cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2669
Views
0
Helpful
3
Replies

Proactive network maintenance plan

fearless.rabbit
Level 1
Level 1

Hello everyone,

I found this community helpful many times before, and I`m testing my luck again This time I would appreciate some guidelines regarding proper network maintenance plan. I basically want to propose this to our management and senior IT personnel (no, they never though of it), and have much more pleasant working time for myself.

So far, I managed to collect some of the pieces, but would greately appreciate additional comments by more experienced guys and galls around here. Is there anything I missed? Some best-practice I didn`t know about? How do you do it?

  • Change control

          Schedule maintenance window for any non-critical tasks. Make sure most recent backup configs exist before changing anything in the network.

  • Documentation

          Keep it electronic. Have centralized backup configs. Enable TFTP on the network and manually, or using archive, backup configs from all the           devices one a week. Have syslog server running as well.

  • Consistency

          Enable NTP server in the network. Schedule IOS upgrades and make sure all devices run same IOS versions. Test backup configs in scheduled maintenance window.

  • Physical maint.

         

          Schedule physical maintenance once every 3 months. Open all the devices, clean the dust and arrange the cables. Consult management for plan

          regarding spare devices and MTBF (meantime before failure)


  • Have monitoring system in place

        We have Zenoss partially running. Plan so far is to get it fully working and SNMP monitoring all the devices.

I know this is just a novice approach, and that is exactly the reason why I`m asking you - what should I add/consider in order to create any decent maintenance plan?

3 Replies 3

Michel Hegeraat
Level 7
Level 7

my 2 cents,

  • Documentation

Use versioning system to keep track of previous configs.

Have some reports made from the syslog that check for obvious things like:

CDP issues

Port errordissabled

Config change

Authentication failures

  • Have monitoring system in place

Get some baseline info so that you can see what is "normal" for response times and interface load on a normal business day.

Make sure you monitor all the links between the devices.  These should be error free, certainly on the backbone

Cheers,

Michel

Thank you for your reply Michel. Every comment counts

fearless.rabbit
Level 1
Level 1

Would you enable NetFlow on all interfaces of all devices in the network? How do you use NetFlow in your monitoring?

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: