Hi,
I have a Zabbix Server in ec2 and the Zabbix backend in Amazon RDS in the same region.
There is also one proxy in a physical data center , the Zabbix busy data sender process is constantly on 75% percent on the proxy. The overall NVPS is 735, 500 is just the Zabbix proxy itself. The Zabbix DB RDS instance is quite big, a db.m3.xlarge size and all monitoring graphs indicates there is no performance issues in terms of CPU/Mem/Disk usage.
However there is an issue we are facing when the Amazon RDS start taking backup of Zabbix DB, when this happens brief IO suspension is expected and this is when there will be performance degradation on both read and write operations. We are seeing some increased write latency (50-60 milliseconds) only during the backup time window and this is when our monitoring is getting affected.
The Zabbix busy data sender process goes on 100% during the backup process and we start getting monitoring alerts all agents unreachable only for the servers under the Zabbix Proxy. This is happening every day during the backup process till it finish.
Since this is not a capacity issue and we have tweak everything we could in Zabbix config ( as far as I know there is no way to increase the number of zabbix data sender process) I'm wondering if anyone has any recommendations regarding performance tuning in Amazon RDS for Zabbix? We are using MySQL and would like tweak it through the database parameter group in Amazon RDS but couldn't find much information on the internet on what / how to do it.
Any help ?
I have a Zabbix Server in ec2 and the Zabbix backend in Amazon RDS in the same region.
There is also one proxy in a physical data center , the Zabbix busy data sender process is constantly on 75% percent on the proxy. The overall NVPS is 735, 500 is just the Zabbix proxy itself. The Zabbix DB RDS instance is quite big, a db.m3.xlarge size and all monitoring graphs indicates there is no performance issues in terms of CPU/Mem/Disk usage.
However there is an issue we are facing when the Amazon RDS start taking backup of Zabbix DB, when this happens brief IO suspension is expected and this is when there will be performance degradation on both read and write operations. We are seeing some increased write latency (50-60 milliseconds) only during the backup time window and this is when our monitoring is getting affected.
The Zabbix busy data sender process goes on 100% during the backup process and we start getting monitoring alerts all agents unreachable only for the servers under the Zabbix Proxy. This is happening every day during the backup process till it finish.
Since this is not a capacity issue and we have tweak everything we could in Zabbix config ( as far as I know there is no way to increase the number of zabbix data sender process) I'm wondering if anyone has any recommendations regarding performance tuning in Amazon RDS for Zabbix? We are using MySQL and would like tweak it through the database parameter group in Amazon RDS but couldn't find much information on the internet on what / how to do it.
Any help ?