47 lines
1.4 KiB
Ruby
47 lines
1.4 KiB
Ruby
# frozen_string_literal: true
|
|
|
|
class UnsetEscalationPoliciesForAlertIncidents < Gitlab::Database::Migration[2.0]
|
|
disable_ddl_transaction!
|
|
|
|
restrict_gitlab_migration gitlab_schema: :gitlab_main
|
|
|
|
class EscalationStatus < MigrationRecord
|
|
include EachBatch
|
|
|
|
self.table_name = 'incident_management_issuable_escalation_statuses'
|
|
|
|
scope :having_alert_policy, -> do
|
|
joins(
|
|
'INNER JOIN alert_management_alerts ' \
|
|
'ON alert_management_alerts.issue_id ' \
|
|
'= incident_management_issuable_escalation_statuses.issue_id'
|
|
)
|
|
end
|
|
end
|
|
|
|
def up
|
|
EscalationStatus.each_batch do |escalation_statuses|
|
|
escalation_statuses
|
|
.where.not(policy_id: nil)
|
|
.having_alert_policy
|
|
.update_all(policy_id: nil, escalations_started_at: nil)
|
|
end
|
|
end
|
|
|
|
def down
|
|
# no-op
|
|
#
|
|
# We cannot retrieve the exact nullified values. We could
|
|
# approximately guess what the values are via the alert's
|
|
# escalation policy. However, that may not be accurate
|
|
# in all cases, as an alert's escalation policy is implictly
|
|
# inferred from the project rather than explicit, like an incident.
|
|
# So we won't backfill potentially incorrect data.
|
|
#
|
|
# This data is functionally safe to delete, as the relevant
|
|
# fields are read-only, and exclusively informational.
|
|
#
|
|
# Re-running the migration will have no effect.
|
|
end
|
|
end
|