Trapper data-send fails without explanation

[michaelNGV]

I'm trying to send data to a trapper item. The data are:

{"request":"sender data","data":[{"host":"Office Zabbix Proxy","key":"Router.ASSET1509","value":"1","clock":1657584151}]}

Zabbix replies:

{"response":"success","info":"processed: 0; failed: 1; total: 1; seconds spent: 0.000021"}

Obviously from the response, Zabbix is recognising and understanding the request, but it fails it. If I send two objects inside "data":[], Zabbix replies

{"response":"success","info":"processed: 0; failed: 2; total: 2; seconds spent: 0.000019"}

How can I get some information about why Zabbix is failing the request? /var/log/Zabbix does not exist, and journalctl -xe shows nothing relevant.

There is a host called Office Zabbix Proxy, and it does have a trapper item with key Router.ASSET1509. See attached screen-shots.

[tim.mooney]

Are you using zabbix_sender to fling the data at your server, or are you generating the JSON manually and then submitting it some other way?

Are you certain that "Office Zabbix Proxy" is the technical name for the host within Zabbix? It cannot be the display name, it must be the host name.

I wouldn't expect "journalctl" to show you anything useful, since Zabbix server is running and not experiencing any serious errors.

I recommend you look at the man page for "zabbix_server", in particular the --runtime-control with log_level_increase. You may be able to identify just a few subsystems that need to have their logging level increased (for example, the trapper, but that alone might not be enough).

Increasing log level to level 4 and then trying your trapper item, and then decreasing the log level back down may provide you with some insight as to what's going on.

[tim.mooney]

Are you using zabbix_sender to fling the data at your server, or are you generating the JSON manually and then submitting it some other way?

Are you certain that "Office Zabbix Proxy" is the technical name for the host within Zabbix? It cannot be the display name, it must be the host name.

I wouldn't expect "journalctl" to show you anything useful, since Zabbix server is running and not experiencing any serious errors.

I recommend you look at the man page for "zabbix_server", in particular the --runtime-control with log_level_increase. You may be able to identify just a few subsystems that need to have their logging level increased (for example, the trapper, but that alone might not be enough).

Increasing log level to level 4 and then trying your trapper item, and then decreasing the log level back down may provide you with some insight as to what's going on.

[michaelNGV]

Thank you! I was using the display name. Now I'm using the technical (host) name and it is working.