PostgreSQL: Streaming lag with {#MASTER} is too high (over 10m in 5m)

[grmnrvs]

Hello everyone, I'm monitoring a PostgreSQL11 database on Zabbix 4.2.6 with the following template (https://git.zabbix.com/projects/ZBX/.../db/postgresql), and everything works correctly, but in my dashboard appears the following trigger:

• "PostgreSQL: Streaming lag with {#MASTER} is too high (over 10m in 5m)"

This is the internal configuration of the template:

• Name: PostgreSQL: Streaming lag with {#MASTER} is too high (over {$PG.REPL_LAG.MAX.WARN} in 5m)
• Expression: {Zabbix server: pgsql.replication.lag.sec["{$PG.HOST}","{$PG.PORT}","{$PG.USER}","{$PG.D B}"].min(5m)} > {$PG.REPL_LAG.MAX.WARN} {$PG.REPL_LAG.MAX.WARN}

And this is the configuration of the item where this trigger is activated:

• Name: Replication: lag in seconds
• Type: Zabbix Agent
• Key: pgsql.replication.lag.sec["{$PG.HOST}","{$PG.PORT}","{$PG.USER}","{$PG.D B}"]
• Host interface: 127.0.0.1 10050
• Type of information: Numeric (float)
• Units: s

It's my only Zabbix server and we don't have replication activated with another server. What do you recommend me to do? Should I deactivate this trigger or should I correct something in the template or script that serves to monitor PostgreSQL?

Thank you.

[beandr]

It may be caused by tons of write traffic. Usually it happened when a lot of replication traffic comes through a slow network or when the nodes are too far from each others (by ping).
If you are using an rsync command, then can try a --bwlimit parameter to the rsync command that copies the WAL files.