Hi guys. I’m bit confused how to deploy distributed poller after reading different docs and how we have deployed it.
My environment consist of three servers:
a.one webserver(nginx),
b.one server which contains db and rrdcache and
c. One poller device.
Webserver and poller are pointinf into the same server(which contains a database and rrdcache). All validations seems to be ok, but there are some strange behaviours that I need to understand.
1.Is it necessary to configure a memcache? When I run validate always appear a warning message about it
2. discovery and polling Cron are running only in poller machine,but when I launch a debug using webgui all tasks are performed by webguiserver not by poller server.
3.I have notice that rrdcache files are both located into poller machine and database server. Checking netstat I could see how poller connect which database and rrd files located in that machine,but I don’t know why is it creating new files. Every file increases disk usage locally. I have configured on poller config.php rrdcacheserver:42217.
Last, I have checked that there are certain event log outdated.Database,time date are ok in all three servers.
But we have noticed using another standalone Lms as a lab and polling same machines without problem (we check all logs,launched different scripts) that we don’t receive all events as example ifOperStatus on time.
So, It’s is possible to deploy distributed environment as we have performed or do you recommend to configure differently? It is a normal behaviour that rrd files are located in different servers?
Can this typology affect receiving updated events via gui?
Thank you very much in advance
Kind Regards