def checkpointAndCheckStatus(checkpoint_file, apache_logs, config):
"""Checkpoint apache_logs into file and return true if still is a master."""
if not liblog.makeValid(checkpoint_file, apache_logs):
logging.error('Error writing checkpoint %s' % checkpoint_file)
return 0
if not isMaster(config):
logging.error('I am not a master. Terminating')
return 0
return 1
def checkpointAndCheckStatus(checkpoint_file, apache_logs, config):
"""Checkpoint apache_logs into file and return true if still is a master."""
if not liblog.makeValid(checkpoint_file, apache_logs):
logging.error('Error writing checkpoint %s' % checkpoint_file)
return 0
if not isMaster(config):
logging.error('I am not a master. Terminating')
return 0
return 1
def CreateLogReport(config, date_str, logs, main_google3_dir,
withResults, topCount, diagnosticTerms,
html_file, valid_file,
new_html_file, new_valid_file):
"""This method generate an aggregate report on search queries over a
period of days."""
logging.info('Creating log report for %s' % date_str)
# see if the report is already valid
if (liblog.checkValid(html_file, valid_file, logs) and
gfile.Exists(html_file)):
logging.info('%s is already valid' % html_file)
return liblog.STILL_VALID
# build the list of args
args = [date_str, new_html_file, withResults, topCount,
diagnosticTerms]
args.extend(map(lambda(x): x.file, logs))
arg_str = string.join( map(commands.mkarg, args) )
stats_cmd = ('cd %s/enterprise/legacy/analyzelogs/scripts; '
'./enterprise_stats.py %s' % (main_google3_dir, arg_str))
(status, output) = liblog.DoCommand(stats_cmd)
if status != 0:
logging.error('Error running enterprise_stats: %s' % output)
return liblog.FAILURE
# make valid file
if not liblog.makeValid(new_valid_file, logs):
logging.error('Error making valid file %s' % new_html_file)
return liblog.FAILURE
logging.info('Done log_report for %s' % new_html_file)
return liblog.SUCCESS
def CreateLogReport(config, date_str, logs, main_google3_dir, withResults,
topCount, diagnosticTerms, html_file, valid_file,
new_html_file, new_valid_file):
"""This method generate an aggregate report on search queries over a
period of days."""
logging.info('Creating log report for %s' % date_str)
# see if the report is already valid
if (liblog.checkValid(html_file, valid_file, logs)
and gfile.Exists(html_file)):
logging.info('%s is already valid' % html_file)
return liblog.STILL_VALID
# build the list of args
args = [date_str, new_html_file, withResults, topCount, diagnosticTerms]
args.extend(map(lambda (x): x.file, logs))
arg_str = string.join(map(commands.mkarg, args))
stats_cmd = ('cd %s/enterprise/legacy/analyzelogs/scripts; '
'./enterprise_stats.py %s' % (main_google3_dir, arg_str))
(status, output) = liblog.DoCommand(stats_cmd)
if status != 0:
logging.error('Error running enterprise_stats: %s' % output)
return liblog.FAILURE
# make valid file
if not liblog.makeValid(new_valid_file, logs):
logging.error('Error making valid file %s' % new_html_file)
return liblog.FAILURE
logging.info('Done log_report for %s' % new_html_file)
return liblog.SUCCESS
def main(argv):
argc = len(argv)
if argc < 6:
sys.exit(__doc__)
config = entconfig.EntConfig(argv[0])
if not config.Load():
sys.exit(__doc__)
pywrapbase.InitGoogleScript('', ['foo',
'--gfs_aliases=%s' % config.var("GFS_ALIASES"),
'--bnsresolver_use_svelte=false',
'--logtostderr'], 0)
gfile.Init()
client = argv[1]
date_arg = argv[2]
html_file = argv[3]
valid_file = argv[4]
new_valid_file = argv[5]
# extract tag and date_range from command line args
date_fields = string.split(date_arg, '_')
date_range = liblog.ParseDateRange(date_fields[0], date_fields[1:])
if not date_range:
sys.exit(__doc__)
first_date, last_date, printable_date, file_date = date_range
if last_date.as_int() < first_date.as_int():
sys.exit(__doc__)
gws_log_dir = liblog.get_gws_log_dir(config)
click_dir = liblog.get_click_dir(config)
collect_dir = liblog.get_collect_dir(config)
apache_dir = liblog.get_apache_dir(config)
directory_map_file = liblog.get_directory_map_file(config)
# we need to collect logs first from all gws nodes and preprocess
# logs first to make sure logs are up to date.
all_machines = config.var('MACHINES')
collect_logs.CollectLogs(all_machines, gws_log_dir, collect_dir)
preprocess_logs.PartitionLogs(config)
# make a vector of Log objects for all apache_logs and click_logs matching
# the given date range and client.
apache_logs = liblog.FindClientLogFiles(apache_dir, directory_map_file,
client, first_date, last_date)
click_logs = liblog.FindClientLogFiles(click_dir, directory_map_file,
client, first_date, last_date)
# If we have valid file and report file, we check to see if the data in
# apache_dir has been changed and if the report is still valid.
if (gfile.Exists(html_file) and gfile.Exists(valid_file) and
liblog.checkValid(html_file, valid_file, apache_logs)):
logging.info('%s still valid.' % html_file)
sys.exit(liblog.STILL_VALID)
# if there is no valid report, we create a new one
DumpApacheAndClickLogs(apache_logs, click_logs)
if not liblog.makeValid(new_valid_file, apache_logs):
logging.error('Error validating %s' % html_file)
sys.exit(liblog.FAILURE)
logging.info('done apache_log, new_valid_file: %s' % new_valid_file)
sys.exit(liblog.SUCCESS)
def main(argv):
argc = len(argv)
if argc < 6:
sys.exit(__doc__)
config = entconfig.EntConfig(argv[0])
if not config.Load():
sys.exit(__doc__)
pywrapbase.InitGoogleScript('', [
'foo',
'--gfs_aliases=%s' % config.var("GFS_ALIASES"),
'--bnsresolver_use_svelte=false', '--logtostderr'
], 0)
gfile.Init()
client = argv[1]
date_arg = argv[2]
html_file = argv[3]
valid_file = argv[4]
new_valid_file = argv[5]
# extract tag and date_range from command line args
date_fields = string.split(date_arg, '_')
date_range = liblog.ParseDateRange(date_fields[0], date_fields[1:])
if not date_range:
sys.exit(__doc__)
first_date, last_date, printable_date, file_date = date_range
if last_date.as_int() < first_date.as_int():
sys.exit(__doc__)
gws_log_dir = liblog.get_gws_log_dir(config)
click_dir = liblog.get_click_dir(config)
collect_dir = liblog.get_collect_dir(config)
apache_dir = liblog.get_apache_dir(config)
directory_map_file = liblog.get_directory_map_file(config)
# we need to collect logs first from all gws nodes and preprocess
# logs first to make sure logs are up to date.
all_machines = config.var('MACHINES')
collect_logs.CollectLogs(all_machines, gws_log_dir, collect_dir)
preprocess_logs.PartitionLogs(config)
# make a vector of Log objects for all apache_logs and click_logs matching
# the given date range and client.
apache_logs = liblog.FindClientLogFiles(apache_dir, directory_map_file,
client, first_date, last_date)
click_logs = liblog.FindClientLogFiles(click_dir, directory_map_file,
client, first_date, last_date)
# If we have valid file and report file, we check to see if the data in
# apache_dir has been changed and if the report is still valid.
if (gfile.Exists(html_file) and gfile.Exists(valid_file)
and liblog.checkValid(html_file, valid_file, apache_logs)):
logging.info('%s still valid.' % html_file)
sys.exit(liblog.STILL_VALID)
# if there is no valid report, we create a new one
DumpApacheAndClickLogs(apache_logs, click_logs)
if not liblog.makeValid(new_valid_file, apache_logs):
logging.error('Error validating %s' % html_file)
sys.exit(liblog.FAILURE)
logging.info('done apache_log, new_valid_file: %s' % new_valid_file)
sys.exit(liblog.SUCCESS)
