I have used successfully new splitter when no --split_file is given, but if I use an existing areas.list splitter looks for cached files and they don't exist; see below: java -Xmx2500M -jar splitter.jar --cache=/160GB/cache_europe --split-file=areas_5.list /160GB/europe.osm Time started: Mon Aug 24 16:24:22 CEST 2009 cache = /160GB/cache_europe split-file = areas_5.list Areas read from file 63240001 (35.947265625,-9.66796875) to (41.044921875,-4.04296875) 63240002 (41.044921875,-9.580078125) to (43.9453125,-4.04296875) 63240003 (36.474609375,-4.04296875) to (39.7265625,1.845703125) 63240004 (39.7265625,-4.04296875) to (43.76953125,-1.142578125) 63240005 (39.7265625,-1.142578125) to (43.41796875,1.845703125) 63240006 (38.759765625,1.845703125) to (42.978515625,4.833984375) Writing out split osm files Mon Aug 24 16:24:22 CEST 2009 Processing 6 areas in a single pass Starting pass 1, processing 6 areas (63240001 to 63240006) Loading and processing nodes Error opening or reading file java.io.FileNotFoundException: /160GB/cache_europe/nodes.bin (No such file or directory) java.io.FileNotFoundException: /160GB/cache_europe/nodes.bin (No such file or directory) at java.io.FileInputStream.open(Native Method) at java.io.FileInputStream.<init>(Unknown Source) at uk.me.parabola.splitter.disk.NodeStoreReader.<init>(NodeStoreReader.java:19) at uk.me.parabola.splitter.BinaryMapLoader.processNodes(BinaryMapLoader.java:92) at uk.me.parabola.splitter.BinaryMapLoader.load(BinaryMapLoader.java:82) at uk.me.parabola.splitter.Main.processMap(Main.java:281) at uk.me.parabola.splitter.Main.writeAreas(Main.java:271) at uk.me.parabola.splitter.Main.split(Main.java:118) at uk.me.parabola.splitter.Main.main(Main.java:92) Time finished: Mon Aug 24 16:24:23 CEST 2009 Total time taken: 0s Regards Carlos