[collector]bugfix recurring tasks caused by priority processing excep…

…tion (#663)
apache · Feb 21, 2023 · 0bbeb01 · 0bbeb01
1 parent 708aeff
commit 0bbeb01
Show file tree

Hide file tree

Showing 4 changed files with 10 additions and 12 deletions.
diff --git a/collector/src/main/java/com/usthe/collector/dispatch/CommonDispatcher.java b/collector/src/main/java/com/usthe/collector/dispatch/CommonDispatcher.java
@@ -219,8 +219,9 @@ public void dispatchCollectData(Timeout timeout, Metrics metrics, CollectRep.Met
             // If the availability collection fails, the next indicator group scheduling will be cancelled and the next round of scheduling will be entered directly.
             // 若metricsSet为null表示执行完成
             // 或判断采集指标组是否优先级为0，即为可用性采集指标组 若可用性采集失败 则取消后面的指标组调度直接进入下一轮调度
-            if (metricsSet == null
-                    || (metrics.getPriority() == (byte) 0 && metricsData.getCode() != CollectRep.Code.SUCCESS)) {
+            boolean isAvailableCollectFailed = metricsSet != null && !metricsSet.isEmpty()
+                    && metrics.getPriority() == (byte) 0 && metricsData.getCode() != CollectRep.Code.SUCCESS;
+            if (metricsSet == null || isAvailableCollectFailed) {
                 // The collection and execution of all index groups of this job are completed.
                 // The periodic task pushes the task to the time wheel again.
                 // First, determine the execution time of the task and the task collection interval.

diff --git a/common/src/main/java/com/usthe/common/entity/job/Job.java b/common/src/main/java/com/usthe/common/entity/job/Job.java
@@ -130,7 +130,7 @@ public class Job {
      * 127 - lastPriorMetrics
      */
     @JsonIgnore
-    private transient List<Set<Metrics>> priorMetrics;
+    private transient LinkedList<Set<Metrics>> priorMetrics;
 
     /**
      * collector use - Temporarily store one-time task indicator group response data
@@ -194,11 +194,10 @@ public synchronized Set<Metrics> getNextCollectMetrics(Metrics metrics, boolean
         if (priorMetrics == null || priorMetrics.isEmpty()) {
             return null;
         }
-        Set<Metrics> metricsSet = priorMetrics.get(0);
+        Set<Metrics> metricsSet = priorMetrics.peek();
         if (first) {
             if (metricsSet.isEmpty()) {
                 log.error("metrics must has one [availability] metrics at least.");
-
             }
             return metricsSet;
         }
@@ -211,11 +210,11 @@ public synchronized Set<Metrics> getNextCollectMetrics(Metrics metrics, boolean
                     id, monitorId, app, metrics.getName());
         }
         if (metricsSet.isEmpty()) {
-            priorMetrics.remove(0);
-            if (priorMetrics.size() == 0) {
+            priorMetrics.poll();
+            if (priorMetrics.isEmpty()) {
                 return null;
             }
-            return priorMetrics.get(0);
+            return priorMetrics.peek();
         } else {
             return Collections.emptySet();
         }

diff --git a/manager/src/main/resources/define/app/app-nacos.yml b/manager/src/main/resources/define/app/app-nacos.yml
@@ -33,7 +33,7 @@ metrics:
       parseType: jsonPath
       parseScript: $.measurements[*]
   - name: memory
-    priority: 0
+    priority: 1
     fields:
       - field: jvm_memory_used
         type: 0

diff --git a/manager/src/main/resources/define/app/app-zookeeper.yml b/manager/src/main/resources/define/app/app-zookeeper.yml
@@ -102,7 +102,7 @@ metrics:
       parseType: netcat
 
   - name: stats
-    priority: 0
+    priority: 1
     # 指标组中的具体监控指标
     fields:
       # 指标信息 包括 field名称   type字段类型:0-number数字,1-string字符串   instance是否为实例主键   unit:指标单位
@@ -163,8 +163,6 @@ metrics:
       - field: zk_min_latency
         type: 0
         unit: ms
-
-
     # 协议
     protocol: ssh
     ssh: