[RBASE-L] - RE: What's up with this?

[Dennis McGrath]

I believe it all boils down to Microsoft's sharing methodology. Sounds like #1 machine is hogging access to the DB since it is on a local disk.   My best guess is that you need to have the DB on a server. You will still see a performance hit when more that one user is connected, but not so drastic as to call it a crawl. And all users should see a reasonable reponse time.   Peep-toPeer sharing of a db has always been a somewhat risky way to save money too. Much safer to have the DB on a clean server where the server is not having to handle a lot of tasks unrelated to the db.   Others here can probably give you more definitive technical info.   Dennis McGrath From: rbase-l@rbase.com [mailto:[EMAIL PROTECTED] On Behalf Of Lee Bailey Sent: Monday, July 10, 2006 9:36 AM To: RBASE-L Mailing List Subject: [RBASE-L] - What's up with this? Hello All-   I’ve been monitoring the users group for a long time, and have been impressed with the expertise revealed in exchanged emails.  I am hoping that one of you networking gurus can point me in the right direction in solving a perplexing problem.   I have been an avid R-Base user since DOS 2.0.  I am currently running version 6.5, and have been for about 5 years.   Here’s the setup— Machine #1 is a 2.2 gighz 64 bit machine with 2 gig’s of ram.  Machine #2 is a 1.9 gighz 32 bit machine with 1 gig of ram.  Both machines are tied together via a router with a transfer rate of 100 mbps.  The router also provided access, by both machines, to the Internet via a DSL modem, also attached to the router.   The 20 meg database files (RB1, RB2, RB3, and RB4) are on machine #1, with application command files on both machines to speed processing in the common database.  Being an old R-Base guy, much of the programming is done via command files, crunching a lot of data, to achieve the full relational data base power.   Here’s the problem-- When machine #1 accesses the database as a sole user, the applications, command files, forms, reports, etc. run very quickly.  When machine #2 accesses the database solely, applications also run very quickly.  When both machines are utilizing the database simultaneously, machine #1 still works great (the data base is on this machine), but machine #2 drops to a painful crawl.   If both computers are utilizing the database, and machine #1 exits the database, machine #2 continues to work painfully slow, even though it has become the sole user of the database at that time.   Setting are: staticdb on, fastfk off, scratch off, multi on, ansi off, feedback on, rules off, rowlocks are used, column verify, and precedence on, sort menu on.   My thought is that at a transfer rate of 100 mbps between the machines, a slow response by machine #2 should never happen—no matter what.      Any ideas as to what is going on, and what can be done to rectify the situation?   Lee   Bailey & Associates E-Mail: [EMAIL PROTECTED] Phone: 772-597-0040 Fax: 772-597-0043