Hi everyone!
I’m trying to import the planet file for map-matching using car, bike and foot vehicles. I’ve been trying for a while with different droplets in DigitalOcean but with no success.
I’m using the following configuration:
datareader.file: planet.osm.pbf
graph.bytes_for_flags: 8
graph.elevation.cache_dir: srtmprovider
graph.elevation.dataaccess: RAM_STORE
graph.elevation.provider: multi
graph.flag_encoders: “car,bike,foot”
graph.location: graph-cache
Last attempt was with bigger memory-optimized droplet in DigitalOcean: 256GB RAM, 32CPUs and 800GB SSD. After more or less two hours, kernel kills the process with a message like this:
kernel: [70667120.897649] Out of memory: Kill process 29957 (java) score 366 or sacrifice child
The configuration used for launching the Java process is : “JAVA_OPTS=”-server -Xconcurrentio -Xmx250g -Xms250g -XX:+UseG1GC -XX:MetaspaceSize=100M""
Could it be that the configuration is not correct? Does anyone know what the problem could be?