It’s been a long week, which is concerning since it’s only Tuesday, but we have a lot going on and it’s been a little hectic, so I can be forgiven for what I am about to say.
I happened to walk into the second floor Main Distribution Facility, or MDF, ten minutes ago, where all of our WAN and network core is and saw one of my network engineers, Leo, removing one of the routers. In fact the router is called core1 since everything goes through it. I had completely forgotten we were doing this today. (I did approve the change yesterday, but like I said it’s been hectic).We chatted for 5 minutes or so (with the router powered off) and I think he was going to run down and grab a cup of coffee before he finished up.
My next stop was down to sales and finance just to make sure no one noticed anything. No one did. Finance had a few questions on retired assets. In the meantime sales continued calling customers and selling product. Our call center never noticed anything either, even the remote users connecting in over VPN using VoIP. The WiFi connection on my laptop never missed a beat and I continued to get ~250Mbs as I walked through the building. It’s OK, I didn’t walk and type, I only checked the connection speed when I stopped….
For us it’s just another day in the life of an Enterasys network, but I’m thinking most IT Directors wouldn’t forget that they were upgrading their core routers. In fact most IT directors probably wouldn’t have slept the night before, or done it during the middle of a workday, but that’s why I’m pretty confident when I talk about Enterasys redundancy…
Now to be fair:
Now to be fair:
- We keep meticulous documentation.
- We spend a lot of time verifying that the configuration of the core routers stay in synch.
- We have been known to test reboot a switch or router after hours to make sure there are no issues.
- We make a lot of changes so have gotten pretty good at them.
- We will “blink link” on a server to move it from the switch we are upgrading to one we aren’t upgrading to avoid any issues, required for servers without redundancy.
But you still have to admit: it’s pretty cool….