Merge pull request #172 from sysdiglabs/staging

Staging to Production

Author: daviddetorres

resources/ceph/INSTALL.md

@@ -1,13 +1,9 @@
 # Prerequisites
-Make sure the Sysdig agent has the right configuration.
-Ceph instruments Prometheus metrics and annotates the manager pod with Prometheus annotations, so there is not further configuration needed.
+Ceph instruments Prometheus metrics and annotates the manager pod with Prometheus annotations. 
 
-For the Sysdig agent to discover and scrape it automatically, enable the `use_promscrape` option in the agent configuration.
+Make sure that the prometheus module is activated in the Ceph cluster by running the following command:
 
-```yaml
-  dragent.yaml: |-
-    use_promscrape: true
-    prometheus:
-      enabled: true
-      prom_service_discovery: true
 ```
+ceph mgr module enable prometheus
+```
+

resources/elasticsearch/INSTALL.md

@@ -1,7 +1,6 @@
 # Installing the exporter
-To install the exporter, do the following:
-1. Verify whether your elastic is secured or not
-2. If it is not secured, download the `exporter_no_credentials.yaml` file and execute:
+## Without credentials
+If it is not secured, download the `exporter_no_credentials.yaml` file and execute:
 ```
 kubectl apply -f exporter_no_credentials.yaml
 ```
@@ -24,15 +23,16 @@ spec:
             value: https://elasticsearch:9200
       ...
 ```
-3. If the elasticsearch is secured similar to the elasticseach deployed in the Sysdig on-prem environments, download the `exporter_with_credentials.yaml` file.
+## With credentials
+If elasticsearch is secured, download the `exporter_with_credentials.yaml` file. 
 
-4. Apply by configuration changes by executing the following:
-```
-kubectl apply -f exporter_with_credentials.yaml
-```
-This configuration will use the user and the password saved as secrets, and will mount a volume with the ca.
+This configuration will use the user and the password saved as secrets, and will mount a volume with the ca certificate.
 
-5. If you are using the exporter with credentials but you are not using sysdig, then edit the secrets name, the service, and the port.
+Create the secret with the ElasticSearch certificates for the exporter and edit in the yaml file:
+* Secrets name (by default `elasticsearch-tls` and `es-certs`)
+* Service
+* Host
+* Port
 
 ```yaml
 ...
@@ -49,12 +49,17 @@ spec:
               key: elasticsearch.adminuser
               name: sysdigcloud-config
         - name: ES_URI
-          value: https://$(ELASTIC_USER):$(ELASTICSEARCH_ADMIN_PASSWORD)@sysdigcloud-elasticsearch:9200
+          value: https://$(ELASTIC_USER):$(ELASTICSEARCH_ADMIN_PASSWORD)@YOUR-HOST:9200
       envFrom:
       - secretRef:
           name: elasticsearch-tls
       ...
 ```
+
+Apply by configuration changes by executing the following:
+```
+kubectl apply -f exporter_with_credentials.yaml
+```
 # Sysdig Agent configuration
 > Requires Sysdig Agent >= v11.3 
 
@@ -67,15 +72,15 @@ spec:
   template:
     metadata:
       annotations:
+        promcat.sysdig.com/integration_type: elasticsearch
         promcat.sysdig.com/port: "9108"
         # Add here the namespace, workload type (deployment, statefulset, replicaset, daemonset) 
         # and workload name of the instance that the exporter will take data from
         promcat.sysdig.com/target_ns: elastic-namespace
         promcat.sysdig.com/target_workload_type: statefulset
         promcat.sysdig.com/target_workload_name: elasticsearch
-        promcat.sysdig.com/integration_type: elasticsearch
 ```
-This way, on the Sysdig Monitor, you can view the associated metrics corresponding to the database pods and the exporter.
+This way, on Sysdig Monitor, you can view the associated metrics corresponding to ElasticSearch pods and the exporter.
 
 After you configure the Sysdig annotations, download the sample configuration file and apply it by:
 ```bash

resources/elasticsearch/include/exporter_with_credentials.yaml

@@ -47,7 +47,7 @@ spec:
                 key: elasticsearch.adminuser
                 name: sysdigcloud-config
           - name: ES_URI
-            value: https://$(ELASTIC_USER):$(ELASTICSEARCH_ADMIN_PASSWORD)@sysdigcloud-elasticsearch:9200
+            value: https://$(ELASTIC_USER):$(ELASTICSEARCH_ADMIN_PASSWORD)@YOURHOST:9200
         envFrom:
         - secretRef:
             name: elasticsearch-tls

resources/nginx-ingress/INSTALL.md

@@ -1,16 +1,15 @@
 # Installing the exporter
 Nginx ingress controller is already instrumented so you don't have to add any extra exporter. However, ensure that your nginx ingress controller is already exposing these metrics.
 
-For exmaple if you are deploying your nginx ingress controller with helm, make sure you have these values.
+For example if you are deploying your nginx ingress controller with helm, make sure you have these values.
 
 ```yaml
 controller:
   metrics:
     port: 10254
-    # if this port is changed, change healthz-port: in extraArgs: accordingly
     enabled: true
   podAnnotations:
-    promcat.sysdig.com/port: "10254"
+    prometheus.io/port: "10254"
 ```
 
 # Sysdig Agent configuration

resources/nginx/INSTALL.md

@@ -1,31 +1,36 @@
 # Installing the exporter
-The exporter can be installed as a sidecar of the pod with the Nginx server. You can find a deployment below with the exporter as a sidecar and the configmap with the configuration required to scrape metrics from the server.
+The exporter can be installed as a sidecar of the pod with the Nginx server. In order to get the nginx a endpoint to scrap the metrics you have to enable the metrics endpoint.
 
-Note that the deployment has a label, `app`. This label will be included in the metrics to be able to use it as a variable in the dashboards and alerts.
-
-# Sysdig Agent configuration
-In the Sysdig Agent configuration, add the following snippet to include the label `app` as a metric label.
 ```yaml
-  dragent.yaml: |-
-    metrics_excess_log: true
-    k8s_cluster_name: YourClusterName
-    10s_flush_enable: true
-    app_checks_enabled: false
-    use_promscrape: true
-    new_k8s: true
-    promscrape_fastproto: true
-    prometheus:
-      enabled: true
-      prom_service_discovery: true
-      log_errors: true
-      max_metrics: 200000
-      max_metrics_per_process: 200000
-      max_tags_per_metric: 100
-      ingest_raw: true
-      ingest_calculated: false
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: nginx-config
+data:
+  nginx.conf: |
+    server {
+      listen       80;
+      server_name  localhost;
+      location /nginx_status {
+        stub_status on;
+        access_log  on;
+        allow all;  # REPLACE with your access policy
+      }
+    }
 ```
 
-You can download the sample configuration file below and apply it by:
-```bash
-kubectl apply -f sysdig-agent-config.yaml
+You can find a deployment below with the exporter as a sidecar and the ConfigMap with the configuration required to scrape metrics from the server.
+
+Also, be sure to add the following annotations to the deployment:
+
+```yaml
+spec:
+  template:
+    metadata:
+      annotations:
+        promcat.sysdig.com/integration_type: nginx
+        promcat.sysdig.com/port: "9113"
 ```
+
+# Sysdig Agent configuration
+Below you can find the configuration for the sysdig agent.

resources/redis/INSTALL.md

@@ -39,8 +39,8 @@ spec:
   template:
     metadata:
       annotations:
+        promcat.sysdig.com/integration_type: redis
         promcat.sysdig.com/port: "9121"
-
         # Add here the namespace, workload type (deployment, statefulset, replicaset, daemonset)
         # and workload name of the Redis instance that the exporter will take data from
         promcat.sysdig.com/target_ns: default