Skip to content

Debugging Elasticsearch

You are viewing an outdated version of the documentation.
Click here to switch to the stable version (v6.6), or use the version switcher on the left to navigate between versions.

Debugging Elasticsearch

Overview

This article shows you how to debug the status and indexing process of your Elasticsearch environment. Ensure that the Debug-Mode is activated in your .env file.

Shopware 6 CLI commands

Cache clear

cache:clear clears the cache

bash
bin/console cache:clear

> Output:

bash
// Clearing the cache for the dev environment with debug
// true
​[OK] Cache for the "dev" environment (debug=true) was successfully cleared.

ES index

es:index creates only the index for ES

bash
bin/console es:index // Creates only the index for ES

> No Output

ES create alias

es:create:alias will create an alias linking to the index after es:index is done. Normally this is done automatically. In the older version, this has to be done.

bash
bin/console es:create:alias

> No Output

DAL refresh index

dal:refresh:index --use-queue creates a complete reindex from the Shopware DAL (ES/SEO/Media/Sitemap...) ALWAYS "--use-queue" since big request can outperform the server!

bash
bin/console dal:refresh:index --use-queue

> Output:

bash
[landing_page.indexer]
1/1 [============================] 100% < 1 sec/< 1 sec 38.5 MiB

[product.indexer]
22/22 [============================] 100% < 1 sec/< 1 sec 40.5 MiB

[customer.indexer]
2/2 [============================] 100% < 1 sec/< 1 sec 40.5 MiB

[sales_channel.indexer]
2/2 [============================] 100% < 1 sec/< 1 sec 40.5 MiB

[category.indexer]
9/9 [============================] 100% < 1 sec/< 1 sec 40.5 MiB

[...]

Messenger consume

messenger:consume -vv starts a message consumer working on all tasks. This could be started X times. When using more than 3 message consumers, you will need something like RabbitMq to handle the data.

bash
bin/console messenger:consume -vv

> Output:

bash
[OK] Consuming messages from transports "default".
​​
// The worker will automatically exit once it has received a stop signal via the messenger:stop-workers command.

// Quit the worker with CONTROL-C.

09:47:28 INFO      [messenger] Received message Shopware\Elasticsearch\Framework\Indexing\ElasticsearchIndexingMessage ["message" => Shopware\Elasticsearch\Framework\Indexing\ElasticsearchIndexingMessage^ { …},"class" => "Shopware\Elasticsearch\Framework\Indexing\ElasticsearchIndexingMessage"]

[...]

Index cleanup

es:index:cleanup to remove unused indices, because each indexing will generate a new Elasticsearch index.

bash
bin/console es:index:cleanup

Helpful Elasticsearch REST APIs

bash
curl -XGET 'http://elasticsearch:9200/?pretty'

> Output:

bash
{
  "name" : "TZzynG6",
  "cluster_name" : "docker-cluster",
  "cluster_uuid" : "tHklOFWPSwm-j8Yn-8PRoQ",
  "version" : {
    "number" : "6.8.1",
    "build_flavor" : "default",
    "build_type" : "docker",
    "build_hash" : "1fad4e1",
    "build_date" : "2019-06-18T13:16:52.517138Z",
    "build_snapshot" : false,
    "lucene_version" : "7.7.0",
    "minimum_wire_compatibility_version" : "5.6.0",
    "minimum_index_compatibility_version" : "5.0.0"
  },
  "tagline" : "You Know, for Search"
}

API for cluster health

Returns the health status of a cluster:

bash
curl -XGET 'http://elasticsearch:9200/_cluster/health?pretty'

> Output:

bash
{
  "cluster_name" : "docker-cluster",
  "status" : "yellow",
  "timed_out" : false,
  "number_of_nodes" : 1,
  "number_of_data_nodes" : 1,
  "active_primary_shards" : 1210,
  "active_shards" : 1210,
  "relocating_shards" : 0,
  "initializing_shards" : 0,
  "unassigned_shards" : 1210,
  "delayed_unassigned_shards" : 0,
  "number_of_pending_tasks" : 0,
  "number_of_in_flight_fetch" : 0,
  "task_max_waiting_in_queue_millis" : 0,
  "active_shards_percent_as_number" : 50.0
}

API for cat indices

Returns high-level information about indices in a cluster, including backing indices for data streams:​

bash
curl -XGET 'http://elasticsearch:9200/_cat/indices/?pretty'

> Output:

bash
yellow open sw1_manufacturer_20210906113224 AYKMT4NJS7eZgU29ww7z6Q 5 1  3 0  33.2kb  33.2kb
yellow open sw1_emotion_20210903165112      he19OP_UR3mMIAKI7ry2mg 5 1  1 0  11.6kb  11.6kb
yellow open sw1_emotion_20210903171353      jBzApKujRPu73CkKA79F7w 5 1  1 0  11.6kb  11.6kb
yellow open sw1_synonym_20210903175037      EexqHsXyTK202XsalUednQ 5 1  1 0     6kb     6kb
yellow open sw1_synonym_20210903170128      NRjlZZ3AQ0Wat1ILB_9L8Q 5 1  0 0   1.2kb   1.2kb

[...]

API to delete the index

With _all it will delete all indices.

bash
curl -X DELETE 'elasticsearch:9200/_all'

> Output:

bash
{"acknowledged":true}

Show the indexing status in the database

Returns the status of your indexing:

sql
select * from message_queue_stats mqs ; 
select count(*) from enqueue e ; 
select count(*) from dead_message dm ;

The number of entries in the enqueue should match the sum of the size values in the message_queue_stats. As long as there are entries in your enqueue, the indexing is in process and your message consumer has to work those messages.

Reset the indexing in the database

Sometimes you want to reset the indexing in your database because your indexing is stuck or you run into an error. If the database queue is used, third-party services will differ. You can do so with the following queries.

sql
truncate enqueue ; 
truncate dead_message ;
truncate message_queue_stats ;
update scheduled_task set status = 'scheduled' where status = 'queued';

Completely reset your Elasticsearch and reindex

This is mainly for debugging purposes and is only meant for testing and staging environments. First, execute the database reset (only working for the database queue):

sql
truncate enqueue ; 
truncate dead_message ;
truncate message_queue_stats ;
update scheduled_task set status = 'scheduled' where status = 'queued';

Now delete the old Elasticsearch index, clear your cache, reindex and ensure that the indexing process is finished:

bash
curl -X DELETE 'elasticsearch:9200/_all'
bin/console cache:clear
bin/console es:index
bin/console messenger:consume -vv

After the last message has been processed, your index should be found in your Storefront else execute:

bash
bin/console es:create:alias

Logfiles and tipps

You can usually find the Elasticsearch logfiles at /var/log/elasticsearch to check for any issues when indexing. Also, tools like Kibana or Cerebro can help you better understand what is happening in your Elasticsearch.