http://phprpms.sourceforge.net/mssql
HTH -----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Petr Klíma Sent: Monday, June 13, 2005 7:18 AM To: CentOS mailing list Subject: [CentOS] Php package for Microsoft-SQL
Hello
I need connect from PHP to MS-SQL. Is there an way how to make just something like php-mssql-x.x.x.rpm without modifiing nad rebuilding full PHP RPMS package (this way I know ...)? I do not want loose abylity of automatic security updates ...
On Mon, 2005-06-13 at 10:49 -0400, Todd Rittinger wrote:
HI! Not sure if this is what you are looking for, but it may be of interest: http://probind.sourceforge.net/
On Mon, 2005-06-13 at 11:37 -0400, Adam Breaux wrote:
On Mon, 2005-06-13 at 12:13 -0400, Dan Anderson wrote:
Has anyone looked at or had any experience with Maintain? http://osuosl.org/projects/maintain/
Not to introduce vaporware, but what I think enterprises are looking for are a true, layer 2 and layer 3 name services product. That's why a number of them are deploying ActiveDirectory Services (ADS) integrated DHCP/DNS, or possibly a 3rd party product like Lucent's or another company (can't seem to remember them).
ISC BIND and DHCP were designed for the Internet with free-form approaches that are horrific on a managed, private network. Deploying BIND and DHCP adds massive amounts of overhead and planning that is unnecessary, and can be a continuing security nightmare. Although these projects attempt to mitigate the management issues, they are not solving the core problem.
An integrated layer 2 + layer 3 name server should do many things.
1. Introduce contemporary peer-replication
The concept that only the "primary" server can be modified is '80s in concept. Peer-replication -- especially across a hierarchial enterprise is required today. I'm sure some of the technologies gained from Red Hat's GPL of Netscape Directory Server could be leveraged.
2. Enforce good system nomenclature
This solves many details. One of the biggest problems with DDNS is that a simple node can overtake a server's name/IP. There should be protected IP masks and/or system names that are delegated to specific systems in a well-integrated layer 2 + layer 3 name server.
3. Enforce good subnetting design
Allow 2 and 3-tier corporate designs in subnetting the reserved, private networks. Not only is this good for IPv4, but also prepares us for IPv6 (which is integrated with layer 2 addressing). It also solves another _huge_ issue for enterprises -- the requirement that reverse zones be classful.
4. Tame legacy protocols
This is my personal favorite. An integrated layer 2 + layer 3 server should support some legacy protocols -- both to proxy them, as well as identify them. E.g., if someone is broadcasting NetBIOS or SAP, the integrated server should not only integrate that name into its tree (assuming some basic logic), but also give an easy way for administrators to identify nodes on the network that are still using them!
On Mon, 2005-06-13 at 12:27, Bryan J. Smith wrote:
- Tame legacy protocols
This is my personal favorite. An integrated layer 2 + layer 3 server should support some legacy protocols -- both to proxy them, as well as identify them. E.g., if someone is broadcasting NetBIOS or SAP, the integrated server should not only integrate that name into its tree (assuming some basic logic), but also give an easy way for administrators to identify nodes on the network that are still using them!
Does anything exist that has that 'basic logic'? The legacy forms work and scale worldwide because the authority to use names is carefully delegated. If two self-issued names are broadcast on the same network, who wins? What if they are on different subnets and can't see each other but you try to integrate them with such a tool? What if they normally live on different networks but are mobile and eventually collide? I'd really prefer not to let anyone's laptop claim to be the company email server and get away with it.