Monitoring Postgres Cluster with Zabbix using pg_monz

[jtech]

Hello

I'm facing some issues on pg_monz under Zabbix v3.2.11 (sadly an old version to support pgCayenne). It is working good and it has even Telegram integrated.

The issue starts when I try to use the Cluster features available on the "Template App PostgreSQL SR Cluster" and "Template App pgpool-II watchdog".


So, I have created 2 "Host groups" according to pg_monz documentation: "PostgreSQL" and "pgpool" and added the respective hosts.
By the way, the other templates such as " Template App pgpool-II-36", "Template App PostgreSQL" and "Template App PostgreSQL SR" works great with the triggers and graphs.

The error messager from "Template App PostgreSQL SR Cluster" are below:

No items for key "psql.primary_server[/usr/local/bin,/usr/local/etc]" in group(s) "PostgreSQL".
No items for key "psql.running[/usr/local/bin,/usr/local/etc]" in group(s) "PostgreSQL".
No items for key "psql.standby_server[/usr/local/bin,/usr/local/etc]" in group(s) "PostgreSQL".

I have tested the respective scripts manually on the Linux box with success, which means the authentication is fine:

[zabbix@standby ~]$ /usr/local/bin/pgsql_primary.sh /usr/local/etc/
0

[zabbix@standby ~]$ /usr/local/bin/pgsql_simple.sh /usr/local/etc/
1

[zabbix@standby ~]$ /usr/local/bin/pgsql_standby.sh /usr/local/etc/
1

Here are the details of one of the items:
Template: "Template App PostgreSQL SR Cluster"
Item name: Number of standby server
Type: Zabbix aggregate
key: grpsum[{$PG_HOST_GROUP},"psql.standby_server[{$PGSCRIPTDIR},{$PGSCRIPT_CONFDIR}]",last,0]
Type of info: Numeric (unsigned)
Data type: Decimal

I have reviewed the items as suggested by Ingus in this post without success

Also, the same problem is happening with the "Template App pgpool-II watchdog":

No items for key "pgpool.have_delegate_ip[/usr/local/bin,/usr/local/etc]" in group(s) "pgpool".
No items for key "pgpool.running[/usr/local/bin,/usr/local/etc]" in group(s) "pgpool".

There are no errors on Zabbix log (configured as debug level 3)
CentOS Linux release 7.6 (Core)
Postgres v11.2 Streaming Replication Cluster (Master / Slave)
pgpool-II v3.6.18 + watchdog + heartbeat

Thanks in advance for any help!

[jtech]

Had reviewed the process today but haven't found the cause of the issue.
Zabbix aggregate syntax:

<grpmax|grpmin|grpsum|grpavg>["Host group(s)", "Item key", "<last|min|max|avg|sum|count>", "parameter"]

Template item syntax:

grpsum[{$PG_HOST_GROUP},"psql.standby_server[{$PGSCRIPTDIR},{$PGSCRIPT_CONFDIR}]",last,0]

Bash script with SQL code located on postgres server (tested with success)

./usr/local/bin/pgsql_standby.sh /usr/local/etc

Zabbix error message:

No items for key "psql.standby_server[/usr/local/bin,/usr/local/etc]" in group(s) "PostgreSQL".

It might the bug ZBX-11658 due to my Zabbix v3.2.11.