deviantony/docker-elk

The Elastic stack powered by Docker and Compose.

Antoine Cotten

@antoineco

@TamilTintin if Logstash gets restarted, Filebeat will automatically queue what it's doing until Logstash is available again. First it will retry every 1s, then it will increase the retry delay progressively up to 60s. See backoff.init and backoff.max settings at Configure the Logstash output)

1 reply

Mansour

@Mansour-J

Hi all, I am trying to use this ELK docker-compose file and push auditbeat logs to elasticsearch. The whole set up is running in a single host. The docker-compose for runs successfully and I have auditbeat installed locally. Auditbeat has a command that test if it can connect to elasticsearch auditbeat test output

When I run the auditbeat connectivity test, I get:

elasticsearch: http://localhost:9200...
  parse url... OK
  connection...
    parse host... OK
    dns lookup... OK
    addresses: ::1, 127.0.0.1
    dial up... OK
  TLS... WARN secure connection disabled
  talk to server... ERROR Connection marked as failed because the onConnect callback failed: cannot retrieve the elasticsearch license from the /_license endpoint, Auditbeat requires the default distribution of Elasticsearch. Please make the endpoint accessible to Auditbeat so it can verify the license.: could not extract license information from the server response: unknown state, received: 'expired'

but when I pull down docker pull docker.elastic.co/kibana/kibana:7.9.2 image directly and run it - it works.

I have tried changing the xpack.license.self_generated.type: trial to basic as well

I am running it on CentOs 7

Antoine Cotten

@antoineco

@Mansour-J are you running a version of Beats which matches the stack version (7.9)? If yes, did you provide ES credentials in your beats config file? You can try with the elastic user (super-admin) first, and if it works, consider switching to a role with lower privileges.

Mansour

@Mansour-J

@antoineco Yes, the auditbeat version matches(not considering patch number) the elasticsearch version and I used the default elastic and changeme as a password(which I think is the super user)

auditbeat version 7.9.2 (amd64), libbeat 7.9.2 [2ab907f5ccecf9fd82fe37105082e89fd871f684 built 2020-09-22 23:14:31 +0000 UTC]

{
  "name" : "19780697fd9b",
  "cluster_name" : "docker-cluster",
  "cluster_uuid" : "CMKnZU0KTZKuOWyAbZJEpg",
  "version" : {
    "number" : "7.9.1",
    "build_flavor" : "default",
    "build_type" : "docker",
    "build_hash" : "083627f112ba94dffc1232e8b42b73492789ef91",
    "build_date" : "2020-09-01T21:22:21.964974Z",
    "build_snapshot" : false,
    "lucene_version" : "8.6.2",
    "minimum_wire_compatibility_version" : "6.8.0",
    "minimum_index_compatibility_version" : "6.0.0-beta1"
  },
  "tagline" : "You Know, for Search"
}

Antoine Cotten

@antoineco

OK, looks good to me. I just noticed the word expired at the end of the error. Could be that Auditbeat requires a paid license, in which case trial (our default) should work for 30 days. Was your stack initialized more than 30 days ago?

Also, please note that Kibana does work just fine with an expired trial license, it just doesn't show all the features (and it should show a warning banner).

Antoine Cotten

@antoineco

I thought beats also worked that way, let me try to reproduce.

Mansour

@Mansour-J

@antoineco I am not sure if Auditbeat requires paid license. However, swapping the docker-compose elastic by running:

docker run -p 9200:9200 -p 9300:9300 -e "discovery.type=single-node" docker.elastic.co/elasticsearch/elasticsearch:7.9.2

Gives the following output:

elasticsearch: http://localhost:9200...
  parse url... OK
  connection...
    parse host... OK
    dns lookup... OK
    addresses: ::1, 127.0.0.1
    dial up... OK
    talk to server... OK

Antoine Cotten

@antoineco

@Mansour-J So it might be an authorization error, because the only difference with the default setup is the absence of xpack options (security). Maybe auditbeat test output didn't read the credentials from your file, did you double check they were defined correctly?

Mansour

@Mansour-J

@antoineco Interesting, because auditbeat is installed at host while elastic search is in container. I would assume that the password in auditbeat would be the same for both of them.

Antoine Cotten

@antoineco

@Mansour-J I mean, you bypassed the docker-elk config with your Docker command, and didn't enable Xpack Security on the command line. By default, ES doesn't require any credentials.

Antoine Cotten

@antoineco

@Mansour-J I wasn't able to reproduce:

$ docker run \
  --cap-add="AUDIT_CONTROL" --cap-add="AUDIT_READ" \
  docker.elastic.co/beats/auditbeat:7.9.2 \
  test output \
  -E 'output.elasticsearch.hosts=["host.docker.internal:9200"]' -E 'output.elasticsearch.username=elastic' -E 'output.elasticsearch.password=changeme'

elasticsearch: http://host.docker.internal:9200...
  parse url... OK
  connection...
    parse host... OK
    dns lookup... OK
    addresses: 192.168.65.2
    dial up... OK
  TLS... WARN secure connection disabled
  talk to server... OK
  version: 7.9.1

(I use host.docker.internal because I run Docker for Desktop, but in your case localhost should work just fine)

RoD

@rodrigoquijarro

Hello everyone!, I am trying to use this ELK docker-compose file and push filebeat logs to elasticsearch. The whole setup is running in a different host (same network). The docker-compose runs successfully when filebeat is installed locally, but no remote. Running the validation command for Filebeat shows the following output. We have checked all possible issues with network configuration and by now everything is ok, however, we're thinking 'discovery.type: single-node' environmental parameter could be a possible issue.
$ sudo filebeat test output
elasticsearch: http://10.2.2.2:5044...
parse url... OK
connection...
parse host... OK
dns lookup... OK
addresses: 10.1.1.20
dial up... ERROR dial tcp 10.1.1.20:5044: i/o timeout
I'm running both machines on Debian 10.

Antoine Cotten

@antoineco

@rodrigoquijarro discovery.type: single-node shouldn't be related to your issue here.
Are you sending your logs to Logstash or directly to Elasticsearch? By default, port 5044 is reserved for Logstash's beats input. If you want to send directly to Elasticsearch, you have to use port 9200, not 5044.
See https://www.elastic.co/guide/en/beats/filebeat/current/configuring-output.html

Antoine Cotten

@antoineco

Here is a quick test I ran locally:

$ docker-compose up -d elasticsearch

$ docker run docker.elastic.co/beats/filebeat:7.9.3 \
    test output \
    -E 'output.elasticsearch.hosts=["host.docker.internal:9200"]' -E 'output.elasticsearch.username=elastic' -E 'output.elasticsearch.password=changeme'

elasticsearch: http://host.docker.internal:9200...
  parse url... OK
  connection...
    parse host... OK
    dns lookup... OK
    addresses: 192.168.65.2
    dial up... OK
  TLS... WARN secure connection disabled
  talk to server... OK
  version: 7.9.2

Antoine Cotten

@antoineco

Another test, this time with Logstash:

$ docker-compose up -d elasticsearch logstash

#  blank config for testing
$ touch filebeat.yml

$ docker run docker.elastic.co/beats/filebeat:7.9.3 \
    -v ${PWD}/filebeat.yml:/usr/share/filebeat/filebeat.yml \
    test output \
    -E 'output.logstash.hosts=["host.docker.internal:5044"]'

logstash: host.docker.internal:5044...
  connection...
    parse host... OK
    dns lookup... OK
    addresses: 192.168.65.2
    dial up... OK
  TLS... WARN secure connection disabled
  talk to server... OK

RoD

@rodrigoquijarro

Thank you very much for your replay @antoineco , I tried to sent logs to logstash, elasticsearch, separately and together. Locally works, yes, however when I try to collect them remotely (from a external host with filebeat installed) its show the timeout message.

Antoine Cotten

@antoineco

@rodrigoquijarro that probably means you have a firewall blocking TCP packets to ports 9200 and 5044 between Filebeat and the machine running the stack. Or maybe the IP address is simply not routable?

Mohammed Gaber

@mgabs

Hi all, I'm new to elk and have been trying to run the stack for a while now with no joy

Elastic is working fine, logstash faces ES unreachable and kibana dies with an error

kibana_1         | {"type":"log","@timestamp":"2020-11-03T14:14:59Z","tags":["fatal","root"],"pid":6,"message":"Error: Setup lifecycle of \"monitoring\" plugin wasn't completed in 30sec. Consider disabling the plugin and re-start.\n    at Timeout.setTimeout (/usr/share/kibana/src/core/utils/promise.js:31:90)\n    at ontimeout (timers.js:436:11)\n    at tryOnTimeout (timers.js:300:5)\n    at listOnTimeout (timers.js:263:5)\n    at Timer.processTimers (timers.js:223:10)"}
kibana_1         | {"type":"log","@timestamp":"2020-11-03T14:14:59Z","tags":["info","plugins-system"],"pid":6,"message":"Stopping all plugins."}
kibana_1         | 
kibana_1         |  FATAL  Error: Setup lifecycle of "monitoring" plugin wasn't completed in 30sec. Consider disabling the plugin and re-start.
kibana_1         | 
elasticsearch_1  | {"type": "server", "timestamp": "2020-11-03T14:15:00,579Z", "level": "INFO", "component": "o.e.c.m.MetadataCreateIndexService", "cluster.name": "docker-cluster", "node.name": "d72b08299b6b", "message": "[ilm-history-2-000001] creating index, cause [api], templates [ilm-history], shards [1]/[0]", "cluster.uuid": "OmP8_jOLQ_C-NUQEPRIb1A", "node.id": "EF4mNoLbSs-aLiiv_uBFmw"  }
elasticsearch_1  | {"type": "server", "timestamp": "2020-11-03T14:15:00,664Z", "level": "INFO", "component": "o.e.x.i.IndexLifecycleTransition", "cluster.name": "docker-cluster", "node.name": "d72b08299b6b", "message": "moving index [ilm-history-2-000001] from [null] to [{\"phase\":\"new\",\"action\":\"complete\",\"name\":\"complete\"}] in policy [ilm-history-ilm-policy]", "cluster.uuid": "OmP8_jOLQ_C-NUQEPRIb1A", "node.id": "EF4mNoLbSs-aLiiv_uBFmw"  }
elasticsearch_1  | {"type": "server", "timestamp": "2020-11-03T14:15:00,845Z", "level": "INFO", "component": "o.e.x.i.IndexLifecycleTransition", "cluster.name": "docker-cluster", "node.name": "d72b08299b6b", "message": "moving index [ilm-history-2-000001] from [{\"phase\":\"new\",\"action\":\"complete\",\"name\":\"complete\"}] to [{\"phase\":\"hot\",\"action\":\"unfollow\",\"name\":\"wait-for-indexing-complete\"}] in policy [ilm-history-ilm-policy]", "cluster.uuid": "OmP8_jOLQ_C-NUQEPRIb1A", "node.id": "EF4mNoLbSs-aLiiv_uBFmw"  }
elasticsearch_1  | {"type": "server", "timestamp": "2020-11-03T14:15:00,913Z", "level": "INFO", "component": "o.e.c.r.a.AllocationService", "cluster.name": "docker-cluster", "node.name": "d72b08299b6b", "message": "Cluster health status changed from [YELLOW] to [GREEN] (reason: [shards started [[ilm-history-2-000001][0]]]).", "cluster.uuid": "OmP8_jOLQ_C-NUQEPRIb1A", "node.id": "EF4mNoLbSs-aLiiv_uBFmw"  }
elasticsearch_1  | {"type": "server", "timestamp": "2020-11-03T14:15:00,991Z", "level": "INFO", "component": "o.e.x.i.IndexLifecycleTransition", "cluster.name": "docker-cluster", "node.name": "d72b08299b6b", "message": "moving index [ilm-history-2-000001] from [{\"phase\":\"hot\",\"action\":\"unfollow\",\"name\":\"wait-for-indexing-complete\"}] to [{\"phase\":\"hot\",\"action\":\"unfollow\",\"name\":\"wait-for-follow-shard-tasks\"}] in policy [ilm-history-ilm-policy]", "cluster.uuid": "OmP8_jOLQ_C-NUQEPRIb1A", "node.id": "EF4mNoLbSs-aLiiv_uBFmw"  }
devian_kibana_1 exited with code 1
logstash_1       | [2020-11-03T14:18:14,409][ERROR][logstash.licensechecker.licensereader] Unable to retrieve license information from license server {:message=>"No Available connections"}
logstash_1       | [2020-11-03T14:18:14,524][WARN ][logstash.licensechecker.licensereader] Attempted to resurrect connection to dead ES instance, but got an error. {:url=>"http://elastic:xxxxxx@elasticsearch:9200/", :error_type=>LogStash::Outputs::ElasticSearch::HttpClient::Pool::HostUnreachableError, :error=>"Elasticsearch Unreachable: [http://elastic:xxxxxx@elasticsearch:9200/][Manticore::ConnectTimeout] connect timed out"}
logstash_1       | [2020-11-03T14:18:22,279][WARN ][logstash.outputs.elasticsearch][main] Attempted to resurrect connection to dead ES instance, but got an error. {:url=>"http://elastic:xxxxxx@elasticsearch:9200/", :error_type=>LogStash::Outputs::ElasticSearch::HttpClient::

I reverted back to repo with git sync and tried again but no joy

apperciate your help

Antoine Cotten

@antoineco

@mgabs seems like networking issues to me. Are you running on CentOS?

Antoine Cotten

@antoineco

If yes, could you please temporarily set SELinux in permissive mode with sudo setenforce 0? (you can re-enable it after we confirm it's not the source of your issue)

Mohammed Gaber

@mgabs

I'm running arch, I am doubting that now as well

not running SEL, only nftables - firewalld

Antoine Cotten

@antoineco

@mgabs then it's very likely that you're running into exactly that issue: https://github.com/deviantony/docker-elk/issues/541#issuecomment-707572713

Firewalld doesn't play nice at all with Docker (Compose especially) and requires some extra configuration to avoid interfering with Docker-generated rules.

Mohammed Gaber

@mgabs

I confirm it's an internet access / networking problem

tried the fix you linked, no joy

Mohammed Gaber

@mgabs

I got the stack running after disabling nftables, thanks

Antoine Cotten

@antoineco

The "fix" depends on the IP address allocated to the docker-elk bridge, so I wouldn't recommend copying those commands.
Glad you managed to isolate the problem though :+1: I hope you find a way to configure firewalld in a suitable way.

Mohammed Gaber

@mgabs

Since i had to dig deeper to get it to work, I wanted to share that It's not really firewalld (only a management interface)

The underlying issue however is iptables vs nftables and the fact that docker doesn't have full support for nftables

1 reply

timiil

@timiil

Please here there is a question.
we have to store a 'Checkin' doc into ES index , we want count a lot of value metrics about this 'Checkin',
for Tenant, for Entity, for Month/Week/Date/Hour/LastMinute/LastHour... the problem is :
should we use which product in ElasticSearch for these usage ? or should I just use Promethues to do this
task ?

soumen0

@soumen0

I am unable to create a docker image for logstash, it is actually failing in the lines ...
ADD config/ /usr/share/logstash/config/ and
ADD pipeline/ /usr/share/logstash/pipeline/
Although both the logstash.conf and logstash.yml exists in the same directory with docker file.
I am trying to customize using the following logstash docker 7.9.3

Antoine Cotten

@antoineco

@soumen0 are you using Compose on macOS or Windows, by any chance?
Also could you please elaborate on "it is actually failing in the lines". What error are you seeing?

neethujacobmec

@neethujacobmec

Hi Everyone, I am using kinesis agent to stream from application logs and lambda function to separate the logs based on some indices. However I am seeing the below error intermittently.

{ "error": { "root_cause": [ { "type": "remote_transport_exception", "reason": "[2f4424726309a9300ffdd0939e5cca77][x.x.x.x:9300][indices:data/write/bulk[s]]" } ], "type": "es_rejected_execution_exception", "reason": "rejected execution of processing of [195195722][indices:data/write/bulk[s][p]]: request: BulkShardRequest [[misc-logs-2020-11-16][0]] containing [index {[misc-logs-2020-11-16][_doc][J9_q0HUBq-O-de1zDKJd], source[{\"message\":\"\\"End of file or no input: Operation interrupted or timed out (60 s recv delay) (60 s send delay)\\"\"}]}], target allocation id: yx7QdcoSRkecIXzgqLLeLg, primary term: 1 on EsThreadPoolExecutor[name = 2f4424726309a9300ffdd0939e5cca77/write, queue capacity = 200, org.elasticsearch.common.util.concurrent.EsThreadPoolExecutor@5205cb35[Running, pool size = 2, active threads = 2, queued tasks = 200, completed tasks = 98568339]]" }, "status": 429 }

I am using multiline option set in agent.json to read multiline log statements together.

AWS Kinesis Agent offers settings like number of nodes and shards. However unsure where queue size may be increased as elasticsearch.yaml doesn't seem to be accessible. Any help would be appreciated.

Antoine Cotten

@antoineco

@neethujacobmec you can adjust the size of thewrite thread pool as described here: https://www.elastic.co/guide/en/elasticsearch/reference/current/modules-threadpool.html

Alan M Kristensen

@Big-al

Hi, in the TLS setup you write "You will be prompted to enter an optional passphrase to protect both the CA and Elasticsearch keys. Please be aware that the passphrase you enter here, if not empty, will have be manually entered on every restart of Elasticsearch."
Where do i update this password? My Elasticsearch node is failing on startup with a wrong certificate password.

Also - let me just add, this is truely awesome work. Makes it so much easier to get going !

Antoine Cotten

@antoineco

@Big-al glad this was useful!

According to this page, the passphrase needs to be added to Elasticsearch's keystore (see step 3): https://www.elastic.co/guide/en/elasticsearch/reference/current/configuring-tls.html#tls-transport

I'm glad you raised this, I need to update the doc with some more accurate info.

Antoine Cotten

@antoineco

README updated on the tls branch.

Davide Pugliese

@Deviad

Hello

I am trying to use fscrawler on docker-elk, but I get an error unfortunately.
https://discuss.elastic.co/t/failed-to-create-elasticsearch-client/256243

Alan M Kristensen

@Big-al

Thanks for taking quick action Antoine! Im actually going to change and rebuild my production image for a large client based on some of your changes here. Its great. This makes it a lot easier for new developers to adopt elastic. Huge fan.

Where communities thrive

People

Repo info

Activity