mirror of
https://github.com/Security-Onion-Solutions/securityonion.git
synced 2026-05-11 05:40:32 +02:00
soup: update postgres backfill comment to reflect reactor removal
The reactor path is gone; so-minion now owns add/delete for new minions. The backfill itself is unchanged — postgres.auth's up_minions fallback fills the aggregate, postgres.telegraf_users creates the roles, and the bash loop fans to per-minion pillar files — so the pre-feature upgrade story still works end-to-end. Just refresh the comment so it isn't misleading.
This commit is contained in:
@@ -491,10 +491,11 @@ post_to_3.1.0() {
|
||||
/usr/sbin/so-kibana-space-defaults
|
||||
|
||||
# One-time backfill for minions that existed before the postgres Telegraf
|
||||
# feature shipped. Generate the aggregate pillar on the manager and create
|
||||
# the per-minion DB roles, then fan each minion's cred into its own pillar
|
||||
# file. Going forward the reactor handles each new salt-key accept with a
|
||||
# targeted fan-out, so a manager highstate no longer needs to iterate.
|
||||
# feature shipped. postgres.auth's up_minions fallback loop generates any
|
||||
# missing aggregate pillar entries; postgres.telegraf_users CREATEs the
|
||||
# matching DB roles; then the bash loop below copies each minion's cred
|
||||
# into its own pillar file. Going forward, so-minion owns add/delete for
|
||||
# every new minion, so this backfill is only needed on the upgrade boundary.
|
||||
echo "Provisioning Telegraf Postgres users for existing minions."
|
||||
salt-call --local state.apply postgres.auth,postgres.telegraf_users queue=True || true
|
||||
|
||||
|
||||
Reference in New Issue
Block a user