What do you guys do for high-availability DR?

We're currently in the process of developing a replacement to our OS X - hosted server applications, but it has hit a problem which may increase the time we need to stay dependent on an OS X Server solution.

Until now, we've had a probably-familiar-to-most-Apple-sysadmins arrangement of a daily server->server backup plus failover... which has been used with a slightly higher degree of frequency than I'd like. And due to various problems we've had - not truly serious but with some notable productivity impact, not to mention that the Enterprise Support we have has been anything but - we need to implement a bolstering of our OS X Server availability in-house, as well as making more offsite availability arrangements.

What do you guys do for high-availabilty, immediate server / site failover scanarios without any data loss for OS X?

Trip.Tucker said:

If you are serious about ensure no business disruption, and you do not have in-house skillsets (which I assume you do not as you are asking on a forum)

Click to expand...

Good point, we have very little in the way of extensive knowledge of the DR products available for OS X. However, we have significant experience of the same on Windows.

I would strongly suggest seeking outside counsel. If downtime will cost you money then invest in your 24/7 uptime by having a 3rd party come in and address your DR needs correctly. They should be able to recommend any number of solutions for each piece of the whole pie from replicated storage on SANS to multiple path redundancy with IP balancing, etc etc.

It is a complex question that you pose and no amount of posts in one forum will ever amount to the professional services that you require from a firm that has this as it's main business.

Spend the dollars, get some consultants in and insure your future.

Click to expand...

Oh we have.

The answers we're getting are not satisfactory - and we already operate redundant SANs with multiple IP redundancy. What we're angling for however is an actually viable multiserver (on a SAN or not), multi-site, zero-downtime, replication-gap-less (or absolutely minimised) solution. And from what I've learned so far - at least from the technical sales guys we talked to - it seems there are several pieces missing in the OS X Server library in order to achieve that.

Which is why I'm asking. This is one occasion in which I'm truly angling for "The Wisdom of the Crowds", for an outline of how you approached high-availability DR especially with one or more backup sites, what compromises you made, and how it has worked out. I'm not necessarily asking for advice - I'm asking for your implementation experiences which I can consider alongside all the other info we have.

As I said, we have.