website/docs/getting-started-guides/fluentd-gcp.yaml

480 lines
20 KiB
YAML

apiVersion: v1
kind: Pod
metadata:
name: fluentd-cloud-logging
namespace: kube-system
labels:
k8s-app: fluentd-logging
# This annotation ensures that fluentd does not get evicted if the node
# supports critical pod annotation based priority scheme.
# Note that this does not guarantee admission on the nodes (#40573).
annotations:
scheduler.alpha.kubernetes.io/critical-pod: ''
spec:
dnsPolicy: Default
containers:
- name: fluentd-cloud-logging
image: gcr.io/google_containers/fluentd-gcp:2.0.2
# If fluentd consumes its own logs, the following situation may happen:
# fluentd fails to send a chunk to the server => writes it to the log =>
# tries to send this message to the server => fails to send a chunk and so on.
# Writing to a file, which is not exported to the back-end prevents it.
# It also allows to increase the fluentd verbosity by default.
command:
- '/bin/sh'
- '-c'
- |-
mkdir /etc/fluent/config.d &&
echo "$FLUENTD_CONFIG" > /etc/fluent/config.d/main.conf &&
/run.sh $FLUENTD_ARGS 2>&1 >>/var/log/fluentd.log
env:
- name: FLUENTD_ARGS
value: --no-supervisor
# Keep this config as close as possible to cluster/addons/fluentd-gcp/fluentd-gcp-configmap.yaml
# Note that backslashes should be doubled, because this is interpreted as shell variable
# TODO(crassirostris): Refactor this
- name: FLUENTD_CONFIG
value: |-
# This configuration file for Fluentd 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 /var/log/containers directory using names that capture the
# pod name and container name. These logs are then submitted to
# Google Cloud Logging which assumes the installation of the cloud-logging plug-in.
#
# Example
# =======
# A line in the Docker log file might look like this JSON:
#
# {"log":"2014/09/25 21:15:03 Got request with path wombat\\n",
# "stream":"stderr",
# "time":"2014-09-25T21:15:03.499185026Z"}
#
# 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:
# /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
# ->
# /var/lib/docker/containers/997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b/997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b-json.log
# 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:
# /var/log/containers/synthetic-logger-0.25lps-pod_default-synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
# This results in the tag:
# var.log.containers.synthetic-logger-0.25lps-pod_default-synth-lgr-997599971ee6366d4a5920d25b79286ad45ff37a74494f262e3bc98d909d0a7b.log
# The record reformer is used is discard the var.log.containers prefix and
# the Docker container ID suffix and "kubernetes." is pre-pended giving the tag:
# kubernetes.synthetic-logger-0.25lps-pod_default-synth-lgr
# Tag is then parsed by google_cloud plugin and translated to the metadata,
# visible in the log viewer
# Example:
# {"log":"[info:2016-02-16T16:04:05.930-08:00] Some log text here\\n","stream":"stdout","time":"2016-02-17T00:04:05.931087621Z"}
<source>
type tail
format json
time_key time
path /var/log/containers/*.log
pos_file /var/log/gcp-containers.log.pos
time_format %Y-%m-%dT%H:%M:%S.%N%Z
tag reform.*
read_from_head true
</source>
<filter reform.**>
type parser
format /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<log>.*)/
reserve_data true
suppress_parse_error_log true
key_name log
</filter>
<match reform.**>
type record_reformer
enable_ruby true
tag raw.kubernetes.${tag_suffix[4].split('-')[0..-2].join('-')}
</match>
# Detect exceptions in the log output and forward them as one log entry.
<match raw.kubernetes.**>
@type copy
<store>
@type prometheus
<metric>
type counter
name logging_line_count
desc Total number of lines generated by application containers
<labels>
tag ${tag}
</labels>
</metric>
</store>
<store>
@type detect_exceptions
remove_tag_prefix raw
message log
stream stream
multiline_flush_interval 5
max_bytes 500000
max_lines 1000
</store>
</match>
# Example:
# 2015-12-21 23:17:22,066 [salt.state ][INFO ] Completed state [net.ipv4.ip_forward] at time 23:17:22.066081
<source>
type tail
format /^(?<time>[^ ]* [^ ,]*)[^\\[]*\\[[^\\]]*\\]\\[(?<severity>[^ \\]]*) *\\] (?<message>.*)$/
time_format %Y-%m-%d %H:%M:%S
path /var/log/salt/minion
pos_file /var/log/gcp-salt.pos
tag salt
</source>
# Example:
# Dec 21 23:17:22 gke-foo-1-1-4b5cbd14-node-4eoj startupscript: Finished running startup script /var/run/google.startup.script
<source>
type tail
format syslog
path /var/log/startupscript.log
pos_file /var/log/gcp-startupscript.log.pos
tag startupscript
</source>
# Examples:
# time="2016-02-04T06:51:03.053580605Z" level=info msg="GET /containers/json"
# time="2016-02-04T07:53:57.505612354Z" level=error msg="HTTP Error" err="No such image: -f" statusCode=404
<source>
type tail
format /^time="(?<time>[^)]*)" level=(?<severity>[^ ]*) msg="(?<message>[^"]*)"( err="(?<error>[^"]*)")?( statusCode=($<status_code>\\d+))?/
path /var/log/docker.log
pos_file /var/log/gcp-docker.log.pos
tag docker
</source>
# Example:
# 2016/02/04 06:52:38 filePurge: successfully removed file /var/etcd/data/member/wal/00000000000006d0-00000000010a23d1.wal
<source>
type tail
# Not parsing this, because it doesn't have anything particularly useful to
# parse out of it (like severities).
format none
path /var/log/etcd.log
pos_file /var/log/gcp-etcd.log.pos
tag etcd
</source>
# Multi-line parsing is required for all the kube logs because very large log
# statements, such as those that include entire object bodies, get split into
# multiple lines by glog.
# Example:
# I0204 07:32:30.020537 3368 server.go:1048] POST /stats/container/: (13.972191ms) 200 [[Go-http-client/1.1] 10.244.1.3:40537]
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/kubelet.log
pos_file /var/log/gcp-kubelet.log.pos
tag kubelet
</source>
# Example:
# I1118 21:26:53.975789 6 proxier.go:1096] Port "nodePort for kube-system/default-http-backend:http" (:31429/tcp) was open before and is still needed
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/kube-proxy.log
pos_file /var/log/gcp-kube-proxy.log.pos
tag kube-proxy
</source>
# Example:
# I0204 07:00:19.604280 5 handlers.go:131] GET /api/v1/nodes: (1.624207ms) 200 [[kube-controller-manager/v1.1.3 (linux/amd64) kubernetes/6a81b50] 127.0.0.1:38266]
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/kube-apiserver.log
pos_file /var/log/gcp-kube-apiserver.log.pos
tag kube-apiserver
</source>
# Example:
# 2017-02-09T00:15:57.992775796Z AUDIT: id="90c73c7c-97d6-4b65-9461-f94606ff825f" ip="104.132.1.72" method="GET" user="kubecfg" as="<self>" asgroups="<lookup>" namespace="default" uri="/api/v1/namespaces/default/pods"
# 2017-02-09T00:15:57.993528822Z AUDIT: id="90c73c7c-97d6-4b65-9461-f94606ff825f" response="200"
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\S+\\s+AUDIT:/
# Fields must be explicitly captured by name to be parsed into the record.
# Fields may not always be present, and order may change, so this just looks
# for a list of key="\\"quoted\\" value" pairs separated by spaces.
# Unknown fields are ignored.
# Note: We can't separate query/response lines as format1/format2 because
# they don't always come one after the other for a given query.
# TODO: Maybe add a JSON output mode to audit log so we can get rid of this?
format1 /^(?<time>\\S+) AUDIT:(?: (?:id="(?<id>(?:[^"\\\\]|\\\\.)*)"|ip="(?<ip>(?:[^"\\\\]|\\\\.)*)"|method="(?<method>(?:[^"\\\\]|\\\\.)*)"|user="(?<user>(?:[^"\\\\]|\\\\.)*)"|groups="(?<groups>(?:[^"\\\\]|\\\\.)*)"|as="(?<as>(?:[^"\\\\]|\\\\.)*)"|asgroups="(?<asgroups>(?:[^"\\\\]|\\\\.)*)"|namespace="(?<namespace>(?:[^"\\\\]|\\\\.)*)"|uri="(?<uri>(?:[^"\\\\]|\\\\.)*)"|response="(?<response>(?:[^"\\\\]|\\\\.)*)"|\\w+="(?:[^"\\\\]|\\\\.)*"))*/
time_format %FT%T.%L%Z
path /var/log/kube-apiserver-audit.log
pos_file /var/log/gcp-kube-apiserver-audit.log.pos
tag kube-apiserver-audit
</source>
# Example:
# I0204 06:55:31.872680 5 servicecontroller.go:277] LB already exists and doesn't need update for service kube-system/kube-ui
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/kube-controller-manager.log
pos_file /var/log/gcp-kube-controller-manager.log.pos
tag kube-controller-manager
</source>
# Example:
# W0204 06:49:18.239674 7 reflector.go:245] pkg/scheduler/factory/factory.go:193: watch of *api.Service ended with: 401: The event in requested index is outdated and cleared (the requested history has been cleared [2578313/2577886]) [2579312]
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/kube-scheduler.log
pos_file /var/log/gcp-kube-scheduler.log.pos
tag kube-scheduler
</source>
# Example:
# I1104 10:36:20.242766 5 rescheduler.go:73] Running Rescheduler
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/rescheduler.log
pos_file /var/log/gcp-rescheduler.log.pos
tag rescheduler
</source>
# Example:
# I0603 15:31:05.793605 6 cluster_manager.go:230] Reading config from path /etc/gce.conf
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/glbc.log
pos_file /var/log/gcp-glbc.log.pos
tag glbc
</source>
# Example:
# I0603 15:31:05.793605 6 cluster_manager.go:230] Reading config from path /etc/gce.conf
<source>
type tail
format multiline
multiline_flush_interval 5s
format_firstline /^\\w\\d{4}/
format1 /^(?<severity>\\w)(?<time>\\d{4} [^\\s]*)\\s+(?<pid>\\d+)\\s+(?<source>[^ \\]]+)\\] (?<message>.*)/
time_format %m%d %H:%M:%S.%N
path /var/log/cluster-autoscaler.log
pos_file /var/log/gcp-cluster-autoscaler.log.pos
tag cluster-autoscaler
</source>
# Logs from systemd-journal for interesting services.
<source>
type systemd
filters [{ "_SYSTEMD_UNIT": "docker.service" }]
pos_file /var/log/gcp-journald-docker.pos
read_from_head true
tag docker
</source>
<source>
type systemd
filters [{ "_SYSTEMD_UNIT": "kubelet.service" }]
pos_file /var/log/gcp-journald-kubelet.pos
read_from_head true
tag kubelet
</source>
# Prometheus monitoring
<source>
@type prometheus
port 80
</source>
<source>
@type prometheus_monitor
</source>
# We use 2 output stanzas - one to handle the container logs and one to handle
# the node daemon logs, the latter of which explicitly sends its logs to the
# compute.googleapis.com service rather than container.googleapis.com to keep
# them separate since most users don't care about the node logs.
<match kubernetes.**>
@type copy
<store>
@type google_cloud
# Set the buffer type to file to improve the reliability and reduce the memory consumption
buffer_type file
buffer_path /var/log/fluentd-buffers/kubernetes.containers.buffer
# Set queue_full action to block because we want to pause gracefully
# in case of the off-the-limits load instead of throwing an exception
buffer_queue_full_action block
# Set the chunk limit conservatively to avoid exceeding the GCL limit
# of 10MiB per write request.
buffer_chunk_limit 2M
# Cap the combined memory usage of this buffer and the one below to
# 2MiB/chunk * (6 + 2) chunks = 16 MiB
buffer_queue_limit 6
# Never wait more than 5 seconds before flushing logs in the non-error case.
flush_interval 5s
# Never wait longer than 30 seconds between retries.
max_retry_wait 30
# Disable the limit on the number of retries (retry forever).
disable_retry_limit
# Use multiple threads for processing.
num_threads 2
</store>
<store>
@type prometheus
<metric>
type counter
name logging_entry_count
desc Total number of log entries generated by either an application container or a system component
<labels>
tag ${tag}
component container
</labels>
</metric>
</store>
</match>
# Keep a smaller buffer here since these logs are less important than the user's
# container logs.
<match **>
@type copy
<store>
@type google_cloud
detect_subservice false
buffer_type file
buffer_path /var/log/fluentd-buffers/kubernetes.system.buffer
buffer_queue_full_action block
buffer_chunk_limit 2M
buffer_queue_limit 2
flush_interval 5s
max_retry_wait 30
disable_retry_limit
num_threads 2
</store>
<store>
@type prometheus
<metric>
type counter
name logging_entry_count
desc Total number of log entries generated by either an application container or a system component
<labels>
tag ${tag}
component system
</labels>
</metric>
</store>
</match>
resources:
limits:
memory: 200Mi
requests:
# Any change here should be accompanied by a proportional change in CPU
# requests of other per-node add-ons (e.g. kube-proxy).
cpu: 100m
memory: 200Mi
volumeMounts:
- name: varlog
mountPath: /var/log
- name: varlibdockercontainers
mountPath: /var/lib/docker/containers
readOnly: true
- name: libsystemddir
mountPath: /host/lib
readOnly: true
# Liveness probe is aimed to help in situarions where fluentd
# silently hangs for no apparent reasons until manual restart.
# The idea of this probe is that if fluentd is not queueing or
# flushing chunks for 5 minutes, something is not right. If
# you want to change the fluentd configuration, reducing amount of
# logs fluentd collects, consider changing the threshold or turning
# liveness probe off completely.
livenessProbe:
initialDelaySeconds: 600
periodSeconds: 60
exec:
command:
- '/bin/sh'
- '-c'
- >
LIVENESS_THRESHOLD_SECONDS=${LIVENESS_THRESHOLD_SECONDS:-300};
STUCK_THRESHOLD_SECONDS=${LIVENESS_THRESHOLD_SECONDS:-900};
if [ ! -e /var/log/fluentd-buffers ];
then
exit 1;
fi;
LAST_MODIFIED_DATE=`stat /var/log/fluentd-buffers | grep Modify | sed -r "s/Modify: (.*)/\1/"`;
LAST_MODIFIED_TIMESTAMP=`date -d "$LAST_MODIFIED_DATE" +%s`;
if [ `date +%s` -gt `expr $LAST_MODIFIED_TIMESTAMP + $STUCK_THRESHOLD_SECONDS` ];
then
rm -rf /var/log/fluentd-buffers;
exit 1;
fi;
if [ `date +%s` -gt `expr $LAST_MODIFIED_TIMESTAMP + $LIVENESS_THRESHOLD_SECONDS` ];
then
exit 1;
fi;
terminationGracePeriodSeconds: 30
volumes:
- name: varlog
hostPath:
path: /var/log
- name: varlibdockercontainers
hostPath:
path: /var/lib/docker/containers
- name: libsystemddir
hostPath:
path: /usr/lib64