Question

我想使用模板来使警报更易于阅读。基本上，我有这样的警报：

alertname =“ service_1的dvvmclust1上部署了3个容器（每个节点预期1个）”
  ... 08:53:46，2018-11-12
  ...标签
  .......严重性=“严重”
  ....... node_name =“ dvvmclust1”
  ....... monitor =“ docker-swarm-host”
  ....... com_docker_swarm_service_name =“ qgis”

希望通过这种方式定义Il：

#include <stdio.h>

int main(void) {
    int A[2][3] = {};
    int B[3][2] = {};
    int i,j;
    for(i=0; i < 2;i++) {
        for(j=0; j < 3;j++) {
            scanf(" %d\n", &A[i][j]);
            getchar();
        }
    }
    return 0;
}

我想我知道只要我不使用“ relabel_configs”这样的方式，这些标签就不会在“ .labels”表中显示：

alert: ${{ labels.com_docker_swarm_service_name }}  has ${{ value }} containers deployed over ${{ labels.node_name }} (1 expected per node)
expr: whatever > 1
for: 30m
labels:
    severity: critical
annotations:
  description:  ${{ labels.com_docker_swarm_service_name }}  has ${{ value }} containers deployed over ${{ labels.node_name }}. You are not expected to have more that 1 at once per node. Please consider redeploying the service among the available nodes or restarting failed nodes.
  summary: ${{ labels.com_docker_swarm_service_name }}   has ${{ value }} containers deployed over ${{ labels.node_name }} (1 expected per node)

尽管，当我从Prometheus API中查看服务发现时，虽然从（ eg < / em>）cAdvisor指标：

relabel_configs - source_labels: [__com_docker_swarm_service_name__] target_label: __com_docker_swarm_service_name

（我从docker-compose文件中添加了一些标签）

所以我的问题是：如何通过cAdvisor或node_exporter从容器中驱动标签（及其“相关名称！”），例如“ serviceName”和“ nodeName”，直到在prometheus中定义了非常警报的状态？ / p>

将标签从容器/探针传送到普罗米修斯警报

0 个答案: