jayson.yang

Vmalert externalLabels override the same labels selected from datasource

3

Following is the vmalert config about external labels. external_labels: region: cn-sh env: production grafana_datasource: Prometheus-CM We have a vmagent which scrapes some various targets in multiple regions. And some of...

Add podAntiAffinity when vmagent in shard mode

1

We have three k8s pod for deploying vm components. Then we config a vmagent cr with 2 shards, every shard with 2 replicas. When I add podAntiAffinity config for vmagent...

All targets display in web UI when vmagent is in shard model

2

When vmagent is in shard model, every shard only display a part of all targets. If I want find a specific target in UI, I would have to flush the...

api/v1/series returned labels beyond range time

4

After upgraded vmcluster to v1.79.2, the api `api/v1/series` returned labels beyond the time range. `curl http://10.1.1.5:8481/select/1/prometheus/api/v1/series?match%5B%5D=trs_engine_up%7Bjob%3D~%22trs-engine%22%7D&start=1663059757&end=1663063357` When using the command above to query specific timeseries, the vmselect return some timeseries...

vmalert external.alert.source configuration: `-external.alert.source="explore?orgId=1&left=[\"now-1h\",\"now\",\"Prometheus-DEV\",{\"expr\": \"{{$expr|quotesEscape|crlfEscape|queryEscape}}\"},{\"mode\":\"Metrics\"},{\"ui\":[true,true,true,\"none\"]}]" ` alert rule expr: `ingress_nginx_request_qps{status=~"5\\d{2}"} / ignoring(status) group_left() sum without (status) (ingress_nginx_request_qps) * 100 > 1` encoded url: `https://grafana.example.com/explore?orgId=1&left=%7B%22datasource%22:%22Prometheus-DEV%22,%22queries%22:%5B%7B%22expr%22:%22ingress_nginx_request_qps%7Bstatus%3D~%5C%225%5C%5Cd%7B2%7D%5C%22%7D%20%2F%20ignoring(status)%20group_left()%20sum%20without%20(status)%20(ingress_nginx_request_qps)%20*%20100%5Cn%20%20%3E%201%5Cn%22,%22refId%22:%22A%22%7D%5D,%22range%22:%7B%22from%22:%22now-1h%22,%22to%22:%22now%22%7D%7D` Grafana reported a syntax error: The...

vmalert evaluate incorrect alert when using absent

7

**Describe the bug** When we use absent at a slow pulled metric for monitoring the metric missing, the vmalert triggered an incorrect alert. alert rule: ``` expr: absent(Mongodb_CPUUtilization) == 1...

VMAgent Deployment add -promscrape.cluster.name to distinct pods

3

When I deploy a vmagent deployment with two replicas, two vmagent pods will scrape same targets and remote write to same vmcluster with same tenant. If vmagents scraped different value...