Sys Admins -- Your thoughts on best practices?

GLaDOS11

I work for a small office...we have about 15 employees (I'm the only real tech guy). We run a Windows Server environment with about 10 physical servers and ~50 VMs. I'm trying to put together a "best practices" guide for internal use and I'd like to get some input from all the industry professionals on here. I'm wondering about things like how often to reboot servers? When to install Windows Updates? How do you handle server monitoring for performance issues and event log errors? Anything along those lines would be really, really helpful.

Thanks very much

Asif Dasl

This is really going to be down to your environment, all servers are not created equal - and you haven't given any details on what they do. Some are more important than others and some updates are more important than others. The short answer is - it depends, you try to keep everything running as long as possible before eventually you have do a reboot.

I haven't worked in a high pressure 24x7 workplace, mostly small businesses, so you may get different answers than mine. Also it depends if you are running WSUS and if you are testing updates before approving updates.

As for server metrics, it depends again if you've got the budget to do it or not. You could just use PerfMon or something like this.

phoeneous

Just curious, what do 15 employees need 10 servers and 50 vm's for? Seems like a lot.

Deathmage

maybe they have tons of unpaid interns

tstrip007

For server monitoring I use Solarwinds SAM and love it. I have node up/down alerts to text as well as drive space precentage alert texts. The alerting possibilites are endless and very easy to use/configure. Highly recommend and with only 50 VMs*, wouldnt be too expensive.

jdancer

Does your department or organization use any kind of issue tracking software? I'm a big fan of ITIL. A couple of full open-source ITIL implementations is iTop and OTRS.

I also use Zabbix for monitoring.