operator
operator copied to clipboard
Kubernetes operator for Victoria Metrics
Currently operator creates one statefulset for all vmstorage nodes and scales it accordingly. This approach is good until something bad happens to one of storage nodes. For example: It is...
I've deployed a `VMCluster` and set the cpu limit to `1000m`. https://github.com/uhthomas/automata/blob/15654a5e2dea23101bf7a358e1f07e76eda5dab8/k8s/unwind/vm/vm_cluster_list.cue#L39 The operator changes this to "1", which breaks my deployment. It should just leave the value alone. ```sh...
Related: #702
In our szenario we deploy all components with the vmoperator. There are some VM components which are not deployable with the vmoperator, which complicates the deployment. One of these components...
It would be nice for [VMServiceScrape](https://docs.victoriametrics.com/operator/api.html#vmservicescrapespec)'s selector, if it can use meta labels defined in [kubernetes_sd_configs](https://docs.victoriametrics.com/sd_configs.html#kubernetes_sd_configs), i.e. `__meta_kubernetes_service_annotation_`. So we can have something like ``` apiVersion: operator.victoriametrics.com/v1beta1 kind: VMServiceScrape metadata:...
It would be great to have an option to define Bearer token or Basic Auth configuration for targets. For instance: ```yaml kind: VMUser spec targetRefs: paths: - static: - ursl:...
Hello! We are using VMOperator + VM CR recourses, and manage them via terraform. We modify Cluster CR inside terraform configuration and apply them via [terraform's kubernetes provider](https://github.com/hashicorp/terraform-provider-kubernetes). In some...
The CRDs do not define patchMergeKey and patchStrategy for both group names and rule alert fields. https://github.com/VictoriaMetrics/operator/blob/master/api/v1beta1/vmrule_types.go The lack of merge keys makes it hard to adjust the rules and...
When in [cluster mode](https://docs.victoriametrics.com/vmagent.html#scraping-big-number-of-targets), vmagent splits the list of targets among the configured shards. One shard starts to scrape only 1/N fraction of all configured targets, and so it shows...
as topic describe