Parce que «Ça peut toujours servir»™ et que de toutes façons il y en a déjà des bouts qui sont dans le dépôt et que c'est chiant de git add -f.
Et puis bon, ça fait que 3Mo
This commit is contained in:
parent
29f50c2ed9
commit
3bde363deb
299 changed files with 17466 additions and 0 deletions
183
archive/utils/syslog-summary
Executable file
183
archive/utils/syslog-summary
Executable file
|
@ -0,0 +1,183 @@
|
|||
#!/usr/bin/python
|
||||
|
||||
"""Summarize the contents of a syslog log file.
|
||||
|
||||
The syslog(3) service writes system log messages in a certain format:
|
||||
|
||||
Jan 17 19:21:50 zeus kernel: klogd 1.3-3, log source = /proc/kmsg started.
|
||||
|
||||
This program summarizes the contents of such a file, by displaying each
|
||||
unique (except for the time) line once, and also the number of times such
|
||||
a line occurs in the input. The lines are displayed in the order they occur
|
||||
in the input.
|
||||
|
||||
Lars Wirzenius <liw@iki.fi>"""
|
||||
|
||||
IGNORE_FILENAME = "/etc/syslog-summary/ignore"
|
||||
STATE_FILENAME = None
|
||||
REPEAT = 0
|
||||
QUIET = 0
|
||||
|
||||
import sys, re, getopt, string, md5
|
||||
|
||||
datepats = [
|
||||
re.compile(r"^(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) [ 0-9][0-9] [ 0-9][0-9]:[0-9][0-9]:[0-9][0-9] "),
|
||||
re.compile(r"^(Mon|Tue|Wed|Thu|Fri|Sat|Sun) (Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) [ 0-9][0-9][0-9][0-9]:[0-9][0-9] "),
|
||||
re.compile(r"^(Mon|Tue|Wed|Thu|Fri|Sat|Sun) (Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) [ 0-9][0-9][0-9][0-9]:[0-9][0-9]:[0-9][0-9] "),
|
||||
]
|
||||
pidpat = re.compile(r"^([^ ]* [^ ]*)\[[0-9][0-9]*\]: ")
|
||||
repeatpat = re.compile(r"^[^ ]* last message repeated (\d+) times$")
|
||||
|
||||
ignore_pats = []
|
||||
|
||||
def read_patterns(filename):
|
||||
pats = []
|
||||
try:
|
||||
f = open(filename, "r")
|
||||
except IOError:
|
||||
return []
|
||||
for line in f.xreadlines():
|
||||
if line[-1:] == "\n":
|
||||
line = line[:-1]
|
||||
pats.append(re.compile(line))
|
||||
f.close()
|
||||
return pats
|
||||
|
||||
def read_states(filename):
|
||||
states = {}
|
||||
if not filename:
|
||||
return states
|
||||
try:
|
||||
f = open(filename, "r")
|
||||
except IOError:
|
||||
return states
|
||||
for line in f.xreadlines():
|
||||
fields = string.split(line)
|
||||
states[fields[0]] = (string.atoi(fields[1]), fields[2])
|
||||
f.close()
|
||||
return states
|
||||
|
||||
def save_states(filename, states):
|
||||
if not filename:
|
||||
return
|
||||
try:
|
||||
f = open(filename, "w")
|
||||
except IOError:
|
||||
return
|
||||
for filename in states.keys():
|
||||
value = states[filename]
|
||||
f.write("%s %d %s\n" % (filename, value[0], value[1]))
|
||||
f.close()
|
||||
|
||||
def should_be_ignored(line):
|
||||
for pat in ignore_pats:
|
||||
if pat.search(line):
|
||||
return 1
|
||||
return 0
|
||||
|
||||
def printable_md5(str):
|
||||
chars = []
|
||||
for char in str:
|
||||
chars.append("%02x" % (ord(char)))
|
||||
return string.join(chars, "")
|
||||
|
||||
def split_date(line):
|
||||
for pat in datepats:
|
||||
m = pat.match(line)
|
||||
if m:
|
||||
return line[:m.end()], line[m.end():]
|
||||
print "line has bad date", "<" + string.rstrip(line) + ">"
|
||||
return None, line
|
||||
|
||||
def summarize(filename, states):
|
||||
counts = {}
|
||||
order = []
|
||||
ignored_count = 0
|
||||
if not QUIET:
|
||||
print "Summarizing %s" % filename
|
||||
file = open(filename, "r")
|
||||
linecount = 0
|
||||
|
||||
md5obj = md5.new()
|
||||
if states.has_key(filename):
|
||||
oldlines, oldmd5 = states[filename]
|
||||
for i in xrange(oldlines):
|
||||
line = file.readline()
|
||||
md5obj.update(line)
|
||||
if printable_md5(md5obj.digest()) != oldmd5:
|
||||
file.seek(0, 0)
|
||||
md5obj = md5.new()
|
||||
else:
|
||||
linecount = oldlines
|
||||
if not QUIET:
|
||||
print "%8d Lines skipped (already processed)" % linecount
|
||||
|
||||
line = file.readline()
|
||||
previous = None
|
||||
while line:
|
||||
md5obj.update(line)
|
||||
linecount = linecount + 1
|
||||
date, rest = split_date(line)
|
||||
if date:
|
||||
found = pidpat.search(rest)
|
||||
if found:
|
||||
rest = found.group(1)+": "+rest[found.end():]
|
||||
|
||||
count = 1
|
||||
repeated = None
|
||||
if REPEAT:
|
||||
repeated=repeatpat.search(rest)
|
||||
if repeated and previous:
|
||||
count = int(repeated.group(1))
|
||||
rest = previous
|
||||
|
||||
if should_be_ignored(rest):
|
||||
ignored_count = ignored_count + count
|
||||
else:
|
||||
if counts.has_key(rest):
|
||||
counts[rest] = counts[rest] + count
|
||||
else:
|
||||
assert count==1
|
||||
counts[rest] = count
|
||||
order.append(rest)
|
||||
|
||||
if not repeated:
|
||||
previous = rest
|
||||
line = file.readline()
|
||||
file.close()
|
||||
md5new = printable_md5(md5obj.digest())
|
||||
states[filename] = (linecount, md5new)
|
||||
if QUIET and order:
|
||||
print "Summarizing %s" % filename
|
||||
if not QUIET or order:
|
||||
print "%8d Patterns to ignore" % len(ignore_pats)
|
||||
print "%8d Ignored lines" % ignored_count
|
||||
for rest in order:
|
||||
print "%8d %s" % (counts[rest], rest),
|
||||
if not QUIET or order:
|
||||
print
|
||||
|
||||
def main():
|
||||
global ignore_pats, IGNORE_FILENAME, STATE_FILENAME, REPEAT, QUIET
|
||||
|
||||
opts, args = getopt.getopt(sys.argv[1:], "i:qs:r", [
|
||||
"ignore=", "quiet", "state=", "repeat" ])
|
||||
|
||||
for opt, optarg in opts:
|
||||
if opt == "-i" or opt == "--ignore":
|
||||
IGNORE_FILENAME = optarg
|
||||
elif opt == "-s" or opt == "--state":
|
||||
STATE_FILENAME = optarg
|
||||
elif opt == "-r" or opt == "--repeat":
|
||||
REPEAT = 1
|
||||
elif opt == "-q" or opt == "--quiet":
|
||||
QUIET = 1
|
||||
|
||||
ignore_pats = read_patterns(IGNORE_FILENAME)
|
||||
states = read_states(STATE_FILENAME)
|
||||
for filename in args:
|
||||
summarize(filename, states)
|
||||
save_states(STATE_FILENAME, states)
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
Loading…
Add table
Add a link
Reference in a new issue