recommendation for taking care of a network

[baselzind]

I'm handling switches at a site ( 6500 core , 4500 server farms , and 3xxx edge switches) I'm wondering what can i do beyond the below point to make sure the network is healthy:

-taking backups of all switches

-drawing a diagram of the network with uplinks ports and ip

-firmaware upgrades

[balaji.bandi · Accepted solution]

Set up a syslog Server (send all devices logs to syslog server)

setup an NMS

- check the Port-utilisations

- check any interface errors 

- check status of port going up ad down. (setup a alerts)

- setup a device reachability in NMS and if lost get alerts

[baselzind]

can i check the port utilization using this output "txload 1/255, rxload 1/255" from "show interface" ?

[balaji.bandi]

yes with NMS, what ever you looking to monitor.

or you can write own script to collect the data and alerts email be possible.