From a0179b829df382a145c3572a95fd3606ff4255f5 Mon Sep 17 00:00:00 2001
From: vincent <vincent@ducamps.win>
Date: Tue, 28 Nov 2023 18:22:13 +0100
Subject: [PATCH] feat: switch to vector for docker log collect

Signed-off-by: vincent <vincent@ducamps.win>
---
 ansible/group_vars/all/nomad         |  2 +-
 nomad-job/{ => decom}/promtail.nomad |  0
 nomad-job/vector.nomad               | 87 ++++++++++++++++++++++++++++
 3 files changed, 88 insertions(+), 1 deletion(-)
 rename nomad-job/{ => decom}/promtail.nomad (100%)
 create mode 100644 nomad-job/vector.nomad

diff --git a/ansible/group_vars/all/nomad b/ansible/group_vars/all/nomad
index f644a75..0f2cfa5 100644
--- a/ansible/group_vars/all/nomad
+++ b/ansible/group_vars/all/nomad
@@ -5,4 +5,4 @@ nomad_docker_allow_caps:
 nomad_vault_enabled: true
 nomad_vault_address: "http://active.vault.service.{{consul_domain}}:8200"
 nomad_vault_role: "nomad-cluster"
-
+nomad_docker_extra_labels: ["job_name", "task_group_name", "task_name", "namespace", "node_name"]
diff --git a/nomad-job/promtail.nomad b/nomad-job/decom/promtail.nomad
similarity index 100%
rename from nomad-job/promtail.nomad
rename to nomad-job/decom/promtail.nomad
diff --git a/nomad-job/vector.nomad b/nomad-job/vector.nomad
new file mode 100644
index 0000000..34fb44a
--- /dev/null
+++ b/nomad-job/vector.nomad
@@ -0,0 +1,87 @@
+job "vector" {
+  datacenters = ["homelab", "hetzner"]
+  priority    = 50
+  type        = "system"
+  meta {
+    forcedeploy = "0"
+  }
+  group "vector" {
+    count = 1
+    network {
+      port api {
+        to = 8686
+      }
+    }
+    task "vector" {
+      driver = "docker"
+      config {
+        image = "timberio/vector:0.34.1-alpine"
+        ports = ["api"]
+        volumes = [
+          "/var/run/docker.sock:/var/run/docker.sock",
+        ]
+
+      }
+      # docker socket volume mount
+      env {
+        VECTOR_CONFIG = "local/vector.toml"
+        VECTOR_REQUIRE_HEALTHY = "true"
+      }
+      # resource limits are a good idea because you don't want your log collection to consume all resources available
+      resources {
+        cpu    = 500 # 500 MHz
+        memory = 256 # 256MB
+      }
+      # template with Vector's configuration
+      template {
+        destination = "local/vector.toml"
+        change_mode   = "signal"
+        change_signal = "SIGHUP"
+        # overriding the delimiters to [[ ]] to avoid conflicts with Vector's native templating, which also uses {{ }}
+        left_delimiter = "[["
+        right_delimiter = "]]"
+        data=<<EOH
+          data_dir = "alloc/data/vector/"
+          [api]
+            enabled = true
+            address = "0.0.0.0:8686"
+            playground = true
+          [sources.logs]
+            type = "docker_logs"
+            exclude_containers= ["loki"]
+          [sources.podman_logs]
+          type = "docker_logs"
+          docker_host = "unix:///var/run/podman/podman.sock"
+          [sinks.loki]
+            type = "loki"
+            inputs = ["logs","podman_logs"]
+            endpoint = "http://[[ range service "loki" ]][[ .Address ]]:[[ .Port ]][[ end ]]"
+            encoding.codec = "json"
+            healthcheck.enabled = true
+            # since . is used by Vector to denote a parent-child relationship, and Nomad's Docker labels contain ".",
+            # we need to escape them twice, once for TOML, once for Vector
+            labels.source = 'vector'
+            labels.job = '{{ label."com.hashicorp.nomad.job_name" }}'
+            labels.task = '{{ label."com.hashicorp.nomad.task_name" }}'
+            labels.group = '{{ label."com.hashicorp.nomad.task_group_name" }}'
+            labels.namespace = '{{ label."com.hashicorp.nomad.namespace" }}'
+            labels.node = '{{ label."com.hashicorp.nomad.node_name" }}'
+            # remove fields that have been converted to labels to avoid having the field twice
+            remove_label_fields = true
+        EOH
+      }
+      service {
+        check {
+          port     = "api"
+          type     = "http"
+          path     = "/health"
+          interval = "30s"
+          timeout  = "5s"
+        }
+      }
+      kill_timeout = "30s"
+    }
+  }
+
+
+}