clear stale docker network endpoints before starting containers

docker_container.running fails with an address-already-in-use error when a
container holds a stale sobridge endpoint - either after its IP changes in
pillar or after an unclean shutdown leaves the endpoint behind.

Add the clear_stale_endpoint macro and call it from every enabled.sls whose
container attaches to sobridge with a static IP. The macro force-disconnects
the container from the network only when its current endpoint IP does not
match the one salt is about to assign, so it is a no-op on a normal highstate.

The strelka states use state IDs that differ from their container names, so
those calls pass state_id explicitly. nginx resolves its IP through
container_config, which varies by role.

salt/docker/macros/docker_endpoint.jinja

@@ -0,0 +1,12 @@
+{% macro clear_stale_endpoint(container, network, ipv4, state_id=None) %}
+{{ container }}_{{ network }}_stale_endpoint:
+  cmd.run:
+    - name: docker network disconnect -f {{ network }} {{ container }} || true
+    - onlyif: docker inspect {{ container }}
+    - unless: >-
+        docker inspect -f
+        '{{ '{{' }} with index .NetworkSettings.Networks "{{ network }}" {{ '}}' }}{{ '{{' }} .IPAMConfig.IPv4Address {{ '}}' }}{{ '{{' }} end {{ '}}' }}'
+        {{ container }} 2>/dev/null | grep -qx '{{ ipv4 }}'
+    - require_in:
+      - docker_container: {{ state_id or container }}
+{% endmacro %}

salt/elastalert/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-elastalert', 'sobridge', DOCKERMERGED.containers['so-elastalert'].ip) }}
 
 include:
   - elastalert.config

salt/elastic-fleet-package-registry/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-elastic-fleet-package-registry', 'sobridge', DOCKERMERGED.containers['so-elastic-fleet-package-registry'].ip) }}
 
 include:
   - elastic-fleet-package-registry.config

salt/elasticagent/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-elastic-agent', 'sobridge', DOCKERMERGED.containers['so-elastic-agent'].ip) }}
 
 include:
   - ca

salt/elasticfleet/enabled.sls

@@ -8,6 +8,7 @@
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'elasticfleet/map.jinja' import ELASTICFLEETMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
 
 {#   This value is generated during node install and stored in minion pillar #}
 {%   set SERVICETOKEN = salt['pillar.get']('elasticfleet:config:server:es_token','') %}
@@ -43,6 +44,8 @@ elasticagent_syncartifacts:
 {% endif %}
 
 {%   if SERVICETOKEN != '' %}
+{{ clear_stale_endpoint('so-elastic-fleet', 'sobridge', DOCKERMERGED.containers['so-elastic-fleet'].ip) }}
+
 so-elastic-fleet:
   docker_container.running:
     - image: {{ GLOBALS.registry_host }}:5000/{{ GLOBALS.image_repo }}/so-elastic-agent:{{ GLOBALS.so_version }}

salt/elasticfleet/tools/sbin_jinja/so-elastic-fleet-setup

@@ -257,6 +257,7 @@ for state_attempt in {1..3}; do
     fi
 done
 
+printf "\nRunning so-elastic-agent-gen-installers\n"
 # Generate installers & install Elastic Agent on the node
 for agent_gen_attempt in {1..3}; do
     if so-elastic-agent-gen-installers; then
@@ -274,4 +275,6 @@ printf "\nApplying elasticfleet.install_agent_grid state\n"
 if ! salt-call state.apply elasticfleet.install_agent_grid queue=True; then
     printf "\nFailure(s) in elasticfleet.install_agent_grid state... Exiting...\n"
     exit 1
-fi
+fi
+
+printf "\nElastic Fleet setup completed successfully\n"

salt/elasticsearch/enabled.sls

@@ -10,6 +10,9 @@
 {%   from 'elasticsearch/config.map.jinja' import ELASTICSEARCH_NODES %}
 {%   from 'elasticsearch/config.map.jinja' import ELASTICSEARCH_SEED_HOSTS %}
 {%   from 'elasticsearch/config.map.jinja' import ELASTICSEARCHMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-elasticsearch', 'sobridge', DOCKERMERGED.containers['so-elasticsearch'].ip) }}
 
 include:
   - ca

salt/hydra/enabled.sls

@@ -14,6 +14,9 @@
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   if 'api' in salt['pillar.get']('features', []) %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-hydra', 'sobridge', DOCKERMERGED.containers['so-hydra'].ip) }}
 
 include:
   - hydra.config

salt/influxdb/enabled.sls

@@ -9,6 +9,9 @@
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   set PASSWORD = salt['pillar.get']('secrets:influx_pass') %}
 {%   set TOKEN = salt['pillar.get']('influxdb:token') %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-influxdb', 'sobridge', DOCKERMERGED.containers['so-influxdb'].ip) }}
 
 include:
   - influxdb.ssl

salt/kafka/enabled.sls

@@ -16,6 +16,9 @@
 {%   set KAFKANODES = salt['pillar.get']('kafka:nodes') %}
 {%   set KAFKA_EXTERNAL_ACCESS = salt['pillar.get']('kafka:config:external_access:enabled', default=False) %}
 {%   if 'gmd' in salt['pillar.get']('features', []) %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-kafka', 'sobridge', DOCKERMERGED.containers['so-kafka'].ip) }}
 
 include:
   - kafka.ca

salt/kibana/enabled.sls

@@ -8,6 +8,9 @@
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'elasticsearch/config.map.jinja' import ELASTICSEARCHMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-kibana', 'sobridge', DOCKERMERGED.containers['so-kibana'].ip) }}
 
 include:
   - kibana.config

salt/kratos/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-kratos', 'sobridge', DOCKERMERGED.containers['so-kratos'].ip) }}
 
 include:
   - kratos.config

salt/logstash/enabled.sls

@@ -10,6 +10,9 @@
 {%   from 'logstash/map.jinja' import LOGSTASH_MERGED %}
 {%   from 'logstash/map.jinja' import LOGSTASH_NODES %}
 {%   set lsheap = LOGSTASH_MERGED.settings.lsheap %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-logstash', 'sobridge', DOCKERMERGED.containers['so-logstash'].ip) }}
 
 include:
   - ca

salt/nginx/enabled.sls

@@ -8,6 +8,7 @@
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'nginx/map.jinja' import NGINXMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
 
 include:
   - nginx.ssl
@@ -31,6 +32,8 @@ make-rule-dir-nginx:
 {%     set container_config = 'so-nginx-fleet-node' %}
 {%   endif %}
 
+{{ clear_stale_endpoint('so-nginx', 'sobridge', DOCKERMERGED.containers[container_config].ip) }}
+
 so-nginx:
   docker_container.running:
     - image: {{ GLOBALS.registry_host }}:5000/{{ GLOBALS.image_repo }}/so-nginx:{{ GLOBALS.so_version }}

salt/postgres/enabled.sls

@@ -8,6 +8,9 @@
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   set SO_POSTGRES_USER = salt['pillar.get']('postgres:auth:users:so_postgres_user:user', 'so_postgres') %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-postgres', 'sobridge', DOCKERMERGED.containers['so-postgres'].ip) }}
 
 include:
   - postgres.auth

salt/redis/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-redis', 'sobridge', DOCKERMERGED.containers['so-redis'].ip) }}
 
 include:
   - ca

salt/registry/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-dockerregistry', 'sobridge', DOCKERMERGED.containers['so-dockerregistry'].ip) }}
 
 include:
   - registry.ssl

salt/soc/defaults.map.jinja

@@ -8,6 +8,7 @@
 {% from 'docker/docker.map.jinja' import DOCKERMERGED -%}
 {% set INFLUXDB_TOKEN = salt['pillar.get']('influxdb:token') %}
 {% import_text 'influxdb/metrics_link.txt' as METRICS_LINK %}
+{% from 'telegraf/map.jinja' import TELEGRAFMERGED %}
 
 {% for module, application_url in GLOBALS.application_urls.items() %}
 {%   do SOCDEFAULTS.soc.config.server.modules[module].update({'hostUrl': application_url}) %}
@@ -24,13 +25,22 @@
 
 {% do SOCDEFAULTS.soc.config.server.modules.elastic.update({'username': GLOBALS.elasticsearch.auth.users.so_elastic_user.user, 'password': GLOBALS.elasticsearch.auth.users.so_elastic_user.pass}) %}
 
-{% do SOCDEFAULTS.soc.config.server.modules.influxdb.update({'hostUrl': 'https://' ~ GLOBALS.influxdb_host ~ ':8086'}) %}
+{% if TELEGRAFMERGED.output == 'POSTGRES' %}
-{% do SOCDEFAULTS.soc.config.server.modules.influxdb.update({'token': INFLUXDB_TOKEN}) %}
+{%   for tool in SOCDEFAULTS.soc.config.server.client.tools %}
-{% for tool in SOCDEFAULTS.soc.config.server.client.tools %}
+{%     if tool.name == "toolInfluxDb" %}
-{%   if tool.name == "toolInfluxDb" and METRICS_LINK | length > 0 %}
+{%       do SOCDEFAULTS.soc.config.server.client.tools.remove(tool) %}
-{%     do tool.update({'link': METRICS_LINK}) %}
+{%     endif %}
-{%   endif %}
+{%   endfor %}
-{% endfor %}
+
+{% else %}
+{%   do SOCDEFAULTS.soc.config.server.modules.influxdb.update({'hostUrl': 'https://' ~ GLOBALS.influxdb_host ~ ':8086'}) %}
+{%   do SOCDEFAULTS.soc.config.server.modules.influxdb.update({'token': INFLUXDB_TOKEN}) %}
+{%   for tool in SOCDEFAULTS.soc.config.server.client.tools %}
+{%     if tool.name == "toolInfluxDb" and METRICS_LINK | length > 0 %}
+{%       do tool.update({'link': METRICS_LINK}) %}
+{%     endif %}
+{%   endfor %}
+{% endif %}
 
 {% do SOCDEFAULTS.soc.config.server.modules.statickeyauth.update({'anonymousCidr': DOCKERMERGED.range, 'apiKey': pillar.sensoroni.config.sensoronikey}) %}
 

salt/soc/enabled.sls

@@ -10,6 +10,9 @@
 {%   from 'soc/merged.map.jinja' import DOCKER_EXTRA_HOSTS %}
 {%   from 'soc/merged.map.jinja' import SOCMERGED %}
 
+{% from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+{{ clear_stale_endpoint('so-soc', 'sobridge', DOCKERMERGED.containers['so-soc'].ip) }}
+
 include:
   - ca
   - soc.config

salt/soc/merged.map.jinja

@@ -7,6 +7,11 @@
 {% from 'soc/defaults.map.jinja' import SOCDEFAULTS with context %}
 {% from 'elasticsearch/config.map.jinja' import ELASTICSEARCH_NODES %}
 {% from 'manager/map.jinja' import MANAGERMERGED %}
+{% from 'telegraf/map.jinja' import TELEGRAFMERGED %}
+{%- set PG_ENTRY = salt['pillar.get']('telegraf:postgres_creds:' ~ grains.id, {}) %}
+{%- set PG_USER = PG_ENTRY.get('user', '') %}
+{%- set PG_PASS = PG_ENTRY.get('pass', '') %}
+
 {% set DOCKER_EXTRA_HOSTS = ELASTICSEARCH_NODES %}
 {% do DOCKER_EXTRA_HOSTS.append({GLOBALS.influxdb_host:pillar.node_data[GLOBALS.influxdb_host].ip}) %}
 
@@ -75,6 +80,20 @@
 {%   do SOCMERGED.config.server.update({'airgapEnabled': false}) %}
 {% endif %}
 
+{# Define the postgresmetrics module if telegraf is setup to only use Postgres #}
+{% if TELEGRAFMERGED.output != 'INFLUXDB' and PG_USER and PG_PASS %}
+{%   do SOCMERGED.config.server.modules.update({
+        'postgresmetrics': {
+          'database': 'so_telegraf',
+          'host': GLOBALS.manager_ip,
+          'password': PG_PASS,
+          'port': 5432,
+          'sslMode': 'allow',
+          'user': PG_USER,
+        }
+      }) %}
+{%   do SOCMERGED.config.server.modules.pop('influxdb') %}
+{% endif %}
 
 {# Define the Detections custom ruleset that should always be present #}
 {% set CUSTOM_RULESET = {

salt/strelka/backend/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-strelka-backend', 'sobridge', DOCKERMERGED.containers['so-strelka-backend'].ip, state_id='strelka_backend') }}
 
 include:
   - strelka.backend.config

salt/strelka/coordinator/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-strelka-coordinator', 'sobridge', DOCKERMERGED.containers['so-strelka-coordinator'].ip, state_id='strelka_coordinator') }}
 
 include:
   - strelka.coordinator.config

salt/strelka/filestream/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-strelka-filestream', 'sobridge', DOCKERMERGED.containers['so-strelka-filestream'].ip, state_id='strelka_filestream') }}
 
 include:
   - strelka.filestream.config

salt/strelka/frontend/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-strelka-frontend', 'sobridge', DOCKERMERGED.containers['so-strelka-frontend'].ip, state_id='strelka_frontend') }}
 
 include:
   - strelka.frontend.config

salt/strelka/gatekeeper/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-strelka-gatekeeper', 'sobridge', DOCKERMERGED.containers['so-strelka-gatekeeper'].ip, state_id='strelka_gatekeeper') }}
 
 include:
   - strelka.gatekeeper.config

salt/strelka/manager/enabled.sls

@@ -7,6 +7,9 @@
 {% if sls.split('.')[0] in allowed_states %}
 {%   from 'docker/docker.map.jinja' import DOCKERMERGED %}
 {%   from 'vars/globals.map.jinja' import GLOBALS %}
+{%   from 'docker/macros/docker_endpoint.jinja' import clear_stale_endpoint %}
+
+{{ clear_stale_endpoint('so-strelka-manager', 'sobridge', DOCKERMERGED.containers['so-strelka-manager'].ip, state_id='strelka_manager') }}
 
 include:
   - strelka.manager.config

salt/telegraf/soc_telegraf.yaml

@@ -5,7 +5,7 @@ telegraf:
     advanced: True
     helpLink: influxdb
   output:
-    description: Selects the backend(s) Telegraf writes metrics to. INFLUXDB keeps the current behavior; POSTGRES writes to the grid's Postgres instance; BOTH dual-writes for migration validation.
+    description: Selects the backend(s) Telegraf writes metrics to. INFLUXDB keeps the current behavior; POSTGRES writes to the grid's Postgres instance; BOTH dual-writes for migration validation. When set to BOTH, the grid screen's metrics are pulled from Postgres, and the InfluxDB tool link remains visible. When set to POSTGRES, the InfluxDB tool link is removed.
     options:
       - INFLUXDB
       - POSTGRES

setup/so-setup

@@ -793,7 +793,7 @@ if ! [[ -f $install_opt_file ]]; then
 		logCmd "so-soc-restart"
 		title "Setting up Elastic Fleet"
 		logCmd "salt-call state.apply elasticfleet.config"	
-		if ! logCmd so-elastic-fleet-setup; then
+		if ! so-elastic-fleet-setup; then
 			fail_setup "Failed to run so-elastic-fleet-setup"
 		fi
 		mark_setup_complete