Ip_checksum:int,ip_src:chararray,ip_dst:chararray,tcp_sport:int, Ts,ip_version:int,ip_header_length:int,ip_tos:int,ip_total_length:int, Hadoop fs -cp file:///home/cloudera/Downloads/packetpigTest/packetpig/lib/*.jar hdfs:///packetpig/ Here are a handful of instructions from my notes: Two issues with PacketPig (or me :) ) slowed me down: (1) I had issues running on the most recent versions of hadoop and pig, running on hadoop 1.0.3 and pig 0.10 worked fine, (2) initially I could only get PacketPig to load the first IP record from the pcap. You will probably want to split your pcap file into smaller pcap files if you have a hadoop cluster to run on. Pig is a language on top of hadoop, these explain it better than I can: and (programming_tool). I looked at it for analysis of some pcaps of MGCP packets, but that project never got started. Hopefully this is still useful to you, and if not maybe it is for someone else. Sorry for the delay, I saw this post but didn't want to recommend a solution without having run it first.
0 kommentar(er)
