Hi,
I'm currently running Zabbix 2.2.6 on my PRD environment (yes, I know, it's outdated, an upgrade is planned
).
It's running smoothly, with these values :
I'm running 300 pollers
We're installing a VMware platform, and I wanted to monitor it, so I added to the zabbix_server.conf file the following lines :
StartVMwareCollectors=5
VMwareFrequency=60
VMwareCacheSize=8M
And that's when hell started.
Immediately, the pollers were all busy (100%), and the zabbix frontend was displaying the infamous red label. On the dashboard, sometimes I could see that the Zabbix server was running (and still get the red ribbon), most of the time it was at not running, despite the fact that it was indeed running.
I first thought it was due to a massive amount of data to retrieve, but after several hours, the situation was the same.
Tried to restart, same result.
I rollbacked the config, removed the vmware part.. and suddently everything went back to normal.
I cannot believe that 5 vmware pollers can cause such a mess.
Does someone already have that kind of problem ?
I'm currently running Zabbix 2.2.6 on my PRD environment (yes, I know, it's outdated, an upgrade is planned
). It's running smoothly, with these values :
Code:
Parameter Value Details Zabbix server is running Yes x.x.x.x:10051 Number of hosts (monitored/not monitored/templates) 1251 1090 / 21 / 140 Number of items (monitored/disabled/not supported) 149569 113798 / 2078 / 33693 Number of triggers (enabled/disabled) [problem/ok] 55722 51246 / 4476 [588 / 50658] Number of users (online) 31 7 Required server performance, new values per second 1312.96 -
We're installing a VMware platform, and I wanted to monitor it, so I added to the zabbix_server.conf file the following lines :
StartVMwareCollectors=5
VMwareFrequency=60
VMwareCacheSize=8M
And that's when hell started.
Immediately, the pollers were all busy (100%), and the zabbix frontend was displaying the infamous red label. On the dashboard, sometimes I could see that the Zabbix server was running (and still get the red ribbon), most of the time it was at not running, despite the fact that it was indeed running.
I first thought it was due to a massive amount of data to retrieve, but after several hours, the situation was the same.
Tried to restart, same result.
I rollbacked the config, removed the vmware part.. and suddently everything went back to normal.
I cannot believe that 5 vmware pollers can cause such a mess.
Does someone already have that kind of problem ?
Comment