upstream/kubernetes
1
0
Fork 0
mirror of https://github.com/kubernetes/kubernetes.git synced 2026-03-21 01:51:01 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions
kubernetes/docs/logging.md

49 lines
6.2 KiB
Markdown
Raw Normal View History

Separated user, dev, and design docs. Renamed: logging.md -> devel/logging.m Renamed: access.md -> design/access.md Renamed: identifiers.md -> design/identifiers.md Renamed: labels.md -> design/labels.md Renamed: namespaces.md -> design/namespaces.md Renamed: security.md -> design/security.md Renamed: networking.md -> design/networking.md Added abbreviated user user-focused document in place of most moved docs. Added docs/README.md explains how docs are organized. Added short, user-oriented documentation on labels Added a glossary. Fixed up some links.
2014-10-16 17:45:16 -04:00
# Logging
## Logging by Kubernetes Components
Kubernetes components, such as kubelet and apiserver, use the [glog](https://godoc.org/github.com/golang/glog) logging library. Developer conventions for logging severity are described in [devel/logging.md](devel/logging.md).
## Logging in Containers
Launch Elasticsearch and Kibana automatically
2015-01-07 18:02:35 -05:00
There are no Kubernetes-specific requirements for logging from within containers. [search](https://www.google.com/?q=docker+container+logging) will turn up any number of articles about logging and
Mention Fluentd in the logging doc, as originally requested by Kiyoto in PR 2727.
2014-12-11 17:29:52 -05:00
Docker containers. However, we do provide an example of how to collect, index, and view pod logs [using Fluentd, Elasticsearch, and Kibana](./getting-started-guides/logging.md)
Introduce some default log verbosity control Move a lot of common error logging into better buckets: glog.Errorf() - Always an error glog.Warningf() - Something unexpected, but probably not an error glog.V(0) - Generally useful for this to ALWAYS be visible to an operator * Programmer errors * Logging extra info about a panic * CLI argument handling glog.V(1) - A reasonable default log level if you don't want verbosity * Information about config (listening on X, watching Y) * Errors that repeat frequently that relate to conditions that can be corrected (pod detected as unhealthy) glog.V(2) - Useful steady state information about the service * Logging HTTP requests and their exit code * System state changing (killing pod) * Controller state change events (starting pods) * Scheduler log messages glog.V(3) - Extended information about changes * More info about system state changes glog.V(4) - Debug level verbosity (for now) * Logging in particularly thorny parts of code where you may want to come back later and check it
2014-09-18 06:46:14 -04:00
Launch Elasticsearch and Kibana automatically
2015-01-07 18:02:35 -05:00
## Logging to Elasticsearch on the GCE platform
Currently the collection of container logs using the [Fluentd](http://www.fluentd.org/) log collector is
enabled by default for clusters created for the GCE platform. Each node uses Fluentd to collect
the container logs which are submitted in [Logstash](http://logstash.net/docs/1.4.2/tutorials/getting-started-with-logstash)
format (in JSON) to an [Elasticsearch](http://www.elasticsearch.org/) cluster which runs as a Kubernetes service.
When you create a cluster the console output reports the URL of both the Elasticsearch cluster as well as
a URL for a [Kibana](http://www.elasticsearch.org/overview/kibana/) dashboard viewer for the logs that have been ingested
into Elasticsearch.
```
Cluster logs are ingested into Elasticsearch running at http://130.211.121.21:9200
Kibana logging dashboard will be available at http://130.211.137.206:5601
```
Visiting the Kibana dashboard URL in a browser should give a display like this:
![Kibana](kibana.png)
Introduce some default log verbosity control Move a lot of common error logging into better buckets: glog.Errorf() - Always an error glog.Warningf() - Something unexpected, but probably not an error glog.V(0) - Generally useful for this to ALWAYS be visible to an operator * Programmer errors * Logging extra info about a panic * CLI argument handling glog.V(1) - A reasonable default log level if you don't want verbosity * Information about config (listening on X, watching Y) * Errors that repeat frequently that relate to conditions that can be corrected (pod detected as unhealthy) glog.V(2) - Useful steady state information about the service * Logging HTTP requests and their exit code * System state changing (killing pod) * Controller state change events (starting pods) * Scheduler log messages glog.V(3) - Extended information about changes * More info about system state changes glog.V(4) - Debug level verbosity (for now) * Logging in particularly thorny parts of code where you may want to come back later and check it
2014-09-18 06:46:14 -04:00
Launch Elasticsearch and Kibana automatically
2015-01-07 18:02:35 -05:00
To learn how to query, fitler etc. using Kibana you might like to look at this [tutorial](http://www.elasticsearch.org/guide/en/kibana/current/working-with-queries-and-filters.html).
You can check to see if any logs are being ingested into Elasticsearch by curling against its URL:
```
$ curl http://130.211.120.118:9200/_search?size=10
{"took":7,"timed_out":false,"_shards":{"total":5,"successful":5,"failed":0},"hits":{"total":3705,"max_score":1.0,"hits":[{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8Durpdh","_score":1.0,"_source":{"message":"I0108 18:30:47.694725 4927 server.go:313] GET /healthz: (9.249us) 200","tag":"kubelet","@timestamp":"2015-01-08T18:30:47+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8Durpdm","_score":1.0,"_source":{"message":"E0108 18:30:52.299372 4927 metadata.go:109] while reading 'google-dockercfg' metadata: http status code: 404 while fetching url http://metadata.google.internal./computeMetadata/v1/instance/attributes/google-dockercfg","tag":"kubelet","@timestamp":"2015-01-08T18:30:52+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8Durpdr","_score":1.0,"_source":{"message":"I0108 18:30:52.317636 4927 docker.go:214] Pulling image kubernetes/kube2sky without credentials","tag":"kubelet","@timestamp":"2015-01-08T18:30:52+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8Durpdw","_score":1.0,"_source":{"message":"I0108 18:30:54.500174 4927 event.go:92] Event(api.ObjectReference{Kind:\"BoundPod\", Namespace:\"default\", Name:\"67cfcb1f-9764-11e4-898c-42010af03582\", UID:\"67cfcb1f-9764-11e4-898c-42010af03582\", APIVersion:\"v1beta1\", ResourceVersion:\"\", FieldPath:\"spec.containers{kube2sky}\"}): status: 'waiting', reason: 'created' Created with docker id ff24ec6eb3b10d1163a2bcb7c63ccef78e6e3e7a1185eba3fe430f6b3d871eb5","tag":"kubelet","@timestamp":"2015-01-08T18:30:54+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8Durpd1","_score":1.0,"_source":{"message":"goroutine 114 [running]:","tag":"kubelet","@timestamp":"2015-01-08T18:30:56+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8Durpd6","_score":1.0,"_source":{"message":"github.com/GoogleCloudPlatform/kubernetes/pkg/kubelet.(*Server).error(0xc2080e0060, 0x7fe0ba496840, 0xc208278840, 0x7fe0ba4881b0, 0xc20822daa0)","tag":"kubelet","@timestamp":"2015-01-08T18:30:56+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRRanI4L8DurpeB","_score":1.0,"_source":{"message":"\t/go/src/github.com/GoogleCloudPlatform/kubernetes/_output/dockerized/go/src/github.com/GoogleCloudPlatform/kubernetes/pkg/kubelet/server.go:94 +0x44","tag":"kubelet","@timestamp":"2015-01-08T18:30:56+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRSanI4L8DurpeJ","_score":1.0,"_source":{"message":"goroutine 114 [running]:","tag":"kubelet","@timestamp":"2015-01-08T18:30:56+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRSanI4L8DurpeO","_score":1.0,"_source":{"message":"github.com/GoogleCloudPlatform/kubernetes/pkg/kubelet.(*Server).error(0xc2080e0060, 0x7fe0ba496840, 0xc208278a80, 0x7fe0ba4881b0, 0xc20822df00)","tag":"kubelet","@timestamp":"2015-01-08T18:30:56+00:00"}},{"_index":"logstash-2015.01.08","_type":"fluentd","_id":"AUrK0hRSanI4L8DurpeT","_score":1.0,"_source":{"message":"\t/go/src/github.com/GoogleCloudPlatform/kubernetes/_output/dockerized/go/src/github.com/GoogleCloudPlatform/kubernetes/pkg/kubelet/server.go:240 +0x45","tag":"kubelet","@timestamp":"2015-01-08T18:30:56+00:00"}}]}}
```
A [demonstration](../cluster/addons/fluentd-elasticsearch/logging-demo/README.md) of two synthetic logging sources can be used
to check that logging is working correctly.
Cluster logging can be turned on or off using the environment variable `ENABLE_NODE_LOGGING` which is defined in the
`config-default.sh` file for each provider. For the GCE provider this is set by default to `true`. Set this
to `false` to disable cluster logging.
The type of logging is used is specified by the environment variable `LOGGING_DESTINATION` which for the
GCE provider has the default value `elasticsearch`. If this is set to `gcp` for the GCE provder then
logs will be sent to the Google Cloud Logging system instead.
When using Elasticsearch the number of Elasticsearch instances can be controlled by setting the
variable `ELASTICSEARCH_LOGGING_REPLICAS` which has the default value of `1`. For large clusters
or clusters that are generating log information at a high rate you may wish to use more
Elasticsearch instances.
Reference in a new issue Copy permalink