Real Time Server Monitoring Tool with web interface
 

Hi all,

i am currently working on a centos 5.7 x64bit os.

i want a tool that should be free & can monitor my servers real time performance & hardware utilization.

also want to monitor this on some front end as we say web based monitoring.

as we are using nagios for server alerting in our organization. but i want a tool that can monitor my server load, memory uses, swap utilization & services etc. with web based monitoring eg. graphs.

if you have any experience or any one currently using these kind of tool plz let me know.

i searched & found below links, but most of all tools are paid versions & network monitoring tools.

http://community.spiceworks.com/how_...ftware-details

http://www.junauza.com/2010/12/free-...-software.html

http://slodive.com/freebies/server-monitoring-tools/

Thanks in advance.... :)

groundwork, hyperic, nagios, zenoss, opennms... they'll all do this for free. Given you're already using Nagios, not sure why you're asking tbh. You can bolt on graphing tools like pnp4nagios or nagiosgraph easily enough.

zabbix.

http://nagios.sourceforge.net/docs/3...g-windows.html

This page is about windows servers. May be helpful for you

http://nagios.sourceforge.net/docs/3_0/addons.html#nrpe

And the above link may also be helpful

no no no no! never use nrpe! it's junk. Just use check_by_ssh instead - more secure, no additional services running, vastly better.

I have just started using nrpe. I am not expert/master in that. Here is something that the above said link says about nrpe and check_by_ssh

I'd rather have a fractional increase in load that a gaping security hole. There is no realistic downside over NOT using NRPE. If nothing else, it's just SO much easier not having another service running. Why would you want to install an additional service if you can do it better with what you already have installed by default?

Agreed, totally. Between the two, the SSH check is much easier, ESPECIALLY if you have lots of servers, since SSH is always running. More services = more to check/make sure is running/keep updated. And I'm monitoring about 1500 Windows/Linux/Unix servers with Nagios, and the 'increased load' is minimal.

One option that is quite common is Cacti http://www.cacti.net/downloads/docs/html/index.html, http://www.cacti.net/screenshots.php

Awesome! everybody is supporting check_by_ssh over nrpe. It must be fun. I will look into it. Thanks for great suggestion :)

Was reading again. So whatever is listed below, I am ignoring it.

Well, that's been acknowledged, but I think you need to put context behind it.

The server where Nagios is running will have a higher CPU load...but that's your monitoring system, and if you have a dedicated box just for that, the higher load won't impact ANY users or services. So, would it matter that your CPU load went up, if it didn't affect anything? Also, while it's true that CPU load will increase...they also don't say how MUCH it will increase. Chances are your CPU load now is trivial...probably less than 5%. I'm monitoring a LOT of systems, and have only ever seen the load spike up to about 25%....which means that I have a CPU that's 75% UNUSED. As with anything, though, the best way to see what fits for you, is for you to try it.

TB0ne, Thank you very for your nice explanation. It is really useful. Can please point to a resource/web-page explaining/comparing check_by_ssh and NRPE?

Thanks again

You're welcome, but no, I can't. Again, the best thing for you to do, is to try both of them, and see what fits your needs best. You can read websites all day, but you won't know what is right for YOU, until you try it.