Skip to content

fix(elasticsearch): gate the rolling-upgrade health wait on expected node count #198

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Oddly merged 1 commit into from
Aug 12, 2026
Merged

fix(elasticsearch): gate the rolling-upgrade health wait on expected node count #198

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 18 additions & 2 deletions roles/elasticsearch/tasks/elasticsearch-rolling-upgrade.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -154,11 +154,21 @@
validate_certs: "{{ elasticsearch_validate_api_certs }}"
force_basic_auth: true
register: elasticsearch_response
until: ((elasticsearch_response.json | default({})).status | default('')) in _elasticsearch_upgrade_health_statuses
# Also gate on the expected node count: elasticsearch_api_host defaults
# to localhost, so a just-restarted node briefly reports
# number_of_nodes: 1 before peer discovery finishes. Without this
# guard, a mixed-version 2-node upgrade can lock at status: yellow +
# unassigned_shards: 1 until the health-retry budget times out —
# even though everything is fine, we're just too early.
until: >-
((elasticsearch_response.json | default({})).status | default('')) in _elasticsearch_upgrade_health_statuses
and ((elasticsearch_response.json | default({})).number_of_nodes | default(0) | int)
>= _elasticsearch_upgrade_expected_nodes | int
retries: "{{ elasticsearch_upgrade_health_retries | int }}"
delay: "{{ elasticsearch_upgrade_health_delay | int }}"
vars:
_elasticsearch_upgrade_health_statuses: "{{ ['green'] if elasticsearch_upgrade_wait_status == 'green' else ['green', 'yellow'] }}"
_elasticsearch_upgrade_expected_nodes: "{{ groups[elasticstack_elasticsearch_group_name] | default([inventory_hostname]) | length }}"
no_log: "{{ elasticstack_no_log }}"
changed_when: false
when: _elasticsearch_cluster_reachable | bool
Expand Down Expand Up @@ -315,13 +325,19 @@
validate_certs: "{{ elasticsearch_validate_api_certs }}"
force_basic_auth: true
register: elasticsearch_response
until: ((elasticsearch_response.json | default({})).status | default('')) in _elasticsearch_upgrade_health_statuses
# Same expected-nodes guard as the pre-node-down wait — see comment
# there for the flake it prevents.
until: >-
((elasticsearch_response.json | default({})).status | default('')) in _elasticsearch_upgrade_health_statuses
and ((elasticsearch_response.json | default({})).number_of_nodes | default(0) | int)
>= _elasticsearch_upgrade_expected_nodes | int
retries: 30
delay: 30
vars:
# Same gate as the pre-node-down health wait so both sides of the
# rolling loop honour the same operator preference.
_elasticsearch_upgrade_health_statuses: "{{ ['green'] if elasticsearch_upgrade_wait_status == 'green' else ['green', 'yellow'] }}"
_elasticsearch_upgrade_expected_nodes: "{{ groups[elasticstack_elasticsearch_group_name] | default([inventory_hostname]) | length }}"
no_log: "{{ elasticstack_no_log }}"
changed_when: false

Expand Down
Loading