Problems via API not in sync with Zabbix Web UI

[StormScorpion]

We use Zabbix API (problem.get) to integrate Zabbix with our Jira instance. We get problems reported for hosts via API that are not visible in Web ui.

I was able to reproduce this issue:

1. Create host, discover items like filesystems
2. Trigger an alert by filling disk space
3. Disable the disk space trigger or item
4. Zabbix wont show you the disk full alert, but problem.get for this host will still report this problem

Are we doing something wrong? Or is this something that Zabbix does not handle well?

{
"jsonrpc": "2.0",
"method": "problem.get",
"params": {
"hostids": [
1234
],
"output": "extend",
"sortfield": [
"eventid"
],
"severities": [
2,
3,
4,
5
],
"sortorder": "DESC",
"symptom": false,
"suppressed": false
},
"id": 1
}

[Andrejs Poddubņaks]

Hi!
When you disable a trigger or item:
- The frontend hides its problems
- but the problem events stay in the DB, so problem.get still returns them. This is clearly described in the documentation on triggers:

From official Zabbix documentation (Zabbix 7.4) https://www.zabbix.com/documentation...hosts/triggers

If you need the API to provide the same data as the Web UI, you need to repeat the logic on your side:
1. First, call problem.get as you do now.
2. Take objectid (triggerid) from the results and do trigger.get with the filter status: 0 (enabled) and, if necessary, selectHosts to check that the host is not disabled.
3. When integrating with Jira, only leave problems whose triggers/hosts are enabled.

[Andrejs Poddubņaks]

Hi!
When you disable a trigger or item:
- The frontend hides its problems
- but the problem events stay in the DB, so problem.get still returns them. This is clearly described in the documentation on triggers:

From official Zabbix documentation (Zabbix 7.4) https://www.zabbix.com/documentation...hosts/triggers

If you need the API to provide the same data as the Web UI, you need to repeat the logic on your side:
1. First, call problem.get as you do now.
2. Take objectid (triggerid) from the results and do trigger.get with the filter status: 0 (enabled) and, if necessary, selectHosts to check that the host is not disabled.
3. When integrating with Jira, only leave problems whose triggers/hosts are enabled.