diff --git a/.ci/.e2e-tests.yaml b/.ci/.e2e-tests.yaml
index dcce7b4553..aa350a4a49 100644
--- a/.ci/.e2e-tests.yaml
+++ b/.ci/.e2e-tests.yaml
@@ -15,12 +15,16 @@ SUITES:
       - "ubuntu-18.04"
     scenarios:
       - name: "Fleet"
+        pullRequestFilter: " && ~debian"
         tags: "fleet_mode_agent"
       - name: "Endpoint Integration"
+        pullRequestFilter: " && ~debian"
         tags: "agent_endpoint_integration"
       - name: "Stand-alone"
+        pullRequestFilter: " && ~ubi8"
         tags: "stand_alone_agent"
       - name: "Backend Processes"
+        pullRequestFilter: " && ~debian"
         tags: "backend_processes"
   - suite: "metricbeat"
     platforms:
@@ -45,6 +49,7 @@ SUITES:
       - name: "Metricbeat"
         tags: "metricbeat"
       - name: "MySQL"
+        pullRequestFilter: " && ~old_versions"
         tags: "integrations && mysql"
       - name: "Oracle"
         tags: "integrations && oracle"
diff --git a/.ci/Jenkinsfile b/.ci/Jenkinsfile
index 4a69b80969..15a7c8e2bd 100644
--- a/.ci/Jenkinsfile
+++ b/.ci/Jenkinsfile
@@ -250,12 +250,13 @@ def checkTestSuite(Map parallelTasks = [:], Map item = [:]) {
   def suite = item.suite
   def platforms = item.platforms
   item.scenarios.each { scenario ->
+    def pullRequestFilter = scenario.containsKey('pullRequestFilter') ? scenario.pullRequestFilter : ''
     def tags = scenario.tags
     def regexps = [ "^e2e/_suites/${suite}/.*", "^.ci/.*", "^cli/.*", "^e2e/.*\\.go" ]
     if ("${FORCE_SKIP_GIT_CHECKS}" == "true" || isGitRegionMatch(patterns: regexps, shouldMatchAll: false)) {
       platforms.each { platform ->
         log(level: 'INFO', text: "Adding ${platform}:${suite}:${tags} test suite to the build execution")
-        parallelTasks["${platform}_${suite}_${tags}"] = generateFunctionalTestStep(platform: "${platform}", suite: "${suite}", tags: "${tags}")
+        parallelTasks["${platform}_${suite}_${tags}"] = generateFunctionalTestStep(platform: "${platform}", suite: "${suite}", tags: "${tags}", pullRequestFilter: "${pullRequestFilter}")
       }
     } else {
       log(level: 'WARN', text: "The ${suite}:${tags} test suite won't be executed because there are no modified files")
@@ -307,6 +308,7 @@ def generateFunctionalTestStep(Map args = [:]){
   def platform = args.get('platform')
   def suite = args.get('suite')
   def tags = args.get('tags')
+  def pullRequestFilter = args.get('pullRequestFilter')?.trim() ? args.get('pullRequestFilter') : ''
 
   // We will decide whether to include the nightly tests in the execution at CI time, only.
   // On the other hand, the developers can use the TAGS environment variable locally.
@@ -317,6 +319,10 @@ def generateFunctionalTestStep(Map args = [:]){
   }
   tags += excludeNightlyTag
 
+  if (isPR() || isUpstreamTrigger(filter: 'PR-')) {
+    tags += pullRequestFilter
+  }
+
   return {
     node("${platform} && immutable && docker") {
       try {
diff --git a/e2e/_suites/fleet/features/agent_endpoint_integration.feature b/e2e/_suites/fleet/features/agent_endpoint_integration.feature
index 0f692c1ad8..d74de99922 100644
--- a/e2e/_suites/fleet/features/agent_endpoint_integration.feature
+++ b/e2e/_suites/fleet/features/agent_endpoint_integration.feature
@@ -3,36 +3,84 @@ Feature: Agent Endpoint Integration
   Scenarios for Agent to deploy Endpoint and sending data to Fleet and Elasticsearch.
 
 @deploy-endpoint-with-agent
-Scenario: Adding the Endpoint Integration to an Agent makes the host to show in Security App
-  Given a "centos" agent is deployed to Fleet with "tar" installer
+Scenario Outline: Adding the Endpoint Integration to an Agent makes the host to show in Security App
+  Given a "<os>" agent is deployed to Fleet with "tar" installer
     And the agent is listed in Fleet as "online"
   When the "Endpoint Security" integration is "added" in the policy
   Then the "Endpoint Security" datasource is shown in the policy as added
     And the host name is shown in the Administration view in the Security App as "online"
 
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
+
 @endpoint-policy-check
-Scenario: Deploying an Endpoint makes policies to appear in the Security App
-  When an Endpoint is successfully deployed with a "centos" Agent using "tar" installer
+Scenario Outline: Deploying an Endpoint makes policies to appear in the Security App
+  When an Endpoint is successfully deployed with a "<os>" Agent using "tar" installer
   Then the policy response will be shown in the Security App
 
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
+
 @set-policy-and-check-changes
-Scenario: Changing an Agent policy is reflected in the Security App
-  Given an Endpoint is successfully deployed with a "centos" Agent using "tar" installer
+Scenario Outline: Changing an Agent policy is reflected in the Security App
+  Given an Endpoint is successfully deployed with a "<os>" Agent using "tar" installer
   When the policy is updated to have "malware" in "detect" mode
   Then the policy will reflect the change in the Security App
 
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
+
 @deploy-endpoint-then-unenroll-agent
-Scenario: Un-enrolling Elastic Agent stops Elastic Endpoint
-  Given an Endpoint is successfully deployed with a "centos" Agent using "tar" installer
+Scenario Outline: Un-enrolling Elastic Agent stops Elastic Endpoint
+  Given an Endpoint is successfully deployed with a "<os>" Agent using "tar" installer
   When the agent is un-enrolled
   Then the agent is listed in Fleet as "inactive"
     And the host name is not shown in the Administration view in the Security App
-    And the "elastic-endpoint" process is in the "stopped" state on the host
+
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
 
 @deploy-endpoint-then-remove-it-from-policy
-Scenario: Removing Endpoint from Agent policy stops the connected Endpoint
-  Given an Endpoint is successfully deployed with a "centos" Agent using "tar" installer
+Scenario Outline: Removing Endpoint from Agent policy stops the connected Endpoint
+  Given an Endpoint is successfully deployed with a "<os>" Agent using "tar" installer
   When the "Endpoint Security" integration is "removed" in the policy
   Then the agent is listed in Fleet as "online"
     But the host name is not shown in the Administration view in the Security App
-    And the "elastic-endpoint" process is in the "stopped" state on the host
+
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
diff --git a/e2e/_suites/fleet/features/backend_processes.feature b/e2e/_suites/fleet/features/backend_processes.feature
index 88ae79bf9b..5e61d0e9f3 100644
--- a/e2e/_suites/fleet/features/backend_processes.feature
+++ b/e2e/_suites/fleet/features/backend_processes.feature
@@ -8,9 +8,15 @@ Scenario Outline: Deploying the <os> agent
   When the "elastic-agent" process is in the "started" state on the host
   Then the "filebeat" process is in the "started" state on the host
     And the "metricbeat" process is in the "started" state on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @enroll
@@ -19,9 +25,15 @@ Scenario Outline: Deploying the <os> agent with enroll and then run on rpm and d
   When the "elastic-agent" process is in the "started" state on the host
   Then the "filebeat" process is in the "started" state on the host
     And the "metricbeat" process is in the "started" state on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @stop-agent
@@ -30,9 +42,15 @@ Scenario Outline: Stopping the <os> agent stops backend processes
   When the "elastic-agent" process is "stopped" on the host
   Then the "filebeat" process is in the "stopped" state on the host
     And the "metricbeat" process is in the "stopped" state on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @restart-agent
@@ -41,9 +59,15 @@ Scenario Outline: Restarting the installed <os> agent
   When the "elastic-agent" process is "restarted" on the host
   Then the "filebeat" process is in the "started" state on the host
     And the "metricbeat" process is in the "started" state on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @restart-host
@@ -53,9 +77,15 @@ Scenario Outline: Restarting the <os> host with persistent agent restarts backen
   Then the "elastic-agent" process is in the "started" state on the host
     And the "filebeat" process is in the "started" state on the host
     And the "metricbeat" process is in the "started" state on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @unenroll
@@ -65,9 +95,15 @@ Scenario Outline: Un-enrolling the <os> agent stops backend processes
   Then the "elastic-agent" process is in the "started" state on the host
     And the "filebeat" process is in the "stopped" state on the host
     And the "metricbeat" process is in the "stopped" state on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @reenroll
@@ -77,9 +113,15 @@ Scenario Outline: Re-enrolling the <os> agent starts the elastic-agent process
     And the "elastic-agent" process is "stopped" on the host
   When the agent is re-enrolled on the host
   Then the "elastic-agent" process is "started" on the host
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @uninstall-host
@@ -89,7 +131,45 @@ Scenario Outline: Un-installing the installed <os> agent
   Then the "elastic-agent" process is in the "stopped" state on the host
     And the "filebeat" process is in the "stopped" state on the host
     And the "metricbeat" process is in the "stopped" state on the host
-Examples:
+
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
+
+@deploy-endpoint-then-unenroll-agent
+Scenario Outline: Un-enrolling Elastic Agent stops Elastic Endpoint
+  Given an Endpoint is successfully deployed with a "<os>" Agent using "tar" installer
+  When the agent is un-enrolled
+  Then the "elastic-endpoint" process is in the "stopped" state on the host
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
+| debian |
+
+@deploy-endpoint-then-remove-it-from-policy
+Scenario Outline: Removing Endpoint from Agent policy stops the connected Endpoint
+  Given an Endpoint is successfully deployed with a "<os>" Agent using "tar" installer
+  When the "Endpoint Security" integration is "removed" in the policy
+  Then the "elastic-endpoint" process is in the "stopped" state on the host
+
+@centos
+Examples: Centos
+| os     |
+| centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
diff --git a/e2e/_suites/fleet/features/fleet_mode_agent.feature b/e2e/_suites/fleet/features/fleet_mode_agent.feature
index 26d77eeddd..31eab5853e 100644
--- a/e2e/_suites/fleet/features/fleet_mode_agent.feature
+++ b/e2e/_suites/fleet/features/fleet_mode_agent.feature
@@ -8,9 +8,15 @@ Scenario Outline: Deploying the <os> agent
   When the "elastic-agent" process is in the "started" state on the host
   Then the agent is listed in Fleet as "online"
     And system package dashboards are listed in Fleet
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @enroll
@@ -19,9 +25,15 @@ Scenario Outline: Deploying the <os> agent with enroll and then run on rpm and d
   When the "elastic-agent" process is in the "started" state on the host
   Then the agent is listed in Fleet as "online"
     And system package dashboards are listed in Fleet
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 # @upgrade-agent
@@ -32,18 +44,26 @@ Scenario Outline: Upgrading the installed <os> agent
     And the "elastic-agent" process is "restarted" on the host
   When agent is upgraded to version "latest"
   Then agent is in version "latest"
-Examples:
-| os     | 
-| debian | 
+
+@debian
+Examples: Debian
+| os     |
+| debian |
 
 @restart-agent
 Scenario Outline: Restarting the installed <os> agent
   Given a "<os>" agent is deployed to Fleet with "tar" installer
   When the "elastic-agent" process is "restarted" on the host
   Then the agent is listed in Fleet as "online"
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @unenroll
@@ -51,9 +71,15 @@ Scenario Outline: Un-enrolling the <os> agent deactivates the agent
   Given a "<os>" agent is deployed to Fleet with "tar" installer
   When the agent is un-enrolled
   Then the agent is listed in Fleet as "inactive"
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @reenroll
@@ -64,9 +90,15 @@ Scenario Outline: Re-enrolling the <os> agent activates the agent in Fleet
     And the agent is re-enrolled on the host
   When the "elastic-agent" process is "started" on the host
   Then the agent is listed in Fleet as "online"
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @revoke-token
@@ -74,9 +106,15 @@ Scenario Outline: Revoking the enrollment token for the <os> agent
   Given a "<os>" agent is deployed to Fleet with "tar" installer
   When the enrollment token is revoked
   Then an attempt to enroll a new agent fails
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
 
 @uninstall-host
@@ -85,7 +123,13 @@ Scenario Outline: Un-installing the installed <os> agent
   When the "elastic-agent" process is "uninstalled" on the host
   Then the file system Agent folder is empty
     And the agent is listed in Fleet as "offline"
-Examples:
+
+@centos
+Examples: Centos
 | os     |
 | centos |
+
+@debian
+Examples: Debian
+| os     |
 | debian |
diff --git a/e2e/_suites/fleet/features/stand_alone_agent.feature b/e2e/_suites/fleet/features/stand_alone_agent.feature
index bdc78beeec..5842cd57cb 100644
--- a/e2e/_suites/fleet/features/stand_alone_agent.feature
+++ b/e2e/_suites/fleet/features/stand_alone_agent.feature
@@ -9,18 +9,30 @@ Scenario Outline: Starting the <image> agent starts backend processes
   When a "<image>" stand-alone agent is deployed
   Then the "filebeat" process is in the "started" state on the host
     And the "metricbeat" process is in the "started" state on the host
-Examples:
+
+@default
+Examples: default
 | image   |
 | default |
+
+@ubi8
+Examples: Ubi8
+| image   |
 | ubi8    |
 
 @deploy-stand-alone
 Scenario Outline: Deploying a <image> stand-alone agent
   When a "<image>" stand-alone agent is deployed
   Then there is new data in the index from agent
-Examples:
+
+@default
+Examples: default
 | image   |
 | default |
+
+@ubi8
+Examples: Ubi8
+| image   |
 | ubi8    |
 
 @stop-agent
@@ -28,7 +40,13 @@ Scenario Outline: Stopping the <image> agent container stops data going into ES
   Given a "<image>" stand-alone agent is deployed
   When the "elastic-agent" docker container is stopped
   Then there is no new data in the index after agent shuts down
-Examples:
+
+@default
+Examples: default
 | image   |
 | default |
+
+@ubi8
+Examples: Ubi8
+| image   |
 | ubi8    |
diff --git a/e2e/_suites/metricbeat/features/integrations.feature b/e2e/_suites/metricbeat/features/integrations.feature
index ea6afe2c09..dbde408369 100644
--- a/e2e/_suites/metricbeat/features/integrations.feature
+++ b/e2e/_suites/metricbeat/features/integrations.feature
@@ -92,20 +92,24 @@ Examples: vSphere
 | integration | version |
 | vsphere     | latest  |
 
-@variants
+@mysql
 Scenario Outline: <integration>-<variant>-<version> sends metrics to Elasticsearch without errors
   Given "<variant>" v<version>, variant of "<integration>", is running for metricbeat
   When metricbeat is installed and configured for "<variant>", variant of the "<integration>" module
   Then there are "<variant>" events in the index
     And there are no errors in the index
 
-@mysql
-Examples: MySQL
+@latest_versions
+Examples: MySQL (latest versions)
 | integration | variant | version  |
-| mysql       | MariaDB | 10.2.23  |
-| mysql       | MariaDB | 10.3.14  |
 | mysql       | MariaDB | 10.4.4   |
-| mysql       | MySQL   | 5.7.12   |
 | mysql       | MySQL   | 8.0.13   |
+| mysql       | MySQL   | 5.7.12   |
+
+@old_versions
+Examples: MySQL (old versions)
+| integration | variant | version  |
+| mysql       | MariaDB | 10.2.23  |
+| mysql       | MariaDB | 10.3.14  |
 | mysql       | Percona | 5.7.24   |
 | mysql       | Percona | 8.0.13-4 |