Hopefully fix the buildfarm. Add some debug level 10 messages
[jra/samba/.git] / source3 / printing / printing.c
index 83b5ac8b432209fbe7442f20caee3c4e6e2a6aec..ce98792096a7317e6a7814d72e71ea7fc13d414c 100644 (file)
@@ -2313,11 +2313,15 @@ static bool allocate_print_jobid(struct tdb_print_db *pdb, int snum, const char
                        if (tdb_error(pdb->tdb) != TDB_ERR_NOEXIST) {
                                DEBUG(0, ("allocate_print_jobid: failed to fetch INFO/nextjob for print queue %s\n",
                                        sharename));
+                               tdb_unlock_bystring(pdb->tdb, "INFO/nextjob");
                                return False;
                        }
+                       DEBUG(10,("allocate_print_jobid: no existing jobid in %s\n", sharename));
                        jobid = 0;
                }
 
+               DEBUG(10,("allocate_print_jobid: read jobid %u from %s\n", jobid, sharename));
+
                jobid = NEXT_JOBID(jobid);
 
                if (tdb_store_int32(pdb->tdb, "INFO/nextjob", jobid)==-1) {
@@ -2329,8 +2333,10 @@ static bool allocate_print_jobid(struct tdb_print_db *pdb, int snum, const char
                /* We've finished with the INFO/nextjob lock. */
                tdb_unlock_bystring(pdb->tdb, "INFO/nextjob");
 
-               if (!print_job_exists(sharename, jobid))
+               if (!print_job_exists(sharename, jobid)) {
                        break;
+               }
+               DEBUG(10,("allocate_print_jobid: found jobid %u in %s\n", jobid, sharename));
        }
 
        if (i > 2) {