Zabbix perfomance after upgrade to 7.0
Hi everyone,
I have a Zabbix proxy that monitors an enterprise-level application, retrieving data in various ways through the API. Some of the data is sent using the standard HTTP-Agent method. Some are handled via an external check script for additional processing.
On version 5.0.18, everything was working well. Occasionally, some items would time out, but it wasn’t critical. The number of API requests registered by the application hovered around 60–80 per minute, which was acceptable for its API.
After upgrading to version 7.0.11, Zabbix started, let’s say, acting like a DoS attack on the application, increasing the number of requests to 200, and then to 400 per minute. This number of requests is not supported by the target application, and it starts returning 502 errors.
At the same time, I didn’t change anything in the monitoring configuration after the update. The only change I made was increasing all timeouts in General -> Timeouts to 30 seconds. This was necessary because some scripts take a long time to execute and don’t always return values within the default 3 seconds.
What could be the root cause of this issue?
I read about MaxConcurrentChecksPerPoller, but that’s not quite what I’m looking for. I have a feeling that the Zabbix proxy started generating more script executes or http requests per unit of time after upgrading to version 7.0.
I have a Zabbix proxy that monitors an enterprise-level application, retrieving data in various ways through the API. Some of the data is sent using the standard HTTP-Agent method. Some are handled via an external check script for additional processing.
On version 5.0.18, everything was working well. Occasionally, some items would time out, but it wasn’t critical. The number of API requests registered by the application hovered around 60–80 per minute, which was acceptable for its API.
After upgrading to version 7.0.11, Zabbix started, let’s say, acting like a DoS attack on the application, increasing the number of requests to 200, and then to 400 per minute. This number of requests is not supported by the target application, and it starts returning 502 errors.
At the same time, I didn’t change anything in the monitoring configuration after the update. The only change I made was increasing all timeouts in General -> Timeouts to 30 seconds. This was necessary because some scripts take a long time to execute and don’t always return values within the default 3 seconds.
What could be the root cause of this issue?
I read about MaxConcurrentChecksPerPoller, but that’s not quite what I’m looking for. I have a feeling that the Zabbix proxy started generating more script executes or http requests per unit of time after upgrading to version 7.0.