (inactive) repeatly in a terminal. 2. run systemctl daemon-reload repeatly in other terminals. After successful reproduce, systemctl list-jobs will list the hang job. Upsteam: systemd/systemd#13124 Resolves: #1829754
diff --git a/src/core/unit.c b/src/core/unit.c
@@ -2918,15 +2918,17 @@ static int unit_add_deserialized_job_coldplug(Unit *u) {
 
 int unit_coldplug(Unit *u, Hashmap *deferred_work) {
         int r;
+        Job *uj;
 
         assert(u);
 
         if (UNIT_VTABLE(u)->coldplug)
                 if ((r = UNIT_VTABLE(u)->coldplug(u, deferred_work)) < 0)
                         return r;
 
-        if (u->job) {
-                r = job_coldplug(u->job);
+        uj = u->job ?: u->nop_job;
+        if (uj) {
+                r = job_coldplug(uj);
                 if (r < 0)
                         return r;
         } else if (u->deserialized_job >= 0)
