Oh, I forgot to mention the main reason we went with Bluecat rather than Infoblox- Command Line Access baby! With Infoblox you only had access to their watered down CLI. With Bluecat you had full access via SSH. That came in handy more times that I care to recount. If I'm going to be responsible for managing a box, I want to have access to all the tools that box has, not just the ones the vendors deems stupid proof. When troubleshooting an issue, I could SSH to the box and look at the files that I needed, change the logging to whatever I wanted, run tcpdumps, everything. With Infoblox that was not an option. Had there been a NIC setting problem with the Infoblox device, there wouldn't have been jack I could do about it. With Bluecat I could actually fix the problem without having to wait for a solution to be developed, tested, packaged and shipped out. Okay, I'll stop now. :)
On Fri, Oct 3, 2008 at 12:01 PM, Dawn Connelly <[EMAIL PROTECTED]>wrote: > My best advice is to FULLY test the appliances before you buy then. Sure, > appliances limits the stupid people headache, but they introduce a bunch > more headaches that you had no idea were coming. Infoblox sales people are > VERY aggressive from what I have seen. Use that to your advantage. Make them > do a bake off so you can really see what you are buying and make sure that > bake off reflects the scale of your environment- not just the functionality. > The Bluecat guys aren't nearly as cut throat but they seem to be a good > group of folks to work with. If you are looking at buying a big enough > environment, they'll send some Canadian down to do a bake off. Make the > vendor early their paycheck. > > I had a similar situation. My personal preference was BIND on Solaris but > because the user base didn't have the necessary skill set to make that > option viable, I *had* to look appliances. We evaluated three- GTM by F5 > networks, Infoblox and Bluecat. GTM wasn't scalable for what we needed so > was out of the race pretty quickly. Infoblox- there were some fundamental > issues that we had that knocked them out. Those issues might be resolved by > now though. It was how the named process was handled. Any time you made any > changes to the named.conf file, it would stop and start the named > process...rather than leveraging rndc commands. That meant that there would > be a rolling blackout in the environment. That's really a non-issue is small > or medium environments, but with thousands of zones, the boxes would be > effectly down for almost two minutes at a time. Not okay. Also the fact that > the database was home grown rather than a standard database platform. If we > ever had to do a restore on a non-infoblox device we would have been SOL. > Like someone else mentioned, there is a compromise between ease of use and > functionality. The compromise on Infoblox was more than I was willing to > give up. We opted for Bluecat in the end. That being said, I wouldn't > exactly be a person they would list as a reference. We had NUMEROUS problems > with their appliances. The major ass biter was the fact that the database > replication between the management servers would periodically just die...and > would have no awareness that it was dead. Deployments would get hung pretty > frequently. The named process on the DNS appliances would die periodically > for no known reason. The solution more often than not was to reboot the > boxes. Even things like the fact that they are using an old Linux kernel > that has problems with forgetting the NIC setting upon reboot. I ended up > having to put a script that would run at boot to hard set the NIC. One thing > I will say for Bluecat is that if you can make a good argument for a needed > feature, you'll get that feature. > > > On Fri, Oct 3, 2008 at 6:36 AM, Linux Addict <[EMAIL PROTECTED]>wrote: > >> Folks, I am looking to re-architecture our NS Infrastructure. Can you >> guys suggest me if there are any Bind-Based Appliances available and >> authorized by ISC itself? >> >> Thanks, LA >> >> >> >> > > > -- > Google for President > YouTube for VP > in any year divisible by 4 > > -- Google for President YouTube for VP in any year divisible by 4
