>>>Regarding SQL Server for Linux, my guess is it's designed to be a daemon that can run in an all-Linux environment, where there is no SMB whatsoever and all clients are running Linux/BSD/*n?x. I don't see that as an issue if it and its clients can talk via, say, TCP/IP.
>>
>>I remember reading that the Linux version is based on something similar to a container with some features missing on the contaner implementation. Did not try to verify, but wondered if this - if true - is a hint that the current version is more a red herring...
>
>There's an interesting article at https://blogs.technet.microsoft.com/dataplatforminsider/2016/12/16/sql-server-on-linux-how-introduction/ . I didn't read the whole thing.

Thx for the link, which is somewhat more pro MS than the article I read. Still, Drawbridge / PAL fit quit well with the things I had gathered.

>
>Not sure what you mean by "red herring". I think MS is serious about this product, they probably want to steal some Oracle sales on various *n?x platforms. Or offer a paid, supported RDMBS on those platforms (other than Oracle). Or maybe they have existing SQL Server customers looking to go to the cloud who want to save money running their databases on Linux rather than paying for Windows Server licenses. There are probably lots of other business reasons they would want to do this.

The old article went on (much broader) that not everything available on Win would be under Linux, sometimes just because no GUI for it exists, in other cases the functionality is not available at all.

Red Herring:
Coupled with the basic difficulty to find certain things in MS products DRagan often laudates, going the Win way sems like the better option unless dire higher authority like company policy supported by BOFH makes that impossible. At least a Linux version might allow marketing droids to postpone decisons against them.


my 0.02€

thomas