How do you monitor the state of the application and server (Linux)?

U

un1t2015-10-26 20:43:34

linux

un1t, 2015-10-26 20:43:34

I would like to see the load on the application and the web server in real time with graphs, processor, network, etc. And also alerts when certain parameters are exceeded.

Reply

Answer the question

In order to leave comments, you need to log in

5 answer(s)

A

Alexey Cheremisin, 2015-10-26
@un1t

I'm using collectd. Fast, simple, undemanding to resources., Reliable. Graphs are generated based on rrd, there is a simple web application in the kit that makes graphs and does cron.
If there are a lot of servers, like we have, 50+, then we put influxdb on them (before that we used graphite) and the grafana frontend. There is already collection and processing in real time. If the data needs to be further thinned out / processed, then we add another statd.
Total for large tasks: collectd (per host) - statd (one or more) - influxdb (one or more) - grafana (one or more).
For a single collectd server - web-collectd.

T

TyzhSysAdmin, 2015-10-26
@POS_troi

zabbix, nagios

I

Igor Vorotnev, 2015-10-26
@HeadOnFire

Nagios, Linux Dash, New Relic. Well, Uptime Robot.
And here are a couple of pages of tips from around the world.

M

Max, 2015-10-26
@MaxDukov

MRTG + get data via snmp

D

devel787, 2015-10-28
@devel787

highly recommended
https://mmonit.com/monit/
but I haven't had a chance to try it myself yet