0152538cb630db0a08d41756a12f61091ff4fcc7 max Wed Apr 18 11:53:21 2018 -0700 little apacheLogParse update diff --git src/utils/apacheLogParse src/utils/apacheLogParse index 8260e17..09a6f74 100755 --- src/utils/apacheLogParse +++ src/utils/apacheLogParse @@ -9,31 +9,31 @@ from os import listdir import optparse, logging, sys, string from itertools import imap from operator import itemgetter import heapq #TEMPDIR = "/dev/shm" # filename to delete on exit removeTmpName = None # where do we store the raw apache logfiles baseLogDir = "/hive/data/inside/wwwstats/RR" # years to analyze -years = ["2009", "2010", "2011", "2012", "2013", "2014", "2015", "2016", "2017"] +years = ["2009", "2010", "2011", "2012", "2013", "2014", "2015", "2016", "2017", "2018", "2019", "2020"] # list of servers to analyze # change these for debugging or if need to focus on single servers servNames = ["hgw1", "hgw2", "hgw3", "hgw4", "hgw5", "hgw6", "hgw7", "hgw8"] # directory for cluster job output jobOutDir = "/hive/data/inside/wwwstats/apacheLogParse" # directory to store csv files for C3 plot htmlDir = "/cluster/home/max/public_html/logParse/plot" # a struct for a cleaned apache log line apacheFields = ['ip', 'time', 'status', "reqType", 'speed', "filePath", "cgiParams", "agent", "referer"] ApacheLine = namedtuple("ApacheLine", apacheFields)