Koschei: Add workaround for database issue

UnappliedChange database table can run out of IDs under normal
operation.  Once this happens, there is backend outage that requires
manual admin intervention.

See: https://github.com/fedora-infra/koschei/issues/234
This commit is contained in:
Mikolaj Izdebski 2022-08-17 07:22:14 +02:00
parent e9bced043b
commit c2c063d22f

View file

@ -66,3 +66,7 @@ koschei_cron_jobs:
- name: track-churchyard
schedule: "31 0-23/3 * * *"
command: "koschei-admin psql <<<'UPDATE package SET tracked = TRUE WHERE id IN (SELECT p.id FROM package p JOIN package_group_relation pgr ON p.base_id = pgr.base_id WHERE NOT p.tracked AND pgr.group_id = 79)'"
# Workaround for https://github.com/fedora-infra/koschei/issues/234
- name: rewind-unapplied-changes
schedule: "42 8 17 */3 *"
command: "koschei-admin psql <<<'BEGIN; DELETE FROM unapplied_change; ALTER SEQUENCE unapplied_change_id_seq RESTART; COMMIT'"