Merge branch '201826-es-field-length-limit' into 'master'

Application limit for ES indexed field length

See merge request gitlab-org/gitlab!24345

db/migrate/20200204070729_add_elasticsearch_indexed_field_length_limit_to_application_settings.rb

+# frozen_string_literal: true
+
+class AddElasticsearchIndexedFieldLengthLimitToApplicationSettings < ActiveRecord::Migration[6.0]
+  DOWNTIME = false
+
+  def up
+    add_column :application_settings, :elasticsearch_indexed_field_length_limit, :integer, null: false, default: 0
+
+    if Gitlab.com?
+      execute 'UPDATE application_settings SET elasticsearch_indexed_field_length_limit = 20000'
+    end
+  end
+
+  def down
+    remove_column :application_settings, :elasticsearch_indexed_field_length_limit
+  end
+end

db/schema.rb

@@ -343,6 +343,7 @@ ActiveRecord::Schema.define(version: 2020_02_06_111847) do
     t.boolean "force_pages_access_control", default: false, null: false
     t.boolean "updating_name_disabled_for_users", default: false, null: false
     t.integer "instance_administrators_group_id"
+    t.integer "elasticsearch_indexed_field_length_limit", default: 0, null: false
     t.index ["custom_project_templates_group_id"], name: "index_application_settings_on_custom_project_templates_group_id"
     t.index ["file_template_project_id"], name: "index_application_settings_on_file_template_project_id"
     t.index ["instance_administration_project_id"], name: "index_applicationsettings_on_instance_administration_project_id"

doc/administration/instance_limits.md

@@ -86,3 +86,22 @@ Plan.default.limits.update!(ci_active_jobs: 500)
 ```
 
 NOTE: **Note:** Set the limit to `0` to disable it.
+
+## Advanced Global Search limits
+
+### Maximum field length
+
+> [Introduced](https://gitlab.com/gitlab-org/gitlab/issues/201826) in GitLab 12.8.
+
+You can set a limit on the content of text fields indexed for Global Search.
+Setting a maximum helps to reduce the load of the indexing processes.  If any
+text field exceeds this limit then the text will be truncated to this number of
+characters and the rest will not be indexed and hence will not be searchable.
+
+- On GitLab.com this is limited to 20000 characters
+- For self-hosted installations it is unlimited by default
+
+This limit can be configured for self hosted installations when [enabling
+Elasticsearch](../integration/elasticsearch.md#enabling-elasticsearch).
+
+NOTE: **Note:** Set the limit to `0` to disable it.

doc/integration/elasticsearch.md

@@ -150,6 +150,7 @@ The following Elasticsearch settings are available:
 | `AWS Region`                                          | The AWS region your Elasticsearch service is located in. |
 | `AWS Access Key`                                      | The AWS access key. |
 | `AWS Secret Access Key`                               | The AWS secret access key. |
+| `Maximum field length`                                | See [the explanation in instance limits.](../administration/instance_limits.md#maximum-field-length). |
 
 ### Limiting namespaces and projects
 

ee/app/helpers/ee/application_settings_helper.rb

@@ -27,6 +27,7 @@ module EE
         :elasticsearch_aws_secret_access_key,
         :elasticsearch_indexing,
         :elasticsearch_replicas,
+        :elasticsearch_indexed_field_length_limit,
         :elasticsearch_search,
         :elasticsearch_shards,
         :elasticsearch_url,

ee/app/models/ee/application_setting.rb

@@ -56,6 +56,10 @@ module EE
                 presence: { message: "can't be blank when using aws hosted elasticsearch" },
                 if: ->(setting) { setting.elasticsearch_indexing? && setting.elasticsearch_aws? }
 
+      validates :elasticsearch_indexed_field_length_limit,
+                presence: true,
+                numericality: { only_integer: true, greater_than_or_equal_to: 0 }
+
       validates :email_additional_text,
                 allow_blank: true,
                 length: { maximum: EMAIL_ADDITIONAL_TEXT_CHARACTER_LIMIT }
@@ -85,6 +89,7 @@ module EE
           elasticsearch_aws_region: ENV['ELASTIC_REGION'] || 'us-east-1',
           elasticsearch_replicas: 1,
           elasticsearch_shards: 5,
+          elasticsearch_indexed_field_length_limit: 0,
           elasticsearch_url: ENV['ELASTIC_URL'] || 'http://localhost:9200',
           email_additional_text: nil,
           lock_memberships_to_ldap: false,

ee/app/views/admin/application_settings/_elasticsearch_form.html.haml

@@ -57,6 +57,12 @@
               = _('How many replicas each Elasticsearch shard has.')
               = recreate_index_text
 
+          .form-group
+            = f.label :elasticsearch_indexed_field_length_limit, _('Maximum field length'), class: 'label-bold'
+            = f.number_field :elasticsearch_indexed_field_length_limit, value: @application_setting.elasticsearch_indexed_field_length_limit, class: 'form-control'
+            .form-text.text-muted
+              = _('If any indexed field exceeds this limit it will be truncated to this number of characters and the rest will not be indexed or searchable. This does not apply to repository and wiki indexing. Setting this to 0 means it is unlimited.')
+
         .sub-section
           %h4= _('Elasticsearch indexing restrictions')
           .form-group

ee/changelogs/unreleased/201826-es-field-length-limit.yml

+---
+title: Add application limit for ES indexed field length
+merge_request: 24345
+author:
+type: added

ee/lib/elastic/instance_proxy_util.rb

@@ -41,10 +41,21 @@ module Elastic
     # them to raise exceptions. When this happens, we still want the remainder
     # of the object to be saved, so silently swallow the errors
     def safely_read_attribute_for_elasticsearch(attr_name)
-      target.send(attr_name) # rubocop:disable GitlabSecurity/PublicSend
+      result = target.send(attr_name) # rubocop:disable GitlabSecurity/PublicSend
+      apply_field_limit(result)
     rescue => err
       target.logger.warn("Elasticsearch failed to read #{attr_name} for #{target.class} #{target.id}: #{err}")
       nil
     end
+
+    def apply_field_limit(result)
+      return result unless result.is_a? String
+
+      limit = Gitlab::CurrentSettings.elasticsearch_indexed_field_length_limit
+
+      return result unless limit > 0
+
+      result[0, limit]
+    end
   end
 end

ee/spec/features/admin/admin_settings_spec.rb

@@ -68,6 +68,7 @@ describe 'Admin updates EE-only settings' do
         check 'Search with Elasticsearch enabled'
         fill_in 'Number of Elasticsearch shards', with: '120'
         fill_in 'Number of Elasticsearch replicas', with: '2'
+        fill_in 'Maximum field length', with: '100000'
 
         click_button 'Save changes'
       end
@@ -77,6 +78,7 @@ describe 'Admin updates EE-only settings' do
         expect(current_settings.elasticsearch_search).to be_truthy
         expect(current_settings.elasticsearch_shards).to eq(120)
         expect(current_settings.elasticsearch_replicas).to eq(2)
+        expect(current_settings.elasticsearch_indexed_field_length_limit).to eq(100000)
         expect(page).to have_content 'Application settings saved successfully'
       end
     end

ee/spec/models/application_setting_spec.rb

@@ -41,6 +41,12 @@ describe ApplicationSetting do
     it { is_expected.not_to allow_value(1.1).for(:elasticsearch_replicas) }
     it { is_expected.not_to allow_value(-1).for(:elasticsearch_replicas) }
 
+    it { is_expected.to allow_value(10).for(:elasticsearch_indexed_field_length_limit) }
+    it { is_expected.to allow_value(0).for(:elasticsearch_indexed_field_length_limit) }
+    it { is_expected.not_to allow_value(nil).for(:elasticsearch_indexed_field_length_limit) }
+    it { is_expected.not_to allow_value(1.1).for(:elasticsearch_indexed_field_length_limit) }
+    it { is_expected.not_to allow_value(-1).for(:elasticsearch_indexed_field_length_limit) }
+
     it { is_expected.to allow_value(nil).for(:required_instance_ci_template) }
     it { is_expected.not_to allow_value("").for(:required_instance_ci_template) }
     it { is_expected.not_to allow_value("  ").for(:required_instance_ci_template) }

ee/spec/models/concerns/elastic/issue_spec.rb

@@ -125,6 +125,32 @@ describe Issue, :elastic do
     expect(issue.__elasticsearch__.as_indexed_json).to eq(expected_hash)
   end
 
+  context 'field length limits' do
+    context 'when there is an elasticsearch_indexed_field_length limit' do
+      it 'truncates to the default plan limit' do
+        stub_ee_application_setting(elasticsearch_indexed_field_length_limit: 10)
+
+        issue = create :issue, description: 'The description is too long'
+
+        indexed_json = issue.__elasticsearch__.as_indexed_json
+
+        expect(indexed_json['description']).to eq('The descri')
+      end
+    end
+
+    context 'when the elasticsearch_indexed_field_length limit is 0' do
+      it 'does not truncate the fields' do
+        stub_ee_application_setting(elasticsearch_indexed_field_length_limit: 0)
+
+        issue = create :issue, description: 'The description is too long'
+
+        indexed_json = issue.__elasticsearch__.as_indexed_json
+
+        expect(indexed_json['description']).to eq('The description is too long')
+      end
+    end
+  end
+
   it_behaves_like 'no results when the user cannot read cross project' do
     let(:record1) { create(:issue, project: project, title: 'test-issue') }
     let(:record2) { create(:issue, project: project2, title: 'test-issue') }

locale/gitlab.pot

@@ -10106,6 +10106,9 @@ msgstr ""
 msgid "Identities"
 msgstr ""
 
+msgid "If any indexed field exceeds this limit it will be truncated to this number of characters and the rest will not be indexed or searchable. This does not apply to repository and wiki indexing. Setting this to 0 means it is unlimited."
+msgstr ""
+
 msgid "If any job surpasses this timeout threshold, it will be marked as failed. Human readable time input language is accepted like \"1 hour\". Values without specification represent seconds."
 msgstr ""
 
@@ -11713,6 +11716,9 @@ msgstr ""
 msgid "Maximum duration of a session."
 msgstr ""
 
+msgid "Maximum field length"
+msgstr ""
+
 msgid "Maximum job timeout"
 msgstr ""