NAV 2009 R2 High availability/redundancy


I recently moved into a company that is using NAV 2009 R2. It’s my first time working with this platform and I’m exploring some expansion/high availability options.

Our current architecture is that all of our users connect via a RemoteApp to an RDS Gateway server, which hosts the actual NAV application. The NAV application on that server then talks to a database server that interfaces with an enterprise grade NAS which contains the NAV db and logs. This seems to work well, and I have no issues with the functionality or performance.

My concern is that there are two potentially significant single points of failure - the RDS server and the DB server. We have ~20 users that work in NAV all day every day, as a company we are expanding, and downtime would severely impact our operations. I’ve tried searching around this forum and Microsoft’s website and I’ve found a few things, but I’m looking for some personal views/experiences -

  • What does your company do for high availability/redundancy of your NAV system?
  • I’ve seen several mentions that NAV 2009 can’t be clustered - how do you load balance, multiple gateways?
  • Do normal SQL2008 HA techniques apply?
  • How do customizations to NAV work in a multiple server environment?

We’ve reached out to our implementation partner about this as well, but we’re trying to expand our knowledge base. Thanks for any information you can provide.

Many of the standard HA solutions can be used with NAV. The challenge is in selecting which ones best fit your needs and budget. HA is sort of like buying insurance. You have to decide how much you need in case something we to happen.

My 1 question is which NAV client to you use?

For the database (SQL) server, the standard SQL HA options (clustering, mirroring, log shipping) are all valid choices. At one poitn or another I’ve used each of these in a NAV system. There are also many third party options but I won’t comment on those as I have little experience with them.

Your RDS environment can be handled with load balancing clusters.

The one part that can be an issue is the middle-tiers for the RTC. Usually we will just put up multiple services on different computers.