Switch to fluentd kubernetes metadata plugin to enrich logs with more searchable…

cluster/addons/fluentd-elasticsearch/fluentd-es-image/Dockerfile

@@ -4,12 +4,13 @@
 # TODO(a-robinson): Use a lighter base image, e.g. some form of busybox.
 # The image acts as an executable for the binary /usr/sbin/td-agent.
 # Note that fluentd is run with root permssion to allow access to
-# log files with root only access under /var/lib/docker/containers/*
+# log files with root only access under /var/log/containers/*
 # Please see http://docs.fluentd.org/articles/install-by-deb for more
 # information about installing fluentd using deb package.
 
 FROM ubuntu:14.04
 MAINTAINER Alex Robinson "arob@google.com"
+MAINTAINER Jimmi Dyson "jimmidyson@gmail.com"
 
 # Ensure there are enough file descriptors for running Fluentd.
 RUN ulimit -n 65536
@@ -19,9 +20,9 @@ ENV DEBIAN_FRONTEND noninteractive
 
 # Install prerequisites.
 RUN apt-get update && \
-    apt-get install -y curl && \
-    apt-get install -y -q libcurl4-openssl-dev make && \
-    apt-get clean
+    apt-get install -y -q curl make g++ && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
 
 # Install Fluentd.
 RUN /usr/bin/curl -L https://td-toolbelt.herokuapp.com/sh/install-ubuntu-trusty-td-agent2.sh | sh
@@ -31,13 +32,10 @@ RUN /usr/bin/curl -L https://td-toolbelt.herokuapp.com/sh/install-ubuntu-trusty-
 RUN sed -i -e "s/USER=td-agent/USER=root/" -e "s/GROUP=td-agent/GROUP=root/" /etc/init.d/td-agent
 
 # Install the Elasticsearch Fluentd plug-in.
-RUN /usr/sbin/td-agent-gem install fluent-plugin-elasticsearch
-
-# Install the record reformer plugin.
-RUN /usr/sbin/td-agent-gem install fluent-plugin-record-reformer
+RUN td-agent-gem install fluent-plugin-kubernetes_metadata_filter fluent-plugin-elasticsearch
 
 # Copy the Fluentd configuration file.
 COPY td-agent.conf /etc/td-agent/td-agent.conf
 
 # Run the Fluentd service.
-CMD /usr/sbin/td-agent "$FLUENTD_ARGS"
+ENTRYPOINT ["td-agent"]

cluster/addons/fluentd-elasticsearch/fluentd-es-image/Makefile

 .PHONY:	build push
 
 IMAGE = fluentd-elasticsearch
-TAG = 1.10
+TAG = 1.11
 
 build:	
 	docker build -t gcr.io/google_containers/$(IMAGE):$(TAG) .

cluster/addons/fluentd-elasticsearch/fluentd-es-image/td-agent.conf

 # This configuration file for Fluentd / td-agent is used
-# to watch changes to Docker log files that live in the
-# directory /var/lib/docker/containers/ and are symbolically
-# linked to from the /varlog directory using names that capture the
-# pod name and container name. These logs are then submitted to
-# Elasticsearch which assumes the installation of the fluentd-elasticsearch plug-in.
-# See https://github.com/uken/fluent-plugin-elasticsearch for
-# more information about the plug-in.
+# to watch changes to Docker log files. The kubelet creates symlinks that
+# capture the pod name, namespace, container name & Docker container ID
+# to the docker logs for pods in the /var/log/containers directory on the host.
+# If running this fluentd configuration in a Docker container, the /var/log
+# directory should be mounted in the container.
+#
+# These logs are then submitted to Elasticsearch which assumes the
+# installation of the fluent-plugin-elasticsearch & the
+# fluent-plugin-kubernetes_metadata_filter plugins.
+# See https://github.com/uken/fluent-plugin-elasticsearch &
+# https://github.com/fabric8io/fluent-plugin-kubernetes_metadata_filter for
+# more information about the plugins.
+# Maintainer: Jimmi Dyson <jimmidyson@gmail.com>
 #
 # Example
 # =======
@@ -31,116 +37,147 @@
 #    },
 # ...
 #
-# The record reformer is used to write the tag to focus on the pod name
-# and the Kubernetes container name. For example a Docker container's logs
-# might be in the directory:
+# The Kubernetes fluentd plugin is used to write the Kubernetes metadata to the log
+# record & add labels to the log record if properly configured. This enables users
+# to filter & search logs on any metadata.
+# For example a Docker container's logs might be in the directory:
+#
 #  /var/lib/docker/containers/997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b
+#
 # and in the file:
+#
 #  997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b-json.log
+#
 # where 997599971ee6... is the Docker ID of the running container.
 # The Kubernetes kubelet makes a symbolic link to this file on the host machine
 # in the /var/log/containers directory which includes the pod name and the Kubernetes
 # container name:
-#    synthetic-logger-0.25lps-pod_default-synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log 
-#    -> 
+#
+#    synthetic-logger-0.25lps-pod_default_synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log 
+#    ->
 #    /var/lib/docker/containers/997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b/997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b-json.log
-# The /var/log directory on the host is mapped to the /varlog directory in the container
+#
+# The /var/log directory on the host is mapped to the /var/log directory in the container
 # running this instance of Fluentd and we end up collecting the file:
-#   /varlog/containers/synthetic-logger-0.25lps-pod_default-synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
+#
+#   /var/log/containers/synthetic-logger-0.25lps-pod_default_synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
+#
 # This results in the tag:
-#  varlog.containers.synthetic-logger-0.25lps-pod_default-synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
-# The record reformer is used is discard the varlog.containers prefix and
-# the Docker container ID suffix and "kubernetes." is pre-pended giving the
-# final tag which is ingested into Elasticsearch:
-#   kubernetes.synthetic-logger-0.25lps-pod_default-synth-lgr
+#
+#  var.log.containers.synthetic-logger-0.25lps-pod_default_synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
+#
+# The Kubernetes fluentd plugin is used to extract the namespace, pod name & container name
+# which are added to the log message as a kubernetes field object & the Docker container ID
+# is also added under the docker field object.
+# The final tag is:
+#
+#   kubernetes.var.log.containers.synthetic-logger-0.25lps-pod_default_synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
+#
+# And the final log record look like:
+#
+# {
+#   "log":"2014/09/25 21:15:03 Got request with path wombat\n",
+#   "stream":"stderr",
+#   "time":"2014-09-25T21:15:03.499185026Z",
+#   "kubernetes": {
+#     "namespace": "default",
+#     "pod_name": "synthetic-logger-0.25lps-pod",
+#     "container_name": "synth-lgr"
+#   },
+#   "docker": {
+#     "container_id": "997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b"
+#   }
+# }
+#
 # This makes it easier for users to search for logs by pod name or by
 # the name of the Kubernetes container regardless of how many times the
 # Kubernetes pod has been restarted (resulting in a several Docker container IDs).
+#
 # TODO: Propagate the labels associated with a container along with its logs
 # so users can query logs using labels as well as or instead of the pod name
-# and container name.
+# and container name. This is simply done via configuration of the Kubernetes
+# fluentd plugin but requires secrets to be enabled in the fluent pod. This is a
+# problem yet to be solved as secrets are not usable in static pods which the fluentd
+# pod must be until a per-node controller is available in Kubernetes.
 
 <source>
   type tail
-  format json
-  time_key time
-  path /varlog/containers/*.log
-  pos_file /varlog/es-containers.log.pos
+  path /var/log/containers/*.log
+  pos_file /var/log/es-containers.log.pos
   time_format %Y-%m-%dT%H:%M:%S
-  tag reform.*
+  tag kubernetes.*
+  format json
   read_from_head true
 </source>
 
-<match reform.**>
-  type record_reformer
-  enable_ruby true
-  tag kubernetes.${tag_suffix[3].split('-')[0..-2].join('-')}
-</match>
-
 <source>
   type tail
   format none
-  path /varlog/salt/minion
-  pos_file /varlog/gcp-salt.pos
+  path /var/log/salt/minion
+  pos_file /var/log/gcp-salt.pos
   tag salt
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/startupscript.log
-  pos_file /varlog/es-startupscript.log.pos
+  path /var/log/startupscript.log
+  pos_file /var/log/es-startupscript.log.pos
   tag startupscript
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/docker.log
-  pos_file /varlog/es-docker.log.pos
+  path /var/log/docker.log
+  pos_file /var/log/es-docker.log.pos
   tag docker
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/etcd.log
-  pos_file /varlog/es-etcd.log.pos
+  path /var/log/etcd.log
+  pos_file /var/log/es-etcd.log.pos
   tag etcd
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/kubelet.log
-  pos_file /varlog/es-kubelet.log.pos
+  path /var/log/kubelet.log
+  pos_file /var/log/es-kubelet.log.pos
   tag kubelet
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/kube-apiserver.log
-  pos_file /varlog/es-kube-apiserver.log.pos
+  path /var/log/kube-apiserver.log
+  pos_file /var/log/es-kube-apiserver.log.pos
   tag kube-apiserver
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/kube-controller-manager.log
-  pos_file /varlog/es-kube-controller-manager.log.pos
+  path /var/log/kube-controller-manager.log
+  pos_file /var/log/es-kube-controller-manager.log.pos
   tag kube-controller-manager
 </source>
 
 <source>
   type tail
   format none
-  path /varlog/kube-scheduler.log
-  pos_file /varlog/es-kube-scheduler.log.pos
+  path /var/log/kube-scheduler.log
+  pos_file /var/log/es-kube-scheduler.log.pos
   tag kube-scheduler
 </source>
 
+<filter kubernetes.**>
+  type kubernetes_metadata
+</filter>
+
 <match **>
    type elasticsearch
    log_level info

cluster/saltbase/salt/fluentd-es/fluentd-es.yaml

@@ -6,24 +6,24 @@ metadata:
 spec:
   containers:
   - name: fluentd-elasticsearch
-    image: gcr.io/google_containers/fluentd-elasticsearch:1.9
+    image: gcr.io/google_containers/fluentd-elasticsearch:1.11
     resources:
       limits:
         cpu: 100m
-    env:
-    - name: "FLUENTD_ARGS"
-      value: "-q"
+    args:
+    - -qq
     volumeMounts:
     - name: varlog
-      mountPath: /varlog
-    - name: containers
+      mountPath: /var/log
+    - name: varlibdockercontainers
       mountPath: /var/lib/docker/containers
+      readOnly: true
   terminationGracePeriodSeconds: 30
   volumes:
   - name: varlog
     hostPath:
       path: /var/log
-  - name: containers
+  - name: varlibdockercontainers
     hostPath:
       path: /var/lib/docker/containers