Agent's synthetic monitoring

[haiyi...@gmail.com]

Hi, Trask,

Even though Glowroot allow users to add synthetic monitor for an agent, say, ping a downstream service URL, It seems the execution of the URL connection checking is performed by central collectors, is this correct? I noticed even I shutdown the agent, the synthetic monitoring will still keep going afterward.

Assuming this is true, if the agent's server has special firewall/network settings, the synthetic monitoring performed at central collector may not reflect real result that performed by the agent; and, I think it would be simpler and less confusing to just configure synthetic monitors at Rollup levels. Is my understanding correct?

Thanks!

Haiying

[Trask Stalnaker]

Hi Haiying,

The idea for the synthetic monitors is to test the service where the agent is running (as opposed to testing a downstream service).  E.g. to alert on when the service itself is not accessible, or gives error / timeout during login.

Thanks,

Trask

[haiyi...@gmail.com]

Trask,

Thanks for explanation. So I assume this also confirms that the synthetic monitors are executed by the central collector. Please correct me if I am wrong.

Haiying

[Trask Stalnaker]

Yes, this is correct, synthetic monitors are executed by the central collector.

Trask