Discovery problem

Elastic Stack Beats
1

Hi,

I am trying to implement a sample example with autodiscover and modules

I created 2 docker :

  • nginx
  • filebeat

I want filebeat container to get logs from nginx container with autodiscover and use nginx module to parse them and send to elastic.

Here is my filebeat.yml, mounted in my container.

`
filebeat.modules:

  • modules:
    nginx:
    enabled: true

filebeat.autodiscover:
providers:
- type: docker
templates:
- condition:
contains:
docker.container.image: nginx
config:
- module: nginx
log:
input:
type: docker
containers.ids:
- "${data.docker.container.id}"

output.elasticsearch:
hosts: ["https://elasticsearch:9200"]
`
I doesn't work but I am wondering why?

I added labels to nginx container (docker-compose) :
labels:
co.elastic.logs/module: nginx
co.elastic.logs/fileset.stdout: access
co.elastic.logs/fileset.stderr: error

any idea?

Regards

2

Kindly format your configuration with </> for better clarity of your configuration.

3

Any luck on this one? I'm having the same issue right now.

As I see it, this is the relevant part for my Nginx container in docker-compose.yml:

nginx:
  container_name: nginx
  build: ./nginx-vts/.
  labels:
    co.elastic.logs/disable: false
    co.elastic.logs/module: nginx
    co.elastic.logs/fileset.stdout: access
    co.elastic.logs/fileset.stderr: error

filebeat:
  container_name: filebeat
  image: docker.elastic.co/beats/filebeat:7.0.0
  user: root
  volumes:
    - ${MY_DOCKER_DATA_DIR}/filebeat/nginx-access-ingest.json:/usr/share/filebeat/module/nginx/access/ingest/default.json
    - ${MY_DOCKER_DATA_DIR}/filebeat/filebeat.autodiscover.docker.yml:/usr/share/filebeat/filebeat.yml:ro
    - filebeat_data:/usr/share/filebeat/data
    - /var/run/docker.sock:/var/run/docker.sock
    - /var/lib/docker/containers/:/var/lib/docker/containers/:ro
    - /var/log/:/var/log/:ro
  environment:
    - ELASTICSEARCH_HOST=elasticsearch:9200
    - KIBANA_HOST=kibana:5601
    command: ["--strict.perms=false"]

My filebeat.yml is deemed Config OK on filebeat test config:

filebeat.config:
  modules:
    path: ${path.config}/modules.d/*.yml
    reload.enabled: true

filebeat.modules:
- module: nginx

filebeat.autodiscover:
  providers:
    - type: docker
      hints.enabled: true
      default.disable: true

output.elasticsearch:
  hosts: 'elasticsearch:9200'

setup.kibana:
  host: "kibana:5601"

I verified that the nginx module is enabled inside my filebeat container. Data is being sent to Elasticsearch, but I get the following error.message when I inspect a log entry in Kibana:

Provided Grok expressions do not match field value: [my.domain.com 192.168.1.1 - my_user [20/Apr/2019:20:21:37 +0000] \"GET /ocs/v2.php/apps/serverinfo/api/v1/info HTTP/2.0\" 200 764 \"-\" \"Go-http-client/2.0\"]

I've used the Kibana dev tools to evaluate the Grok-pattern I use for Nginx, that is present in the following file:

/usr/share/filebeat/module/nginx/access/ingest/default.json

My altered Nginx pattern works in the debugger. This is the content of my grok "patterns": [...] for my nginx access pipeline:

"\"%{IPORHOST:nginx.access.host} %{IPORHOST:nginx.access.remote_ip_list} - %{DATA:user.name} \\[%{HTTPDATE:nginx.access.time}\\] \"%{WORD:http.request.method} %{DATA:url.original} HTTP/%{NUMBER:http.version}\" %{NUMBER:http.response.status_code:long} %{NUMBER:http.response.body.bytes:long} \"%{DATA:http.request.referrer}\" \"%{DATA:user_agent.original}\"\""

The pattern I actually input in the debugger, and that works:

%{IPORHOST:nginx.access.host} %{IPORHOST:nginx.access.remote_ip_list} - %{DATA:user.name} \[%{HTTPDATE:nginx.access.time}\] \"%{WORD:http.request.method} %{DATA:url.original} HTTP/%{NUMBER:http.version}\" %{NUMBER:http.response.status_code:long} %{NUMBER:http.response.body.bytes:long} \"%{DATA:http.request.referrer}\" \"%{DATA:user_agent.original}\"

So what I'm looking for is how to debug my setup - or tips on what I'm doing wrong so that the fields are properly "nginx-parsed" when they enter Elasticsearch.

Sorry for hijacking the thread - but I hope OP already found the answer.

4

I finally solved this!

For anyone finding this post, I'll go ahead and say what solved it for me.

Almost all my configs were spot on. What made a difference started with doing a slight adjustment in my filebeat.yml; I added the following:

setup.template.overwrite: true

Why? The first time you set up filebeat, if you follow the setup procedures, you are supposed to do something like this:

# enable module
./filebeat modules enable nginx
# check that your module list is ok
./filebeat modules list
# set up your output (in my case elastic) and kibana
./filebeat setup -e

The thing is, I re-defined both the enabled modules and the Grok pattern for my Nginx module access ingest pipeline after the first time I ran the initial setup procedure. So when I ran the setup command after changing my module list later on, it didn't do anything to Elasticsearch, because the default behaviour is to leave the index alone. Setting "setup.template.overwrite: true" made the difference! My Grok pattern "took" and now things are working as expected.

I suggest updating the docs to make this a little more obvious for beginners. Although I caught up on it after a few days of reading the docs up and down times over, I'm still not completely sure if I have to reload the templates into Elasticsearch when I redefine my Grok pattern but I suppose it's needed if what the template is doing is creating field definitions in Elasticsearch based upon the Grok pattern matching definition.