[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Logging (aggregated, FluentD)



Hi Rich,

Thanks.  So first should I not be pointing Kibana at the same host as my OSE Web Interface?  If not, how would you suggest to install/reconfigure.  I'm in AWS and I don't really follow what you mean by openshift.deployment.subdomain.  Would that be another instance in AWS?

Re:  OSE version - yes I'm running OSE 3.4/Kube 1.4.  Here's the output of oc version:
===========================

[root ip-172-31-45-158 ~]# oc version

oc v3.4.1.12

kubernetes v1.4.0+776c994

features: Basic-Auth GSSAPI Kerberos SPNEGO

Server https://ip-172-31-45-158.us-east-2.compute.internal:8443

openshift v3.4.1.12

kubernetes v1.4.0+776c994

===========================

Here's the fluentD conf:

[root ip-172-31-45-158 ~]# oc project logging 

Now using project "logging" on server "https://ip-172-31-45-158.us-east-2.compute.internal:8443".

[root ip-172-31-45-158 ~]# oc get configmap logging-fluentd -o yaml

apiVersion: v1

data:

  fluent.conf: |

    # This file is the fluentd configuration entrypoint. Edit with care.

    @include configs.d/openshift/system.conf

    # In each section below, pre- and post- includes don't include anything initially;

    # they exist to enable future additions to openshift conf as needed.

    ## sources

    ## ordered so that syslog always runs last...

    @include configs.d/openshift/input-pre-*.conf

    @include configs.d/dynamic/input-docker-*.conf

    @include configs.d/dynamic/input-syslog-*.conf

    @include configs.d/openshift/input-post-*.conf

    ##

    <label @INGRESS>

    ## filters

      @include configs.d/openshift/filter-pre-*.conf

      @include configs.d/openshift/filter-retag-journal.conf

      @include configs.d/openshift/filter-k8s-meta.conf

      @include configs.d/openshift/filter-kibana-transform.conf

      @include configs.d/openshift/filter-k8s-record-transform.conf

      @include configs.d/openshift/filter-syslog-record-transform.conf

      @include configs.d/openshift/filter-common-data-model.conf

      @include configs.d/openshift/filter-post-*.conf

    ##

    ## matches

      @include configs.d/openshift/output-pre-*.conf

      @include configs.d/openshift/output-operations.conf

      @include configs.d/openshift/output-applications.conf

      # no post - applications.conf matches everything left

    ##

    </label>

  secure-forward.conf: |

    # @type secure_forward


    # self_hostname ${HOSTNAME}

    # shared_key <SECRET_STRING>


    # secure yes

    # enable_strict_verification yes


    # ca_cert_path /etc/fluent/keys/your_ca_cert

    # ca_private_key_path /etc/fluent/keys/your_private_key

      # for private CA secret key

    # ca_private_key_passphrase passphrase


    # <server>

      # or IP

    #   host server.fqdn.example.com

    #   port 24284

    # </server>

    # <server>

      # ip address to connect

    #   host 203.0.113.8

      # specify hostlabel for FQDN verification if ipaddress is used for host

    #   hostlabel server.fqdn.example.com

    # </server>

  throttle-config.yaml: |

    # Logging example fluentd throttling config file


    #example-project:

    #  read_lines_limit: 10

    #

    #.operations:

    #  read_lines_limit: 100

kind: ConfigMap

metadata:

  creationTimestamp: 2017-04-12T17:20:48Z

  labels:

    logging-infra: support

  name: logging-fluentd

  namespace: logging

  resourceVersion: "188321"

  selfLink: /api/v1/namespaces/logging/configmaps/logging-fluentd

  uid: 5ee29731-1fa4-11e7-b524-0a7a32c48dc3

========================

I'm happy to give you access to my environment if that would help.  


On Tue, Apr 18, 2017 at 4:47 PM, Rich Megginson <rmeggins redhat com> wrote:
On 04/18/2017 01:51 PM, Shepp wrote:
Hello,

I've posted over on the FluentD Google Groups but was directed here.
https://groups.google.com/forum/#!topic/fluentd/Uo2E6kQzM5E <https://groups.google.com/forum/#%21topic/fluentd/Uo2E6kQzM5E>

I've got an OpenShift test lab in AWS, all the Aggregated Logging PODs are deployed and running, and I believe I've also configured the OpenShift Master Node master-config.yaml to display the FluentD aggregated logs in the OpenShift Web Interface, however I don't see the aggregated logs for any of our running images.  I don't think the logs are throwing any kind of severe error that I can tell is causing this to not work.  What I think might be a problem is in the KIBANA 'ROUTES' in OpenShift.  I've attached a .png of the routes, one thing I can say is the logging-kibana route looks close to right but when I try to open that URL there's nothing there at that URL. In this setup port 8443 is open, not 443 (if that is what Kibana is trying to use), and the OpenShift console is at URL/console, yet that and the port seems to have been ignored?

The way Routes work in OpenShift is that the hostname is used as the discriminator for the route.  I'm not sure if you can use the same hostname for both the OpenShift console and Kibana even though they use different ports.  Usually, Kibana is deployed at kibana.{{ openshift deployment subdomain }} and you will have to set up either DNS to have the correct IP address for kibana.{{ openshift deployment subdomain }} or just hack your /etc/hosts file.  The Route listens at host on port 443 (https).  It will redirect requests to the logging-kibana Service, which may be on a different port.

Also the second ROUTE kibna-ops.example.com <http://kibna-ops.example.com/> - that is obvisouly some sort of default URL, what would this need to be set to?

You can ignore it if you deployed without the separate ops cluster.


In the I've copied out and zipped each PODs log file. and the image of the ROUTES for Kibana. I don't know what else to check.


2017-04-13 06:30:18 -0400 [warn]: temporarily failed to flush the buffer. next_retry=2017-04-13 06:30:19 -0400 error_class="NoMethodError" error="undefined method `status' for nil:NilClass" plugin_id="object:1fe9bb4"
  2017-04-13 06:30:18 -0400 [warn]: /usr/share/gems/gems/elasticsearch-transport-1.0.18/lib/elasticsearch/transport/transport/base.rb:265:in `rescue in perform_request'
  2017-04-13 06:30:18 -0400 [warn]: /usr/share/gems/gems/elasticsearch-transport-1.0.18/lib/elasticsearch/transport/transport/base.rb:247:in `perform_request'
  2017-04-13 06:30:18 -0400 [warn]: /usr/share/gems/gems/elasticsearch-transport-1.0.18/lib/elasticsearch/transport/transport/http/faraday.rb:20:in `perform_request'
  2017-04-13 06:30:18 -0400 [warn]: /usr/share/gems/gems/elasticsearch-transport-1.0.18/lib/elasticsearch/transport/client.rb:128:in `perform_request'

Are you using 1.4/3.4?  And did you recently upgrade from 1.3/3.3?

If so, looks like your fluent.conf is incorrect.  Can you provide this?

oc get configmap logging-fluentd -o yaml


Any help would be greatly appreciated.

Thanks and best wishes,

Joel

Inline image 2



_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users


_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]