@ -11,7 +11,7 @@
# Notes: Reboots are handled directly on each host by apt_weekly.sh.
# Notes: Reboot staggering: docker_14 first, docker_69 second, docker_10 third.
# Notes: Container monitoring is dynamic with binary_sensor status preferred over switch state.
# Notes: Includes Portainer stack status repairs and scheduled image prune.
# Notes: Includes Portainer stack status repairs , 20-minute Joanna dispatch for persistent container outages, and scheduled image prune.
######################################################################
input_datetime:
@ -257,7 +257,7 @@ template:
icon : mdi:docker
state : >-
{% set ns = namespace(keys=[], down=[]) %}
{% set monitored = state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) %}
{% set monitored = state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) | list %}
{% set telemetry_degraded = is_state('binary_sensor.docker_container_telemetry_degraded', 'on') %}
{% for switch_entity in monitored %}
{% set key = switch_entity | replace('switch.', '') | regex_replace('_container(?:_2)?$', '') %}
@ -297,19 +297,58 @@ template:
{% set ns.down = ns.down + [key] %}
{% endif %}
{% endfor %}
{{ ns.down | sort | join(', ') if (ns.down | count > 0) else 'none' }}
{{ 'ok' if (ns.down | count == 0) else 'down' }}
attributes:
down_containers : >-
{% set ns = namespace(keys=[], down=[]) %}
{% set monitored = state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) | list %}
{% set telemetry_degraded = is_state('binary_sensor.docker_container_telemetry_degraded', 'on') %}
{% for switch_entity in monitored %}
{% set key = switch_entity | replace('switch.', '') | regex_replace('_container(?:_2)?$', '') %}
{% if key not in ns.keys %}
{% set ns.keys = ns.keys + [key] %}
{% endif %}
{% endfor %}
{% for key in ns.keys | sort %}
{% set status_entity = 'binary_sensor.' ~ key ~ '_status' %}
{% set status_entity_alt = status_entity ~ '_2' %}
{% set state_entity = 'sensor.' ~ key ~ '_state' %}
{% set state_entity_alt = state_entity ~ '_2' %}
{% set switch_entity = 'switch.' ~ key ~ '_container' %}
{% set switch_entity_alt = switch_entity ~ '_2' %}
{% set resolver = namespace(state='unknown', chosen=false) %}
{% for candidate in [status_entity, status_entity_alt, state_entity, state_entity_alt, switch_entity, switch_entity_alt] %}
{% if not resolver.chosen and expand(candidate) | count > 0 %}
{% set candidate_state = states(candidate) | lower %}
{% if candidate_state not in ['unknown', 'unavailable', ''] %}
{% set resolver.state = candidate_state %}
{% set resolver.chosen = true %}
{% endif %}
{% endif %}
{% endfor %}
{% if not resolver.chosen %}
{% for candidate in [status_entity, status_entity_alt, state_entity, state_entity_alt, switch_entity, switch_entity_alt] %}
{% if not resolver.chosen and expand(candidate) | count > 0 %}
{% set resolver.state = states(candidate) | lower %}
{% set resolver.chosen = true %}
{% endif %}
{% endfor %}
{% endif %}
{% set effective_state = resolver.state %}
{% if effective_state in ['off', 'stopped'] %}
{% set ns.down = ns.down + [key] %}
{% elif not telemetry_degraded and effective_state in ['unknown', 'unavailable'] %}
{% set ns.down = ns.down + [key] %}
{% endif %}
{% endfor %}
{{ ns.down | sort }}
- name : "Docker Containers Down Count"
unique_id : docker_containers_down_count
icon : mdi:counter
state : >-
{% set down_list = states('sensor.docker_containers_down_list') %}
{% set normalized = down_list | lower %}
{% if normalized in ['unknown', 'unavailable', 'none', ''] %}
0
{% else %}
{{ down_list.split(',') | map('trim') | reject('equalto', '') | list | count }}
{% endif %}
{% set down_items = state_attr('sensor.docker_containers_down_list', 'down_containers') | default([], true) | list %}
{{ down_items | count }}
- name : "Docker Stacks Down List"
unique_id : docker_stacks_down_list
@ -475,6 +514,60 @@ script:
data:
topic : "DOCKER"
message : "{{ container_name }} is {{ effective_state }} for over 5 minutes."
- delay:
minutes : 15
- variables:
persistent_effective_state : >-
{% set candidates = [status_entity, status_entity_alt, state_entity, state_entity_alt, switch_entity, switch_entity_alt] %}
{% set resolver = namespace(state='unknown', chosen=false) %}
{% for candidate in candidates %}
{% if not resolver.chosen and expand(candidate) | count > 0 %}
{% set candidate_state = states(candidate) | lower %}
{% if candidate_state not in ['unknown', 'unavailable', ''] %}
{% set resolver.state = candidate_state %}
{% set resolver.chosen = true %}
{% endif %}
{% endif %}
{% endfor %}
{% if not resolver.chosen %}
{% for candidate in candidates %}
{% if not resolver.chosen and expand(candidate) | count > 0 %}
{% set resolver.state = states(candidate) | lower %}
{% set resolver.chosen = true %}
{% endif %}
{% endfor %}
{% endif %}
{{ resolver.state }}
container_name : "{{ state_attr(effective_entity, 'friendly_name') | default(container_key, true) }}"
- condition : template
value_template : >-
{{ persistent_effective_state in down_states and
not (is_state('binary_sensor.docker_container_telemetry_degraded', 'on') and
persistent_effective_state in ['unknown', 'unavailable']) }}
- condition : state
entity_id : binary_sensor.docker_container_alerts_snoozed
state : "off"
- service : script.joanna_dispatch
data:
trigger_context : >-
HA automation docker_state_sync_repairs_dynamic
(Docker State Sync - Repairs (Dynamic))
source : "home_assistant_automation.docker_state_sync_repairs_dynamic"
summary : "{{ container_name }} container has remained {{ persistent_effective_state }} for 20 minutes"
entity_ids:
- "{{ effective_entity }}"
- "{{ switch_entity }}"
diagnostics : >-
issue_id={{ issue_id }},
spook_issue_id={{ spook_issue_id }},
container_key={{ container_key }},
effective_entity={{ effective_entity }},
switch_entity={{ switch_entity }},
effective_state_initial={{ effective_state }},
effective_state_20m={{ persistent_effective_state }}
request : >-
Troubleshoot and resolve the persistent Docker container outage if possible.
Use Duplicati and the related host/container telemetry to verify recovery.
- conditions : "{{ op == 'clear' }}"
sequence:
- variables:
@ -695,37 +788,54 @@ automation:
- alias : "Docker State Sync - Repairs (Dynamic)"
id : docker_state_sync_repairs_dynamic
description : "Detect Docker container/stack state transitions and delegate Repairs sync."
mode : parallel
mode : queued
max : 50
max_exceeded : silent
trigger:
- platform : event
event_type : state_changed
variables:
entity_id : "{{ trigger.event.data.entity_id | default('') }}"
old_state : "{{ (trigger.event.data.old_state.state if trigger.event.data.old_state is not none else '') | lower }}"
new_state : "{{ (trigger.event.data.new_state.state if trigger.event.data.new_state is not none else '') | lower }}"
monitored_switches : "{{ state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) | list }}"
is_monitored_container_event : >-
{% set ent = entity_id %}
{% if ent.startswith('switch.') and (ent.endswith('_container') or ent.endswith('_container_2')) %}
{{ ent in monitored_switches }}
{% elif ent.startswith('binary_sensor.') and (ent.endswith('_status') or ent.endswith('_status_2')) %}
{% set key = ent | replace('binary_sensor.', '') | regex_replace('_status(?:_2)?$', '') %}
{{ ('switch.' ~ key ~ '_container') in monitored_switches or ('switch.' ~ key ~ '_container_2') in monitored_switches }}
{% elif ent.startswith('sensor.') and (ent.endswith('_state') or ent.endswith('_state_2')) %}
{% set key = ent | replace('sensor.', '') | regex_replace('_state(?:_2)?$', '') %}
{{ ('switch.' ~ key ~ '_container') in monitored_switches or ('switch.' ~ key ~ '_container_2') in monitored_switches }}
{% else %}
false
{% endif %}
is_monitored_stack_event : >-
{% set ent = entity_id %}
{% if ent.startswith('binary_sensor.') and ent.endswith('_stack_status') %}
{% set stack_key = ent | replace('binary_sensor.', '') | regex_replace('_stack_status$', '') %}
{{ expand('sensor.' ~ stack_key ~ '_stack_containers_count') | count > 0 }}
{% else %}
false
{% endif %}
condition:
- condition : template
value_template : "{{ trigger.event.data.old_state is not none and trigger.event.data.new_state is not none }}"
- condition : template
value_template : "{{ trigger.event.data.old_state.state != trigger.event.data.new_state.state }}"
value_template : "{{ old_state != new_state }}"
- condition : template
value_template : "{{ is_monitored_container_event or is_monitored_stack_event }}"
action:
- variables:
entity_id : "{{ trigger.event.data.entity_id | default('') }}"
old_state : "{{ trigger.event.data.old_state.state | lower }}"
new_state : "{{ trigger.event.data.new_state.state | lower }}"
monitored : "{{ state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) }}"
- delay:
seconds : 2
- condition : template
value_template : "{{ states(entity_id) | lower == new_state }}"
- choose:
- conditions:
- condition : template
value_template : >-
{% set ent = entity_id %}
{% if ent.startswith('switch.') and (ent.endswith('_container') or ent.endswith('_container_2')) %}
{{ ent in monitored }}
{% elif ent.startswith('binary_sensor.') and (ent.endswith('_status') or ent.endswith('_status_2')) %}
{% set key = ent | replace('binary_sensor.', '') | regex_replace('_status(?:_2)?$', '') %}
{{ ('switch.' ~ key ~ '_container') in monitored or ('switch.' ~ key ~ '_container_2') in monitored }}
{% elif ent.startswith('sensor.') and (ent.endswith('_state') or ent.endswith('_state_2')) %}
{% set key = ent | replace('sensor.', '') | regex_replace('_state(?:_2)?$', '') %}
{{ ('switch.' ~ key ~ '_container') in monitored or ('switch.' ~ key ~ '_container_2') in monitored }}
{% else %}
false
{% endif %}
value_template : "{{ is_monitored_container_event }}"
sequence:
- variables:
down_states : [ 'off' , 'stopped' , 'exited' , 'dead' , 'unknown' , 'unavailable' ]
@ -748,14 +858,7 @@ automation:
operation : "clear"
- conditions:
- condition : template
value_template : >-
{% set ent = entity_id %}
{% if ent.startswith('binary_sensor.') and ent.endswith('_stack_status') %}
{% set stack_key = ent | replace('binary_sensor.', '') | regex_replace('_stack_status$', '') %}
{{ expand('sensor.' ~ stack_key ~ '_stack_containers_count') | count > 0 }}
{% else %}
false
{% endif %}
value_template : "{{ is_monitored_stack_event }}"
sequence:
- variables:
down_states : [ 'off' , 'unknown' , 'unavailable' ]
@ -785,7 +888,7 @@ automation:
minutes : "/55"
action:
- variables:
monitored_switches : "{{ state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) }}"
monitored_switches : "{{ state_attr('group.docker_monitored_containers', 'entity_id') | default([], true) | list }}"
- repeat:
for_each : "{{ monitored_switches }}"
sequence:
@ -805,7 +908,7 @@ automation:
{% endif %}
{% endif %}
{% endfor %}
{{ ns.items }}
{{ ns.items | list }}
- repeat:
for_each : "{{ stack_status_entities }}"
sequence:
@ -832,7 +935,9 @@ automation:
data:
title : "Docker Maintenance Check"
value1 : "{{ states('sensor.docker_containers_down_count') }} containers are currently down."
value2 : "Down: {{ states('sensor.docker_containers_down_list') }}"
value2 : >-
{% set down_items = state_attr('sensor.docker_containers_down_list', 'down_containers') | default([], true) | list %}
Down : {{ down_items | join(', ') if (down_items | count > 0) else 'none' }}
who : "carlo"
group : "maintenance"
title1 : "Yes, snooze 1h"
@ -846,7 +951,8 @@ automation:
topic : "DOCKER"
message : >-
Maintenance prompt sent to Carlo ({{ states('sensor.docker_containers_down_count') }} down:
{{ states('sensor.docker_containers_down_list') }}).
{% set down_items = state_attr('sensor.docker_containers_down_list', 'down_containers') | default([], true) | list %}
{{ down_items | join(', ') if (down_items | count > 0) else 'none' }}).
- alias : "Docker Maintenance Snooze 1H"
id : docker_maintenance_snooze_1h
@ -885,7 +991,9 @@ automation:
topic : "DOCKER"
message : >-
Maintenance snooze declined with {{ states('sensor.docker_containers_down_count') }}
containers down ({{ states('sensor.docker_containers_down_list') }}).
containers down (
{% set down_items = state_attr('sensor.docker_containers_down_list', 'down_containers') | default([], true) | list %}
{{ down_items | join(', ') if (down_items | count > 0) else 'none' }}).
- alias : "Docker Telemetry Template Refresh"
id : docker_telemetry_template_refresh
Carlo Costanzo
1 week ago