Add a VictoriaLogs cluster and point logs-ingest at it #488
Reference in New Issue
Block a user
Delete Branch "benvin/vlcluster"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
The k8s log pipeline stores to ClickHouse via NATS+vector, while the VM estate ships journald to a separate puppet-managed VictoriaLogs cluster. Consolidating on VictoriaLogs in-cluster collapses the two paths, and the logs-ingest gateway has no clients yet so it can be repointed now, ahead of the puppet change.
logsat v1.52.0 (2 vlinsert, 2 vlselect, 3 vlstorage, 180d retention, 250Gi each on cephrbd-fast-delete)vlinsert-logs:9481Nothing is removed here; NATS, ClickHouse, vector, logarchiver and logviewer keep running until a follow-up drops them.
apps/base/logging/vlcluster.yaml:28,31—retentionPeriod: 180dis unreachable behind a 220GiB/node cap (660GiB total). This repo's own ingest sizing (apps/base/logging/kustomization.yaml:28-31, ~33 GiB/day compressed) puts 180d at ~5.8TiB, so retention is disk-bound at weeks and the~75% of the 180d estimatecomment does not reconcile with it → size the PVCs for the retention you actually want, or drop180dand state the real window.apps/base/grafana/grafanadatasource.yaml:41— operator v0.66.1 buildsvlselect-logsheadless (ClusterIP: None+publishNotReadyAddresses: true, unconditional inbuildVLSelectService), so Grafana's proxy pins a single pod and can dial a not-ready replica mid-rollout → add a ClusterIP service viaspec.vlselect.serviceSpecand point the datasource at it, or enablespec.requestsLoadBalancer.apps/base/logging/*(VLCluster + logs-ingest repoint) →logging-au-syd1, projectlogging.apps/base/grafana/grafanadatasource.yaml(datasource + plugin) →observability-grafana, projectobservability. It syncs independently, so as written it lands beforevlselect-logsexists, and the plugin install restarts Grafana.apps/base/logging/vlcluster.yaml:29-30— 2-line explanatory comment block; convention is minimal comments (the over-commented neighbours inapps/base/logging/are not the standard) → one line or none.apps/base/logging/vlcluster.yaml:37—cephrbd-fast-delete(reclaimPolicyDelete) for the only copy of 180 days of logs;cephrbd-fast-retainexists. NATS/ClickHouse use-deletebecause they are 72h/short buffers; this store is not.