-
Problem report
-
Resolution: Fixed
-
Trivial
-
4.4.6
-
None
-
centos7 + zabbix-server-pgsql from officical repo, postgresql-11 + pgbouncer.
-
Sprint 66 (Jul 2020)
-
0.25
Steps to reproduce:
- Set up postgresql plus pgboucer as reverse proxy for zabbix-server
- Set up zabbix server (PG build) from official repo, point to pgbouncer as DB host
- Start up postgresql, pgbouncer, zabbix-server
- systemctl stop pgbouncer
- systemctl stop zabbix-server
Result:
Zabbix server cannot stop forever complaining that DB is down (see screenshot). So, during machine reboot zabbix stopping will run forever, preventing whole server from normal rebooting.
Expected:
Zabbix server WILL stop after some time if SIGTERM received AND DB is unreachable.
Also, it is expected to have:
[Unit] After=pgbouncer.service After=postgresql-9.4.service After=postgresql-9.5.service After=postgresql-9.6.service After=postgresql-10.service After=postgresql-11.service After=postgresql-12.service
in /usr/lib/systemd/system/zabbix-server.service (to properly ordering zabbix-server and PGDG official builds shutdown sequence).