Johnson, Shaunn wrote: > Howdy: > > This is probably a silly question. Let's say > that I want to pull data from a remote database > (say, NT or Mainframe).
OK. > > For DB2, am I to understand that I need to > install client software package, connect to > that client and THEN run the perl DBI::DB2 > mods to get data from the remote DB2 > server? Is this the same for Oracle? Yes, and Yes. > I am guessing that it is, because when I > installed the Oracle DBD, it asked about > the ORA_HOME path ... but I don't recall > doing this for PostgrSQL ... *shrug* ... DBD::Pg requires libpg AFAIK, so you need PostgreSQL client software. > > What I don't get (and this is why I'm lurking > in the beginners mail list), is why can't I > just find what the remote server is > and connect directly to it rather than > having to deal with client software and, perhaps, > authentication problems and identifying the > node that I want a connection from. Each database has its own protocols for networking, so you need a client piece to handle that. Typically, the DBI driver talks to the database client libs, which handle the networking internally in some fashion. However, a really slick technique for DBI access across a network when installing the database middleware is impossible or impractical is to use DBD::Proxy (client) and DBI::ProxyServer (server). This is a Perl-only DBI middleware protocol that runs anywhere Perl runs. We are using this for accessing MS SQL Server on NT server from FreeBSD clients. > > Of course, I could be way off in my own world of > make believe (most likely) and could use a > primer in this as well. -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
