Hi, you have to use empirical method! 1. Try optimize serverconfig.ini parameters. We use this parameters with a repositories with many maps and layers LibraryCacheSize = 32 SessionCacheSize = 2 DBPageSize = 32 DBXMLPageSize = 32 LibraryLogBufferSize = 50 SessionLogBufferSize = 20 DBMaxTransactions = 3000 SessionDBPageSize = 2 SessionDBXMLPageSize = 0.5 DBTimeout = 0.2 DBMaxLockers = 3000
2. Forget SHP and use SDF for best performance and stability 3. Try to understand if pink map occurs with only SHP layers, or with only Postgis Layers. Perhaps the problem is caused by a specific data source 4. Try to avoid FDO filters on layers with many object, use View on database side 5. We got pink screen one year ago caused by a malformed polygon geometry (in Oracle). We found the problem because pink map happened always on the same map extent!! 6. Try to decrease pink map occurrences, because resolve definitely is vary hard in my opinion without a good log file created by application 7. Enable MapGuide Performance Log to get important parameters about memory consumption and so on. Graphic in Excel on Performance Log are very cool I hope this helps johnny -- View this message in context: http://osgeo-org.1560.x6.nabble.com/Mapguide-server-3-0-crashes-often-and-got-pink-screen-in-the-fusion-viewer-tp5255669p5256013.html Sent from the MapGuide Users mailing list archive at Nabble.com. _______________________________________________ mapguide-users mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/mapguide-users
