Eric Stewart: Running Off At The Mouth

Attempting (in my scatterbrained manner) to cover all the bases of an interesting issue I encountered recently and failed to fix without assistance. It illustrates why I’m not a CCIE yet, and why, no matter what you see, you shouldn’t assume anything when working through odd issues.

As is not unusual, I run into an edge case that might be specific to us. While attempting to monitor IPv6 reachability of an NTP server I had just upgraded, I ran into an interesting case where NTP was responding via IPv4, but not IPv6, and they appeared to be identically configured.

Just a quickie, probably subject to some editing later on, about how to use the “bind” option multiple times with a given service – mainly, to have a TFTP server operate from behind a virtual IP as the virtual IP (instead of becoming its real IP in the middle of the transfer).

I finally got the job I’ve been wanting for, oh, nearly a decade now (if not more). In this job I’m the go-to admin for the servers (almost all CentOS based) that support a large (3000+ device) network. Cisco routers run the network, and we have servers that are connected to “monitor” ports. One of the principle tasks of these servers are to allow network engineers to run packet captures. Thing is, interesting things can happen as far as IPv6 is concerned when you have systems getting IPv6 router advertisements on server interfaces that don’t actually do more than just accept packets …

THIS IS A LONG ONE. I’ve been reading up on IPv6, but haven’t had much of a chance to actually play with it. For the average user, once implemented, it’s supposed to be a seamless switch. I doubt that’s going to be the case. For the geek (particularly the network guru), it’s going to be a major change. Here’s what it is, and what I think about it …