Merge "Report Tool: Package and add plugins/correlator"

This commit is contained in:
Zuul 2022-12-13 18:18:52 +00:00 committed by Gerrit Code Review
commit 882eab3cd4
22 changed files with 2045 additions and 911 deletions

View File

@ -2941,6 +2941,56 @@ collect_subclouds()
fi
}
############################################################################
#
# Name : get_report_tool
#
# Purpose : Fetch report tool from current host
#
# Parameters: $1 - local path destination
#
############################################################################
function get_report_tool()
{
local local_dest=${1}
mkdir -p ${local_dest}
cp -r /usr/local/bin/report/tool ${local_dest}
local rc=${?}
if [ ${rc} -ne ${PASS} ] ; then
report_error "failed to get report tool from /usr/local/bin" ${rc}
else
ilog "copied report tool from host"
fi
}
############################################################################
#
# Name : get_report_plugins
#
# Purpose : Fetch plugins for report tool from current host
#
# Parameters: $1 - local path destination
#
############################################################################
function get_report_plugins()
{
local local_dest=${1}
mkdir -p ${local_dest}
cp -r /etc/collect/plugins ${local_dest}
local rc=${?}
if [ ${rc} -ne ${PASS} ] ; then
report_error "failed to get report plugins from /etc/collect" ${rc}
else
ilog "copied plugins for report tool from host"
fi
}
############################################################################
#
# Handle subcloud and system hosts batched collect
@ -3031,6 +3081,17 @@ echo -n "creating ${COLLECT_TYPE} tarball ${TARBALL_NAME} ... "
remove_file_local ${COLLECT_ERROR_LOG}
remove_file_local ${HOST_COLLECT_ERROR_LOG}
get_report_tool ${COLLECT_DIR}/report
get_report_plugins ${COLLECT_DIR}/report
cd ${COLLECT_DIR}
tar -czf report_tool.tgz report
rc=${?}
if [ ${rc} -ne ${PASS} ] ; then
report_error "failed to tar report tool" ${rc}
else
rm -r report
fi
/usr/bin/expect << EOF
log_user ${USER_LOG_MODE}

View File

@ -0,0 +1,81 @@
Refer to report.py file header for a description of the tool
Example:
Consider the following collect bundle structure
SELECT_NODES_20220527.193605
├── controller-0_20220527.193605
│  ├── etc
│  ├── root
│  └── var
├── controller-1_20220527.193605
│  ├── etc
│  ├── root
│  └── var
├── report
├── plugins (where the plugin files will be placed)
│  ├── alarm
│  ├── substring
│  └── ...
├── tool (where the tool will be placed)
└── output (where the output files will be placed)
> cat plugins/alarm
algorithm=alarm
alarm_exclude=400., 800.
entity_exclude=subsystem=vim
> cat plugins/substring
algorithm=substring
files=var/log/mtcAgent.log, var/log/sm.log
hosts=controllers
substring=operation failed
substring=Failed to send message
> report/tool/report.py --start 20220501 --end 20220530
Running the command above will populate the report folder with output files.
The tool also provides default values, more details are in 'report.py -h'.
The substring algorithm creates an output file for every host of the
specified host type. The files will contain log events within the
provided date range containing the substring 'operation failed' and 'Failed
to send message'.
The alarm algorithm creates two output file: 'log' and 'alarm'
'log' contains customer log messages created within the provided date range,
and 'alarm' contains system alarms created within the provided date range, as
long as the alarm ids and entity ids are not included in the alarm plugin file.
For more detailed information about an algorithm use 'report.py <algorithm> -h'.
Here is the report directory after running the above command
report
├── output
│ └── SELECT_NODES_20220527.193605 (collect bundle that the report tool was run on)
│ ├── plugins (output files for plugins)
│ │ ├── alarm
│ │ └── ...
│ ├── correlator_failures
│ ├── correlator_events
│ ├── correlator_state_changes
│ ├── report.log (log file for report tool)
│ └── untar.log (log file for untarring collect bundle and host tar files)
├── plugins (where the plugins files are)
└── tool (where the report tool is)
The report tool also allows users to point it at any collect bundle and
have it automatically extract the tarball and tar files for each host
before running.
> report/tool/report.py -d CGTS-19143
Users may specify if they want the correlator to only find events
and state changes for a specific host.
> report/tool/report.py --hostname controller-0

View File

@ -9,8 +9,12 @@
# Algorithm string constants
ALARM = "alarm"
AUDIT = "audit"
PROCESS_FAILURE = "process_failure"
PUPPET = "puppet"
DAEMON_FAILURES = "daemon_failures"
HEARTBEAT_LOSS = "heartbeat_loss"
MAINTENANCE_ERR = "maintenance_errors"
PROCESS_FAILURES = "process_failures"
PUPPET_ERRORS = "puppet_errors"
STATE_CHANGES = "state_changes"
SUBSTRING = "substring"
SWACT = "swact"
SWACT_ACTIVITY = "swact_activity"
SYSTEM_INFO = "system_info"

View File

@ -0,0 +1,583 @@
########################################################################
#
# Copyright (c) 2022 Wind River Systems, Inc.
#
# SPDX-License-Identifier: Apache-2.0
#
########################################################################
#
# This file contains the Correlator class.
# The Correlator class contains algorithms that search for failures.
#
# The Correlator class reads through all the output files created by
# the plugins and detects failures. A summary of the failures and their
# causes are printed to standard output and an output file is created
# in the report directory.
#
# TODO: Modularize code and separate methods into their own files
#
########################################################################
from datetime import datetime
from datetime import timedelta
import logging
import os
import re
logger = logging.getLogger(__name__)
class Correlator:
def __init__(self, plugin_output_dir):
"""Constructor for the Correlator class
Parameters:
plugin_output_dir (string): Path to directory with output files
from plugins
"""
self.plugin_output_dir = plugin_output_dir
def run(self, hostname):
"""Searches through the output files created by the plugins for
failures and determines their causes, as well as extracts significant
events and state changes
Errors:
FileNotFoundError
"""
failures = []
try:
failures += self.uncontrolled_swact()
except FileNotFoundError as e:
logger.error(e)
try:
failures += self.mtc_errors()
except FileNotFoundError as e:
logger.error(e)
events = []
try:
events += self.get_events(hostname)
except FileNotFoundError as e:
logger.error(e)
alarms = []
try:
alarms += self.get_alarms(hostname)
except FileNotFoundError as e:
logger.error(e)
state_changes = []
try:
state_changes += self.get_state_changes(hostname)
except FileNotFoundError as e:
logger.error(e)
return (sorted(failures), sorted(events), sorted(alarms),
sorted(state_changes))
def uncontrolled_swact(self):
"""Searches through the output file created by the swact activity
plugin for uncontrolled swacts and determines their causes through
other indicators, like the log "Neighbour [..] is now in the down"
Errors:
FileNotFoundError
"""
data = []
# Variables to keep track of indicators for failure causes
start_time = end_time = svc_failed = None
ctrlr_down = None # Active controller that went down, causing swact
ctrlr_svc_fail = None # Active controller where service failed
ctrlr_link_down = None # Orig. active controller when link went down
hb_loss = active_failed = go_active_failed = link_down = False
# Open output file from swact activity plugin and read it
file_path = os.path.join(self.plugin_output_dir, "swact_activity")
with open(file_path, "r") as swact_activity:
for line in swact_activity:
if "Uncontrolled swact" in line and not start_time:
start_time = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
if ("Host from active to failed, Peer from standby to "
"active" in line):
link_down = True
ctrlr_link_down = re.findall(
r"\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d{3} (.+) "
"sm:", line)[0]
elif (re.search("Neighbor (.+) is now in the down", line)
and start_time and not ctrlr_down):
ctrlr_down = re.findall(
r"Neighbor \((.+)\) received event", line)[0]
elif (re.search("Service (.+) is failed and has reached max "
"failures", line) and not svc_failed):
svc_failed = re.findall(
r"Service \((.+)\) is failed", line)[0]
ctrlr_svc_fail = re.findall(
r"\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d{3} (.+) sm:",
line)[0]
elif (svc_failed and re.search(
"active-failed\\s+\\| disabling-failed\\s+\\| "
+ svc_failed, line)):
if re.search(r"\| go-active-failed\s+\|", line):
go_active_failed = True
else:
active_failed = True
elif "Swact update" in line and start_time and not end_time:
end_time = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
if ctrlr_down:
try:
hb_loss = self.search_hb_loss(
start_time, end_time, ctrlr_down)
except FileNotFoundError as e:
logger.error(e)
start_time = start_time.strftime("%Y-%m-%dT%H:%M:%S")
end_time = end_time.strftime("%Y-%m-%dT%H:%M:%S")
if link_down:
data.append(start_time + " to " + end_time
+ " Uncontrolled swact, refer to SM logs "
"for in-depth analysis, original active "
"controller: " + ctrlr_link_down + "\n")
elif ctrlr_down:
if hb_loss:
data.append(start_time + " to " + end_time
+ " Uncontrolled swact due to "
"spontaneous reset of active "
"controller " + ctrlr_down + "\n")
else:
data.append(start_time + " to " + end_time
+ " Uncontrolled swact likely due to "
"spontaneous reset of active "
"controller " + ctrlr_down + "\n")
elif svc_failed:
if active_failed and go_active_failed:
data.append(start_time + " to " + end_time
+ " Uncontrolled swact due to service "
"failure (" + svc_failed + ") twice "
"in 2 minutes was unsuccessful so "
"\"bounced back\" to original active "
"controller " + ctrlr_svc_fail + "\n")
elif active_failed:
data.append(start_time + " to " + end_time
+ " Uncontrolled swact due to service "
"failure (" + svc_failed + ") twice "
"in 2 minutes on active controller "
+ ctrlr_svc_fail + "\n")
else:
data.append(start_time + " to " + end_time
+ " Uncontrolled swact likely due to "
"service failure (" + svc_failed
+ ") twice in 2 minutes on active "
"controller " + ctrlr_svc_fail + "\n")
start_time = end_time = svc_failed = None
ctrlr_down = ctrlr_svc_fail = ctrlr_link_down = None
hb_loss = active_failed = go_active_failed = False
link_down = False
return data
def mtc_errors(self):
"""Searches through the output file created by the maintenance errors
plugin for failures and determines their causes through other
indicators, like the log "Loss Of Communication for 5 seconds"
Errors:
FileNotFoundError
"""
data = []
# Variables to keep track of indicators for failure causes
goenable_start = goenable_end = goenable_host = None
goenable_tst_f = config_tst_f = None # Tests failed
config_start = config_end = config_host = puppet_error = None
hb_loss_start = hb_loss_end = hb_loss_host = None
daemon_fail = comm_loss = auto_recov_dis = False
# Open output file from maintenance errors plugin and read it
file_path = os.path.join(self.plugin_output_dir, "maintenance_errors")
with open(file_path, "r") as mtc:
for line in mtc:
if "auto recovery disabled" in line and not auto_recov_dis:
# Check if previous failure recorded was go-enable,
# configuration or heartbeat failure
if (data and
re.search(r"Go-enable|[cC]onfiguration|Heartbeat",
data[-1])):
host = re.findall(r"failure on ([^\s]+)", data[-1])
# Check if host in auto recovery disabled mode is same
# as host with previous failure
if (host and re.search(
host[0] + " auto recovery disabled", line)):
old = data[-1].split("due", 1)
if len(old) == 1:
data[-1] = (data[-1][:-1]
+ " (auto recovery disabled)\n")
else:
data[-1] = (old[0]
+ "(auto recovery disabled) due"
+ old[1])
auto_recov_dis = True
elif "GOENABLED Failed" in line and not goenable_start:
goenable_start, auto_recov_dis = line[0:19], False
goenable_host = re.findall(
"Error : (.+) got GOENABLED Failed", line)[0]
elif ("configuration failed or incomplete" in line
and not config_start):
config_start = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
auto_recov_dis = False
config_host = re.findall(
"Error : (.+) configuration failed", line)[0]
elif "Heartbeat Loss" in line:
# Check if previous failure recorded was heartbeat loss
# due to missing heartbeat messages
if ("(during recovery soak)" in line and data and
re.search("missing heartbeat messages", data[-1])):
host = re.findall(
"failure on (.+) due to", data[-1])[0]
# Check if host with hearbeat loss failure is the same
# as host with previous failure
if (re.search(host + " (.+) Heartbeat Loss (.+) "
"\\(during recovery soak\\)", line)):
old = data[-1]
data[-1] = (old[0:23] + line[0:19] + old[42:-1]
+ " (recovery over disabled due to "
"heartbeat soak failure)\n")
else:
hb_loss_start = line[0:19]
comm_loss = auto_recov_dis = False
hb_loss_host = re.findall("Error : (.+) [CM]", line)[0]
# Check if previous failure recorded was heartbeat loss due to
# missing heartbeat messages
elif ("regained MTCALIVE from host that has rebooted" in line
and data and re.search(r"Heartbeat loss failure (.+) "
r"\(recovery over disabled\)",
data[-1])):
host = re.findall("failure on (.+) due to", data[-1])[0]
if re.search(host + " regained MTCALIVE", line):
old = data[-1].split("due", 1)[0]
data[-1] = (old[0:23] + line[0:19] + old[42:]
+ "due to uncontrolled reboot\n")
elif (hb_loss_start and not comm_loss and hb_loss_host and
re.search(hb_loss_host + " Loss Of Communication for 5 "
"seconds", line)):
comm_loss = True
elif re.search("mtcClient --- (.+)Error : FAILED:", line):
if goenable_start and not goenable_tst_f:
goenable_tst_f = re.findall(
r"Error : FAILED: (.+) \(\d", line)[0]
elif config_start and not config_tst_f:
config_tst_f = re.findall(
r"Error : FAILED: (.+) \(\d", line)[0]
elif (goenable_host and not goenable_end and
re.search(goenable_host + " Task: In-Test Failure, "
"threshold reached", line)):
goenable_end = line[0:19]
if goenable_tst_f:
data.append(goenable_start + " to " + goenable_end
+ " Go-enable test failure on "
+ goenable_host + " due to failing of "
+ goenable_tst_f + "\n")
else:
data.append(goenable_start + " to " + goenable_end
+ " Go-enable test failure on "
+ goenable_host + " due to unknown test "
"failing\n")
goenable_start = goenable_end = goenable_host = None
goenable_tst_f = None
elif (config_host and not config_end and
re.search(config_host + " Task: Configuration failure, "
"threshold reached", line)):
config_end = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
if (config_tst_f
!= "/etc/goenabled.d/config_goenabled_check.sh"):
try:
daemon_fail = self.search_daemon_fail(
config_start, config_end, config_host)
except FileNotFoundError as e:
logger.error(e)
if (config_tst_f ==
"/etc/goenabled.d/config_goenabled_check.sh"
or daemon_fail):
try:
puppet_error = self.search_puppet_error(
config_start, config_end)
except FileNotFoundError as e:
logger.error(e)
config_start = config_start.strftime(
"%Y-%m-%dT%H:%M:%S")
config_end = config_end.strftime("%Y-%m-%dT%H:%M:%S")
if puppet_error:
data.append(config_start + " to " + config_end
+ " Configuration failure on "
+ config_host + " due to:\n"
+ puppet_error)
else:
data.append(config_start + " to " + config_end
+ " Configuration failure on "
+ config_host
+ " due to unknown cause\n")
else:
config_start = config_start.strftime(
"%Y-%m-%dT%H:%M:%S")
config_end = config_end.strftime("%Y-%m-%dT%H:%M:%S")
data.append(config_start + " to " + config_end
+ " Possible configuration failure on "
+ config_host + "\n")
config_start = config_end = config_host = None
config_tst_f = puppet_error = None
daemon_fail = False
elif (hb_loss_start and not hb_loss_end and hb_loss_host and
re.search(hb_loss_host + " Connectivity Recovered ",
line)):
hb_loss_end = line[0:19]
data.append(hb_loss_start + " to " + hb_loss_end
+ " Heartbeat loss failure on " + hb_loss_host
+ " due to too many missing heartbeat "
"messages\n")
hb_loss_start = hb_loss_end = hb_loss_host = None
comm_loss = False
elif (hb_loss_start and comm_loss and not hb_loss_end and
hb_loss_host and re.search(
hb_loss_host + " Graceful Recovery Wait", line)):
hb_loss_end = line[0:19]
data.append(hb_loss_start + " to " + hb_loss_end
+ " Heartbeat loss failure on " + hb_loss_host
+ " due to too many missing heartbeat "
"messages (recovery over disabled)\n")
hb_loss_start = hb_loss_end = hb_loss_host = None
comm_loss = False
return data
def search_hb_loss(self, start_time, end_time, host):
"""Searches through the output file created by the heartbeat loss
plugin for "Heartbeat Loss" message from host between one minute before
start_time and end_time
Errors:
FileNotFoundError
"""
hb_loss = False
# Open output file from heartbeat loss plugin and read it
file_path = os.path.join(self.plugin_output_dir, "heartbeat_loss")
with open(file_path, "r") as heartbeat_loss:
for line in heartbeat_loss:
if (re.search("Error : " + host + " (.+) Heartbeat Loss ",
line)):
date = datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
if (date >= start_time - timedelta(minutes=1)
and date <= end_time):
hb_loss = True
break
return hb_loss
def search_daemon_fail(self, start_time, end_time, host):
"""Searches through the output file created by the daemon failures
plugin for "Failed to run the puppet manifest" message from host
between 10 seconds before start_time and end_time
Errors:
FileNotFoundError
"""
daemon_fail = False
# Open output file from daemon failures plugin and read it
file_path = os.path.join(self.plugin_output_dir, "daemon_failures")
with open(file_path, "r") as daemon_failures:
for line in daemon_failures:
if (re.search("\\d " + host
+ " (.+) Failed to run the puppet manifest",
line)):
date = datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
if (date >= start_time - timedelta(seconds=10)
and date <= end_time):
daemon_fail = True
break
return daemon_fail
def search_puppet_error(self, start_time, end_time):
"""Searches through the output file created by the puppet errors
plugin for "Error:" message between 10 seconds before start_time and
end_time and returns it
Errors:
FileNotFoundError
"""
puppet_log = None
# Open output file from puppet errors plugin and read it
file_path = os.path.join(self.plugin_output_dir, "puppet_errors")
with open(file_path, "r") as puppet_errors:
for line in puppet_errors:
if "Error: " in line:
date = datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
if (date >= start_time - timedelta(seconds=10)
and date <= end_time):
puppet_log = line
break
return puppet_log
def get_events(self, hostname):
"""Searches through the output files created by the plugins for
significant events and summarizes them, such as "force failed by SM"
Errors:
FileNotFoundError
"""
data = []
# Variables to keep track of details for events
mnfa_start, mnfa_hist = None, ""
# Open output file from maintenance errors plugin and read it
file_path = os.path.join(self.plugin_output_dir, "maintenance_errors")
with open(file_path, "r") as mtc:
for line in mtc:
if "force failed by SM" in line:
host = re.findall("Error : (.+) is being", line)[0]
if hostname == "all" or host == hostname:
data.append(line[0:19] + " " + host
+ " force failed by SM\n")
elif "Graceful Recovery Failed" in line:
host = re.findall("Info : (.+) Task:", line)[0]
if hostname == "all" or host == hostname:
data.append(line[0:19] + " " + host
+ " graceful recovery failed\n")
elif "MNFA ENTER" in line:
mnfa_start = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
elif "MNFA POOL" in line:
pool_hosts = len(line.split("MNFA POOL: ")[1].split())
if mnfa_start:
mnfa_hist += (" " + str(pool_hosts))
else:
data_len = len(data)
for n in range(0, data_len):
event = data[data_len - 1 - n]
if "Multi-node failure" in event:
temp = " " + str(pool_hosts) + ")\n"
data[data_len - 1 - n] = event[:-2] + temp
break
elif "MNFA EXIT" in line:
mnfa_duration = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
mnfa_duration -= mnfa_start
mnfa_start = mnfa_start.strftime("%Y-%m-%dT%H:%M:%S")
data.append(mnfa_start + " Multi-node failure avoidance "
+ "(duration: " + str(mnfa_duration)
+ "; history:" + mnfa_hist + ")\n")
mnfa_start, mnfa_hist = None, ""
# Open output file from swact activity plugin and read it
file_path = os.path.join(self.plugin_output_dir, "swact_activity")
with open(file_path, "r") as swact_activity:
for line in swact_activity:
if (re.search("Service (.+) is failed and has reached max "
"failures", line)):
host = re.findall(
r"\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d{3} (.+) sm:",
line)[0]
svc_failed = re.findall(
r"Service \((.+)\) is failed", line)[0]
if hostname == "all" or host == hostname:
data.append(line[0:19] + " " + host
+ " service failure (" + svc_failed
+ ")\n")
return data
def get_alarms(self, hostname):
"""Searches through the 'alarm' output file created by the alarm plugin
and summarizes which alarms were found as well as the number of times
they were set and cleared
Errors:
FileNotFoundError
"""
data = []
# Open 'alarm' output file from alarm plugin and read it
file_path = os.path.join(self.plugin_output_dir, "alarm")
with open(file_path, "r") as alarm:
extract = False
for line in alarm:
if re.search(" \\d", line) and extract:
if line.split()[2] == "set":
data[-1]["set"] += 1
else:
data[-1]["clear"] += 1
elif hostname == "all" or hostname in line:
extract = True
alarm = {
"name": line[:-1],
"set": 0,
"clear": 0,
}
data.append(alarm)
else:
extract = False
temp = []
for entry in data:
temp.append(entry["name"] + " - set: " + str(entry["set"])
+ ", clear: " + str(entry["clear"]) + "\n")
data = temp
return data
def get_state_changes(self, hostname):
"""Searches through the output files created by the state changes
plugin and summarizes the changes of state of the hosts, such as
"is ENABLED"
Errors:
FileNotFoundError
"""
data = []
# Open output file from state changes plugin and read it
file_path = os.path.join(self.plugin_output_dir, "state_changes")
with open(file_path, "r") as state_changes:
for line in state_changes:
if "is ENABLED" in line:
host = re.findall("Info : (.+) is ENABLED", line)[0]
state = re.findall("is (.+)\n", line)[0].lower()
if hostname == "all" or hostname in host:
data.append(line[0:19] + " " + host + " " + state
+ "\n")
elif "locked-disabled" in line:
host = re.findall(
"Info : (.+) u?n?locked-disabled", line)[0]
if hostname == "all" or host == hostname:
data.append(line[0:19] + " " + host + " disabled\n")
return data

View File

@ -0,0 +1,869 @@
########################################################################
#
# Copyright (c) 2022 Wind River Systems, Inc.
#
# SPDX-License-Identifier: Apache-2.0
#
########################################################################
#
# This file contains the ExecutionEngine class.
# The ExecutionEngine class contains all the available algorithms.
#
# The ExecutionEngine class runs plugins and gathers relevant logs and
# information, creating output files in the report directory.
#
# TODO: Modularize code and separate plugin algorithms into their own
# files
#
########################################################################
from datetime import datetime
import gzip
import logging
import os
import re
import shutil
import subprocess
import tarfile
import algorithms
from correlator import Correlator
logger = logging.getLogger(__name__)
class ExecutionEngine:
def __init__(self, opts, output_directory):
"""Constructor for the ExecutionEngine class
Parameters:
opts (dictionary): Options from command line
"""
self.opts = opts
self.hosts = {"controllers": {}, "workers": {}, "storages": {}}
self.active_controller_directory = None
# Uncompresses host tar files if not already done
with open(os.path.join(output_directory, "untar.log"), "a") as logfile:
for obj in (os.scandir(self.opts.directory)):
info = os.path.splitext(obj.name)
if (obj.is_file() and obj.name != "report_tool.tgz" and
tarfile.is_tarfile(obj.path) and not
os.path.isdir(os.path.join(self.opts.directory,
info[0]))):
try:
subprocess.run(["tar", "xzfC", obj.path,
self.opts.directory],
stderr=logfile, check=True)
subprocess.run(["echo", "uncompressed", obj.name],
check=True)
except subprocess.CalledProcessError as e:
logger.error(e)
for folder in (f.path for f in os.scandir(self.opts.directory)):
database_path = os.path.join(folder, "var", "extra", "database")
host_info_path = os.path.join(folder, "var", "extra", "host.info")
if os.path.isdir(database_path) and os.listdir(database_path):
self.active_controller_directory = folder
if os.path.exists(host_info_path):
hostname, subfunction = self._extract_subfunction(
host_info_path)
if "controller" in subfunction:
self.hosts["controllers"][hostname] = folder
elif "worker" in subfunction:
self.hosts["workers"][hostname] = folder
elif "storage" in subfunction:
self.hosts["storages"][hostname] = folder
if not self.active_controller_directory:
raise ValueError("Active controller not found")
def execute(self, plugins, output_directory):
"""Run a list of plugins
Parameters:
plugins (Plugin list): List of plugins to run
Errors:
FileNotFoundError
"""
plugin_output_dir = os.path.join(output_directory, "plugins")
os.makedirs(plugin_output_dir, exist_ok=True)
for plugin in plugins:
processing = "Processing plugin: " + os.path.basename(plugin.file)
hosts = {}
if (
plugin.state["hosts"] and len(plugin.state["hosts"]) >= 1
): # if host list is given
logger.info(
f"Processing plugin: {os.path.basename(plugin.file)}")
for h in plugin.state["hosts"]:
if h == "all":
hosts.update(self.hosts["workers"])
hosts.update(self.hosts["storages"])
hosts.update(self.hosts["controllers"])
else:
hosts.update(self.hosts[h])
for hostname, folderpath in hosts.items():
events = []
if plugin.state["algorithm"] == algorithms.SUBSTRING:
events = self.substring(
plugin.state["substring"],
[
os.path.join(folderpath, file)
for file in plugin.state["files"]
],
)
# creating output file
output_file = os.path.join(
plugin_output_dir,
f"substring_{hostname}",
)
if self.opts.verbose:
logger.info("output at "
+ os.path.relpath(output_file))
with open(output_file, "w") as file:
file.write(
f"Date range: {self.opts.start} until "
f"{self.opts.end}\n"
)
file.write(
f"substrings: "
f"{' '.join(plugin.state['substring'])}\n"
)
for line in events:
if line[-1] == "\n":
file.write(line)
else:
file.write(line + "\n")
else:
if plugin.state["algorithm"] == algorithms.SYSTEM_INFO:
info = self.system_info()
system_info_output = os.path.join(plugin_output_dir,
"system_info")
with open(system_info_output, "w") as file:
for i in info:
file.write(i + "\n")
for k, v in self.hosts.items():
file.write(f"{k}: {','.join(v.keys())}\n")
if self.opts.verbose:
logger.info(processing + ", output at "
+ os.path.relpath(system_info_output))
else:
logger.info(processing)
elif plugin.state["algorithm"] == algorithms.AUDIT:
hosts = {}
hosts.update(self.hosts["workers"])
hosts.update(self.hosts["storages"])
hosts.update(self.hosts["controllers"])
for hostname, folderpath in hosts.items():
self._create_output_file(
f"{hostname}_audit",
plugin_output_dir,
self.audit(
plugin.state["start"],
plugin.state["end"],
os.path.join(
folderpath, "var", "log", "dcmanager",
"audit.log"
),
),
processing,
)
elif plugin.state["algorithm"] == algorithms.SWACT_ACTIVITY:
self._create_output_file(
"swact_activity", plugin_output_dir,
self.swact_activity(), processing
)
elif plugin.state["algorithm"] == algorithms.PUPPET_ERRORS:
self._create_output_file(
"puppet_errors", plugin_output_dir,
self.puppet_errors(), processing
)
elif plugin.state["algorithm"] == algorithms.PROCESS_FAILURES:
self._create_output_file(
"process_failures", plugin_output_dir,
self.process_failures(), processing
)
elif plugin.state["algorithm"] == algorithms.ALARM:
alarms, logs = self.alarm(
plugin.state["alarm_exclude"],
plugin.state["entity_exclude"]
)
alarm_output = os.path.join(plugin_output_dir, "alarm")
log_output = os.path.join(plugin_output_dir, "log")
# creating output alarm file
with open(alarm_output, "w") as file:
for k, v in alarms.items():
file.write(f"{k}:\n")
for date in v["dates"]:
file.write(f" {date}\n")
# creating output log file
with open(log_output, "w") as file:
for k, v in logs.items():
file.write(f"{k}: {v['count']}\n")
file.write("\n")
for k, v in logs.items():
file.write(f"{k}:\n")
for date in v["dates"]:
file.write(f" {date}\n")
if self.opts.verbose:
logger.info(processing + ", output at "
+ os.path.relpath(alarm_output)
+ ", " + os.path.relpath(log_output))
else:
logger.info(processing)
elif plugin.state["algorithm"] == algorithms.HEARTBEAT_LOSS:
self._create_output_file(
"heartbeat_loss", plugin_output_dir,
self.heartbeat_loss(), processing
)
elif plugin.state["algorithm"] == algorithms.MAINTENANCE_ERR:
self._create_output_file(
"maintenance_errors", plugin_output_dir,
self.maintenance_errors(), processing
)
elif plugin.state["algorithm"] == algorithms.DAEMON_FAILURES:
self._create_output_file(
"daemon_failures", plugin_output_dir,
self.daemon_failures(), processing
)
elif plugin.state["algorithm"] == algorithms.STATE_CHANGES:
self._create_output_file(
"state_changes", plugin_output_dir,
self.state_changes(), processing
)
if not self.opts.verbose:
logger.info("Output files for plugins can be found at " +
os.path.relpath(plugin_output_dir))
# Running the correlator and printing the output from it
self.run_correlator(output_directory, plugin_output_dir)
# Built-in algorithms ------------------------------
def alarm(self, alarm_exclude=[], entity_exclude=[]):
"""Alarm algorithm
Gathers list of alarms and customer logs
Parameters:
alarm_exclude (string list) : List of alarm id patterns to not
search for
entity_exclude (string list): List of entity id patterns to not
search for
"""
alarm_data = {}
log_data = {}
with open(
os.path.join(
self.active_controller_directory,
"var", "extra", "database", "fm.db.sql.txt"
)
) as file:
start = False
for line in file:
# start of event log
if re.search(r"COPY (public\.)?event_log", line):
start = True
elif start and line == "\\.\n":
break
elif start:
entry = re.split(r"\t", line)
INDEX_ALARM_ID = 5
INDEX_ACTION = 6
INDEX_ENTITY_ID = 8
INDEX_ALARM_DATE = 9
INDEX_SEVERITY = 10
alarm_id = entry[INDEX_ALARM_ID]
entity_id = entry[INDEX_ENTITY_ID]
action = entry[INDEX_ACTION]
severity = entry[INDEX_SEVERITY]
alarm_date = entry[INDEX_ALARM_DATE]
entry_date = alarm_date.replace(
" ", "T"
) # making time format of alarm the same
if (self.opts.start <= entry_date
and entry_date <= self.opts.end):
cont = True
# Checks if the alarm is in the user specified list of
# alarm or entity ids
for id in alarm_exclude:
if id in alarm_id:
cont = False
break
for entity in entity_exclude:
if entity in entity_id:
cont = False
break
if not cont:
continue
try:
if action == "log":
log_info = log_data[
f"{alarm_id} {entity_id} {severity}"
]
log_info["count"] += 1
log_info["dates"].append(alarm_date)
else:
alarm_info = alarm_data[
f"{alarm_id} {entity_id} {severity}"
]
alarm_info["dates"].append(
f"{alarm_date} {action}")
except KeyError:
if entry[6] != "log":
alarm_data[
f"{alarm_id} {entity_id} {severity}"
] = {
"dates": [f"{alarm_date} {action}"],
}
else:
log_data[
f"{alarm_id} {entity_id} {severity}"
] = {
"count": 1,
"dates": [alarm_date],
}
for _, v in alarm_data.items():
v["dates"] = sorted(v["dates"])
temp = []
temp.append(v["dates"][0])
for i in range(1, len(v["dates"])):
if v["dates"][i].split()[2] != v["dates"][i-1].split()[2]:
temp.append(v["dates"][i])
v["dates"] = temp
for _, v in log_data.items():
v["dates"] = sorted(v["dates"])
return alarm_data, log_data
def substring(self, substr, files):
"""Substring algorithm
Looks for substrings within files
Parameters:
substr (string list): List of substrings to look for
files (string list): List of absolute filepaths to search in
Errors:
FileNotFoundError
"""
# don't analyze older files, continue with current file
CONTINUE_CURRENT = 0
# analyze older files, continue with current file
CONTINUE_CURRENT_OLD = 1
data = []
for file in files:
try:
if not os.path.exists(file):
if (re.search("controller-1_(.+)/var/log/mtcAgent.log",
file)):
continue
raise FileNotFoundError(f"File not found: {file}")
cont = True
# Searching through file
command = (f"""grep -Ea "{'|'.join(s for s in substr)}" """
f"""{file} 2>/dev/null""")
status = self._continue(file)
if (status == CONTINUE_CURRENT
or status == CONTINUE_CURRENT_OLD):
# continue with current file
if status == CONTINUE_CURRENT:
cont = False
self._evaluate_substring(data, command)
# Searching through rotated log files that aren't compressed
n = 1
while os.path.exists(f"{file}.{n}") and cont:
command = (f"""grep -Ea "{'|'.join(s for s in substr)}" """
f"""{file}.{n} 2>/dev/null""")
status = self._continue(f"{file}.{n}")
if (status == CONTINUE_CURRENT
or status == CONTINUE_CURRENT_OLD):
if status == CONTINUE_CURRENT:
cont = False
self._evaluate_substring(data, command)
n += 1
# Searching through rotated log files
while os.path.exists(f"{file}.{n}.gz") and cont:
command = (f"""zgrep -E "{'|'.join(s for s in substr)}" """
f"""{file}.{n}.gz 2>/dev/null""")
status = self._continue(f"{file}.{n}.gz", compressed=True)
if (status == CONTINUE_CURRENT
or status == CONTINUE_CURRENT_OLD):
if status == CONTINUE_CURRENT:
cont = False
self._evaluate_substring(data, command)
n += 1
except FileNotFoundError as e:
logger.error(e)
continue
return sorted(data)
def system_info(self):
"""System info algorithm
Presents basic information about the system
"""
data = []
with open(
os.path.join(
self.active_controller_directory, "etc", "platform",
"platform.conf"
)
) as file:
for line in file:
if "system_mode" in line:
data.append(
f"System Mode: "
f"{re.match('^system_mode=(.*)', line).group(1)}"
)
elif "system_type" in line:
data.append(
f"System Type: "
f"{re.match('^system_type=(.*)', line).group(1)}"
)
elif "distributed_cloud_role" in line:
role = re.match('^distributed_cloud_role=(.*)',
line).group(1)
data.append(f"Distributed cloud role: {role}")
elif "sw_version" in line:
data.append(
f"SW Version: "
f"{re.match('^sw_version=(.*)', line).group(1)}"
)
with open(
os.path.join(self.active_controller_directory, "etc", "build.info")
) as file:
for line in file:
if "BUILD_TYPE" in line:
data.append(
f"Build Type: "
f"{re.match('^BUILD_TYPE=(.*)', line).group(1)}"
)
elif re.match("^OS=(.*)", line):
data.append(f"OS: {re.match('^OS=(.*)', line).group(1)}")
return data
def swact_activity(self):
"""Swact activity algorithm
Presents all swacting activity in the system
"""
data = []
sm_files = []
sm_customer_files = []
swact_start = None
swact_in_progress = False
swact_end = None
for _, folder in self.hosts["controllers"].items():
sm_path = os.path.join(folder, "var", "log", "sm.log")
sm_files.append(sm_path)
sm_customer_path = os.path.join(folder, "var", "log",
"sm-customer.log")
sm_customer_files.append(sm_customer_path)
sm_substrings = ["Uncontrolled swact", "Swact has started,",
"Neighbor (.+) is now in the down",
"Service (.+) has reached max failures",
"Swact update"]
data = self.substring(sm_substrings, sm_files)
for i, line in enumerate(data):
if "Swact has started," in line and not swact_in_progress:
swact_in_progress = True
swact_start = datetime.strptime(line[0:19],
"%Y-%m-%dT%H:%M:%S")
elif "Swact update" in line and swact_in_progress:
swact_in_progress = False
swact_end = datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
line += f" SWACT TOOK {swact_end - swact_start} \n"
data[i] = line
sm_customer_substrings = [
"swact", "active-failed\\s+\\| disabling-failed\\s+\\|"
]
data += self.substring(sm_customer_substrings, sm_customer_files)
return sorted(data)
def puppet_errors(self):
"""Puppet errors algorithm
Presents log errors from puppet logs
"""
data = []
for host_type in self.hosts.keys():
for _, folder in self.hosts[host_type].items():
puppet_folder = os.path.join(folder, "var", "log", "puppet")
command = (f"""grep -rh "[m ]Error: " {puppet_folder} """
f"""2>/dev/null""")
self._evaluate_substring(data, command)
return sorted(data)
def process_failures(self):
"""Process failures algorithm
Presents log errors from pmond
"""
data = []
files = []
for host_type in self.hosts.keys():
for _, folder in self.hosts[host_type].items():
pmond = os.path.join(folder, "var", "log", "pmond.log")
files.append(pmond)
data = self.substring(["Error :"], files)
return data
def heartbeat_loss(self):
"""Heartbeat loss algorithm
Presents all heartbeat loss error messages in the system
"""
data = []
hb_files = []
for _, folder in self.hosts["controllers"].items():
hb_path = os.path.join(folder, "var", "log", "hbsAgent.log")
hb_files.append(hb_path)
hb_substrings = ["Heartbeat Loss"]
data = self.substring(hb_substrings, hb_files)
return sorted(data)
def maintenance_errors(self):
"""Maintenance errors algorithm
Presents maintenance errors and other relevant log messages in system
"""
data = []
mtc_files = []
for _, folder in self.hosts["controllers"].items():
agent = os.path.join(folder, "var", "log", "mtcAgent.log")
mtc_files.append(agent)
for host_type in self.hosts.keys():
for _, folder in self.hosts[host_type].items():
client = os.path.join(folder, "var", "log", "mtcClient.log")
mtc_files.append(client)
mtc_substrings = ["Error : ", "Configuration failure",
"In-Test Failure", "Loss Of Communication",
"Graceful Recovery Wait ",
"regained MTCALIVE from host that has rebooted",
"Connectivity Recovered ; ",
"auto recovery disabled", "Graceful Recovery Failed",
"MNFA ENTER", "MNFA EXIT", "MNFA POOL"]
data = self.substring(mtc_substrings, mtc_files)
return sorted(data)
def daemon_failures(self):
"""Daemon failures algorithm
Presents all failed puppet manifest messages in the system
"""
data = []
daemon_files = []
for host_type in self.hosts.keys():
for _, folder in self.hosts[host_type].items():
daemon_path = os.path.join(folder, "var", "log", "daemon.log")
daemon_files.append(daemon_path)
daemon_substrings = ["Failed to run the puppet manifest"]
data = self.substring(daemon_substrings, daemon_files)
return sorted(data)
def state_changes(self):
"""State changes algorithm
Presents all messages in the system regarding the state of hosts
"""
data = []
sc_files = []
for _, folder in self.hosts["controllers"].items():
sc_path = os.path.join(folder, "var", "log", "mtcAgent.log")
sc_files.append(sc_path)
sc_substrings = ["is ENABLED", "allStateChange (.+)locked-disabled"]
data = self.substring(sc_substrings, sc_files)
return sorted(data)
def audit(self, start, end, audit_log_path):
"""Counts audit events in dcmanager within a specified date range
Parameters:
start (string) : start date in YYYY-MM-DD HH:MM:SS format
end (string) : end date in YYYY-MM-DD HH:MM:SS format
audit_log_path (string) : absolute path of augit log file
"""
if not shutil.which("lnav"):
raise ValueError("Lnav program not found")
SECONDS_PER_HOUR = 3600
fmt = "%Y-%m-%d %H:%M:%S"
d1 = datetime.strptime(start, fmt)
d2 = datetime.strptime(end, fmt)
seconds = (d2 - d1).total_seconds()
log_texts = [
"Triggered subcloud audit%",
"Trigger patch audit%",
"Trigger load audit%",
"Triggered firmware audit%",
"Triggered kubernetes audit%",
# Counts sum of audits from all subclouds
]
INDEX_MIDDLE_WORD = 1
data = [("These rates and totals represent the sum of audits from "
+ "all subclouds")]
def command(text):
return (
f'lnav -R -n -c ";SELECT count(log_body) AS '
f'{text.split(" ")[INDEX_MIDDLE_WORD]}_total from '
f'openstack_log WHERE '
f'(log_time > \\"{start}\\" AND not log_time > \\"{end}\\")'
f' AND log_body like \\"{text}\\"" "{audit_log_path}"'
)
for text in log_texts:
p = subprocess.Popen(command(text), shell=True,
stdout=subprocess.PIPE)
for line in p.stdout:
line = line.decode("utf-8").strip()
if line.isnumeric():
data.append(
f"rate "
f"{round((int(line)/seconds * SECONDS_PER_HOUR), 3)} "
f"per hour. total: {line}"
)
else:
data.append(line)
return data
# -----------------------------------
def run_correlator(self, output_directory, plugin_output_dir):
"""Runs the correlator and prints the results differently based on if
the tool was run with or without the verbose option
Parameters:
output_directory (string) : directory to place output files from
correlator
plugin_output_dir (string) : directory with output files from
plugins
"""
correlator = Correlator(plugin_output_dir)
failures, events, alarms, state_changes = correlator.run(
self.opts.hostname)
failures_len, events_len = len(failures), len(events)
alarms_len, state_changes_len = len(alarms), len(state_changes)
failures.append("\nTotal failures found: " + str(failures_len) + "\n")
events.append("\nTotal events found: " + str(events_len) + "\n")
alarms.append("\nTotal alarms found: " + str(alarms_len) + "\n")
state_changes.append("\nTotal state changes found: "
+ str(state_changes_len) + "\n")
logger.info("\nRunning correlator...")
self._create_output_file("correlator_failures", output_directory,
failures, "")
self._create_output_file("correlator_events", output_directory,
events, "")
self._create_output_file("correlator_alarms", output_directory,
alarms, "")
self._create_output_file("correlator_state_changes", output_directory,
state_changes, "")
if not self.opts.verbose:
logger.info("Output can be found at "
+ os.path.relpath(output_directory) + "\n")
logger.info("Failures: " + str(failures_len))
for f in failures[:-1]:
if "Uncontrolled swact" in f:
logger.info(f[0:19] + " "
+ re.findall("active controller:? (.+)\n",
f)[0] + " uncontrolled swact")
elif "failure on" in f:
host = re.findall(r"failure on ([^\s]+) ", f)[0]
logger.info(f[0:19] + " " + host + " "
+ re.findall("^(.+) failure on ",
f[43:])[0].lower() + " failure")
else:
logger.info(f[:-1])
if failures_len != 0:
logger.info("\nEvents: " + str(events_len))
else:
logger.info("Events: " + str(events_len))
logger.info("Alarms: " + str(alarms_len))
logger.info("State Changes: " + str(state_changes_len))
else:
logger.info("\nFailures: " + str(failures_len))
for f in failures[:-1]:
logger.info(f[:-1])
# Dictionary to keep track of number of times events happens on
# each host
events_summ = {}
for e in events[:-1]:
k = e[20:-1].split(" (", 1)[0]
if not events_summ.get(k):
events_summ[k] = 1
else:
events_summ[k] += 1
if failures_len != 0:
logger.info("\nEvents: " + str(events_len))
else:
logger.info("Events: " + str(events_len))
for k, v in sorted(events_summ.items()):
logger.info(k + ": " + str(v) + " time(s)")
if events_len != 0:
logger.info("\nAlarms: " + str(alarms_len))
else:
logger.info("Alarms: " + str(alarms_len))
logger.info("The full list of alarms can be found at "
+ os.path.relpath(output_directory)
+ "/correlator_alarms")
# Dictionary to keep track of number of times state changes
# happens on each host
state_changes_summ = {}
for s in state_changes[:-1]:
k = s[20:-1]
if "enabled" in k:
k = k.split("enabled", 1)[0] + "enabled"
if not state_changes_summ.get(k):
state_changes_summ[k] = 1
else:
state_changes_summ[k] += 1
if alarms_len != 0:
logger.info("\nState Changes: " + str(state_changes_len))
else:
logger.info("State Changes: " + str(state_changes_len))
for k, v in sorted(state_changes_summ.items()):
logger.info(k + ": " + str(v) + " time(s)")
def _continue(self, file, compressed=False):
# don't analyze older files, continue with current file
CONTINUE_CURRENT = 0
# analyze older files, continue with current file
CONTINUE_CURRENT_OLD = 1
# don't analyze current file, continue to older files
CONTINUE_OLD = 2
# check date of first log event and compare with provided
# start, end dates
first = ""
if not compressed:
with open(file) as f:
line = f.readline()
first = line[0:19]
else:
with gzip.open(file, "rb") as f:
line = f.readline().decode("utf-8")
first = line[0:19]
try:
datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
first = line[0:19]
except ValueError:
return CONTINUE_CURRENT_OLD
if first < self.opts.start:
return CONTINUE_CURRENT
elif first < self.opts.end and first > self.opts.start:
return CONTINUE_CURRENT_OLD
elif first > self.opts.end:
return CONTINUE_OLD
def _evaluate_substring(self, data, command):
p = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE)
for line in p.stdout:
line = line.decode("utf-8")
# different date locations for log events
dates = [line[0:19], line[2:21]]
for date in dates:
try:
datetime.strptime(date, "%Y-%m-%dT%H:%M:%S")
if date > self.opts.start and date < self.opts.end:
if line[0] == "|": # sm-customer.log edge case
line = line[1:].strip()
line = re.sub("\\s+", " ", line)
data.append(line)
break
except ValueError:
if date == dates[-1]:
data.append(line)
def _extract_subfunction(self, host_info_path):
GROUP_ONE = 1
with open(host_info_path) as file:
for line in file:
hostname_match = re.match(
r"\s*hostname =>\s*\"?([^\"]*)(\n|\"\s*,?\s*\n)", line)
subfunction_match = re.match(
r"\s*subfunction =>\s*\"?([^\"]*)(\n|\"\s*,?\s*\n)", line)
if subfunction_match:
subfunction = subfunction_match.group(GROUP_ONE)
if hostname_match:
hostname = hostname_match.group(GROUP_ONE)
return hostname, subfunction
def _create_output_file(self, filename, directory, data, processing):
with open(os.path.join(directory, filename), "w") as file:
for i in data:
if i[-1] == "\n":
file.write(i)
else:
file.write(i + "\n")
if self.opts.verbose:
output = ("output at "
+ os.path.relpath(os.path.join(directory, filename)))
if processing == "":
logger.info(output)
else:
logger.info(processing + ", " + output)
elif processing != "":
logger.info(processing)

View File

@ -40,8 +40,8 @@ class Plugin:
"files": [],
"hosts": [],
"substring": [],
"alarm_ids": [],
"entity_ids": [],
"alarm_exclude": [],
"entity_exclude": [],
"start": None,
"end": None,
}
@ -93,10 +93,11 @@ class Plugin:
self.state["substring"].append(data[1])
elif label == "hosts":
self.state["hosts"] = value.replace(" ", "").split(",")
elif label == "alarm_ids":
self.state["alarm_ids"] = value.replace(" ", "").split(",")
elif label == "entity_ids":
self.state["entity_ids"] = value.replace(" ", "").split(",")
elif label == "alarm_exclude":
self.state["alarm_exclude"] = value.replace(" ", "").split(",")
elif label == "entity_exclude":
self.state["entity_exclude"] = value.replace(
" ", "").split(",")
elif label == "files":
self.state["files"] = value.replace(" ", "").split(",")
elif label == "start":
@ -117,74 +118,77 @@ class Plugin:
"""
plugin_name = os.path.basename(self.file)
HOSTS_ERR = f"plugin: {plugin_name} should not have hosts specified"
if self.state["algorithm"] == algorithms.SUBSTRING:
if len(self.state["files"]) == 0:
raise ValueError(
f"plugin: {plugin_name} needs files specified for substring algorithm"
)
if len(self.state["hosts"]) == 0:
raise ValueError(
f"plugin: {plugin_name} needs hosts specified for substring algorithm"
)
if len(self.state["substring"]) == 0:
raise ValueError(
f"plugin: {plugin_name} need substring specified for substring algorithm"
)
self.validate_state(plugin_name, "files")
self.validate_state(plugin_name, "hosts")
self.validate_state(plugin_name, "substring")
elif self.state["algorithm"] == algorithms.ALARM:
if len(self.state["hosts"]) > 0:
raise ValueError(
f"plugin: {plugin_name} should not have hosts to be specified"
)
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.SYSTEM_INFO:
if len(self.state["hosts"]) > 0:
raise ValueError(
f"plugin: {plugin_name} should not have hosts to be specified"
)
elif self.state["algorithm"] == algorithms.SWACT:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.SWACT_ACTIVITY:
if len(self.state["hosts"]) > 0:
raise ValueError(
f"plugin: {plugin_name} should not have hosts to be specified"
)
elif self.state["algorithm"] == algorithms.PUPPET:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.PUPPET_ERRORS:
if len(self.state["hosts"]) > 0:
raise ValueError(
f"plugin: {plugin_name} should not have hosts to be specified"
)
elif self.state["algorithm"] == algorithms.PROCESS_FAILURE:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.PROCESS_FAILURES:
if len(self.state["hosts"]) > 0:
raise ValueError(
f"plugin: {plugin_name} should not have hosts to be specified"
)
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.HEARTBEAT_LOSS:
if len(self.state["hosts"]) > 0:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.MAINTENANCE_ERR:
if len(self.state["hosts"]) > 0:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.DAEMON_FAILURES:
if len(self.state["hosts"]) > 0:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.STATE_CHANGES:
if len(self.state["hosts"]) > 0:
raise ValueError(HOSTS_ERR)
elif self.state["algorithm"] == algorithms.AUDIT:
if len(self.state["hosts"]) > 0:
raise ValueError(
f"plugin: {plugin_name} should not have hosts to be specified"
)
raise ValueError(HOSTS_ERR)
try:
datetime.strptime(self.state["start"], "%Y-%m-%d %H:%M:%S")
except:
raise ValueError(
f"plugin : {plugin_name} needs a start time in YYYY-MM-DD HH:MM:SS format"
f"plugin : {plugin_name} needs a start time in YYYY-MM-DD "
f"HH:MM:SS format"
)
try:
datetime.strptime(self.state["end"], "%Y-%m-%d %H:%M:%S")
except:
raise ValueError(
f"plugin : {plugin_name} needs an end time in YYYY-MM-DD HH:MM:SS format"
f"plugin : {plugin_name} needs an end time in YYYY-MM-DD "
f"HH:MM:SS format"
)
else:
raise ValueError(
f"plugin: {plugin_name} unknown algorithm {self.state['algorithm']}"
f"plugin: {plugin_name} unknown algorithm "
f"{self.state['algorithm']}"
)
for host in self.state["hosts"]:
if host not in ["controllers", "workers", "storages", "all"]:
raise ValueError(
f"host not recognized: '{host}', accepted hosts are 'controllers', 'workers', 'storages', 'all'"
f"host not recognized: '{host}', accepted hosts are "
f"'controllers', 'workers', 'storages', 'all'"
)
def validate_state(self, plugin_name, key):
if len(self.state[key]) == 0:
raise ValueError(
f"plugin: {plugin_name} needs {key} specified for "
f"substring algorithm"
)
def __str__(self) -> str:
return f"{json.dumps(self.state)} File: {self.file}"

View File

@ -0,0 +1,3 @@
algorithm=alarm
alarm_exclude=400., 800.
entity_exclude=subsystem=vim

View File

@ -0,0 +1 @@
algorithm=daemon_failures

View File

@ -0,0 +1 @@
algorithm=heartbeat_loss

View File

@ -0,0 +1 @@
algorithm=maintenance_errors

View File

@ -0,0 +1 @@
algorithm=process_failures

View File

@ -0,0 +1 @@
algorithm=puppet_errors

View File

@ -0,0 +1 @@
algorithm=state_changes

View File

@ -0,0 +1,5 @@
algorithm=substring
files=var/log/mtcAgent.log, var/log/sm.log
hosts=controllers
substring=operation failed
substring=Failed to send message

View File

@ -0,0 +1 @@
algorithm=swact_activity

View File

@ -0,0 +1 @@
algorithm=system_info

View File

@ -0,0 +1,360 @@
#!/usr/bin/env python3
########################################################################
#
# Copyright (c) 2022 Wind River Systems, Inc.
#
# SPDX-License-Identifier: Apache-2.0
#
########################################################################
#
# Description: The Report tool is used to gather relevant log events
# and information about the system from a collect bundle.
#
# The report tool allows user created plugins which decides relevance
# for log events. Plugins contain an algorithm label which instructs the
# tool what information to search and how to search for it.
#
# The report tool requires the collect bundle and host tarballs to be
# untarred.
#
# The report tool reads user plugins from the report directory in the
# top level of the collect bundle, and outputs files containing files
# containing relevant logs to this directory as well.
#
# Typical Usage:
# command line functionality
# ------------------------------- ----------------------------------
# > report.py - Run all plugins in directory
# > report.py [plugin ...] - Run only specified plugins
# > report.py <algorithm> [labels] - Run algorithm with labels
# > report.py --help - help message
# > report.py <algorithm> --help - algorithm specific help
#
# See --help output for a complete list of full and abbreviated
# command line options and examples of plugins.
#
# Refer to README file for more usage and output examples
#######################################################################
import argparse
from cmath import log
from datetime import datetime
from datetime import timedelta
from datetime import timezone
import logging
import os
import subprocess
import sys
import time
from execution_engine import ExecutionEngine
from plugin import Plugin
now = datetime.now(timezone.utc)
base_dir = os.path.dirname(os.path.realpath(__file__))
parent_dir = os.path.dirname(base_dir)
default_path = os.path.dirname(parent_dir)
plugins = []
parser = argparse.ArgumentParser(
description="Log Event Reporter",
epilog="Place plugins in 'plugins' directory found in 'report' directory "
"at top level of collect bundle.\nOutput files will be placed in 'report' "
"directory.\nThis tool will create a report.log and untar.log file along "
"with other output files.",
)
parser.add_argument(
"-v",
"--verbose",
action="store_true",
help="Verbose output",
)
parser.add_argument(
"-s",
"--start",
default="20000101",
help="Specify a start date in YYYYMMDD format for analysis "
"(default:20000101)",
)
parser.add_argument(
"-e",
"--end",
default=datetime.strftime(now + timedelta(days=1), "%Y%m%d"),
help="Specify an end date in YYYYMMDD format for analysis "
"(default: current date)",
)
parser.add_argument(
"-p",
"--plugin",
default=None,
nargs="*",
help="Specify what plugins to run (default: runs every plugin in plugins "
"folder)",
)
parser.add_argument(
"-d",
"--directory",
default=default_path,
help="Specify top level of collect bundle to analyze "
"(default: two levels above tool directory)",
)
parser.add_argument(
"--hostname",
default="all",
help="Specify host for correlator to find significant events and state "
"changes for (default: all hosts)",
)
subparsers = parser.add_subparsers(help="algorithms", dest="algorithm")
# substring algorithm arguments
parser_substring = subparsers.add_parser(
"substring",
formatter_class=argparse.RawTextHelpFormatter,
help="""Searches through specified files for lines containing specified
substring. There will be an output file for each host of the host
type specified.""",
epilog="Plugin file example:\n"
" algorithm=substring\n"
" files=var/log/mtcAgent.log, var/log/sm.log\n"
" hosts=controllers\n"
" substring=operation failed\n"
" substring=Failed to send message",
)
substring_required = parser_substring.add_argument_group("required arguments")
substring_required.add_argument(
"--files",
required=True,
nargs="+",
help="Files to perform substring analysis on (required)",
)
substring_required.add_argument(
"--substring", nargs="+", required=True,
help="Substrings to search for (required)"
)
substring_required.add_argument(
"--hosts",
choices=["controllers", "workers", "storages", "all"],
required=True,
nargs="+",
help="Host types to perform analysis on (required)",
)
# alarm algorithm arguments
parser_alarm = subparsers.add_parser(
"alarm",
formatter_class=argparse.RawTextHelpFormatter,
help="Searches through fm.db.sql.txt for alarms and logs except for those "
"specified. There are 2 output files: 'alarm', and 'log'",
epilog="Plugin file example:\n"
" algorithm=alarm\n"
" alarm_exclude=400., 800.\n"
" entity_exclude=subsystem=vim\n",
)
parser_alarm.add_argument(
"--alarm_exclude",
nargs="+",
required=False,
default=[],
help="Alarm id patterns to not search for (not required)",
)
parser_alarm.add_argument(
"--entity_exclude",
nargs="+",
required=False,
default=[],
help="Entity id patterns to not search for (not required)",
)
# system info algorithm
parser_system_info = subparsers.add_parser(
"system_info",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents information about the system",
epilog="Plugin file example:\n" " algorithm=system_info\n",
)
# swact activity algorithm
parser_swact_activity = subparsers.add_parser(
"swact_activity",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents system swacting activity",
epilog="Plugin file example:\n" " algorithm=swact_activity\n",
)
# puppet errors algorithm
parser_puppet_errors = subparsers.add_parser(
"puppet_errors",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any puppet errors",
epilog="Plugin file example:\n" " algorithm=puppet_errors\n",
)
# process failures algorithm
parser_process_failures = subparsers.add_parser(
"process_failures",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any process failures from pmond.log",
epilog="Plugin file example:\n" " algorithm=process_failures\n",
)
# daemon failures algorithm
parser_daemon_failures = subparsers.add_parser(
"daemon_failures",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any puppet manifest failures from daemon.log",
epilog="Plugin file example:\n" " algorithm=daemon_failures\n",
)
# heartbeat loss algorithm
parser_heartbeat_loss = subparsers.add_parser(
"heartbeat_loss",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any heartbeat loss error messages from hbsAgent.log",
epilog="Plugin file example:\n" " algorithm=heartbeat_loss\n",
)
# maintenance errors algorithm
parser_maintenance_errors = subparsers.add_parser(
"maintenance_errors",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents errors and other relevant messages from mtcAgent.log and "
"mtcClient.log",
epilog="Plugin file example:\n" " algorithm=maintenance_errors\n",
)
# state changes algorithm
parser_state_changes = subparsers.add_parser(
"state_changes",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any messages from mtcAgent.log regarding the state of "
"hosts, such as enabled/disabled",
epilog="Plugin file example:\n" " algorithm=state_changes\n",
)
# audit algorithm
parser_audit = subparsers.add_parser(
"audit",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents information about audit events in dcmanager.\n"
"The rates and totals represents the sum of audits on all subclouds ",
epilog="Plugin file example:\n"
" algorithm=audit\n"
" start=2022-06-01 10:00:00\n"
" end=2022-06-01 04:00:00\n",
)
parser_audit.add_argument(
"--start",
required=False,
default=datetime.strftime(now - timedelta(days=7), "%Y-%m-%d %H:%M:%S"),
type=str,
help="Specify a start date in YYYY-MM-DD HH:MM:SS format for analysis "
"(not required, default: 1 week ago)"
)
parser_audit.add_argument(
"--end",
required=False,
default=datetime.strftime(now, "%Y-%m-%d %H:%M:%S"),
type=str,
help="Specify an end date in YYYY-MM-DD HH:MM:SS format for analysis "
"(not required, default: today)"
)
args = parser.parse_args()
args.start = datetime.strptime(args.start, "%Y%m%d").strftime(
"%Y-%m-%dT%H:%M:%S")
args.end = datetime.strptime(args.end, "%Y%m%d").strftime("%Y-%m-%dT%H:%M:%S")
if args.directory.endswith("/"):
output_directory = os.path.join(
default_path, "report", "output",
os.path.basename(os.path.dirname(args.directory))
)
else:
output_directory = os.path.join(
default_path, "report", "output", os.path.basename(args.directory)
)
# creating report log
os.makedirs(output_directory, exist_ok=True)
open(os.path.join(output_directory, "report.log"), "w").close()
# setting up logger
formatter = logging.Formatter("%(message)s")
logger = logging.getLogger()
logging.basicConfig(
filename=os.path.join(output_directory, "report.log"),
level=logging.INFO,
format="%(asctime)s %(levelname)s: %(message)s",
datefmt="%Y-%m-%dT%H:%M:%S",
)
logging.Formatter.converter = time.gmtime
ch = logging.StreamHandler()
ch.setLevel(logging.INFO)
ch.setFormatter(formatter)
logger.addHandler(ch)
if not os.path.isdir(args.directory):
sys.exit("Top level of collect bundle given to analyze is not a directory")
else:
for obj in (os.scandir(args.directory)):
info = os.path.splitext(obj.name)
# TODO: ask user which file to report on if more than one tarball in
# directory
# Check if collect tarball is in given directory and extracts it if
# not already done
if (obj.is_file() and info[1] == ".tar"):
try:
result = subprocess.check_output(["tar", "tf", obj.path],
encoding="UTF-8")
result = result.split("\n", 1)
if not os.path.isdir(os.path.join(args.directory,
os.path.dirname(result[0]))):
subprocess.run(["tar", "xfC", obj.path, args.directory],
check=True)
subprocess.run(["echo", "extracted", obj.name], check=True)
args.directory = os.path.join(args.directory,
os.path.dirname(result[0]))
break
except subprocess.CalledProcessError as e:
logger.error(e)
try:
engine = ExecutionEngine(args, output_directory)
except ValueError as e:
logger.error(str(e))
sys.exit("Confirm you are running the report tool on a collect bundle")
if args.algorithm:
plugins.append(Plugin(opts=vars(args)))
else:
if args.plugin:
for p in args.plugin:
path = os.path.join(default_path, "report", "plugins", p)
if os.path.exists(path):
try:
plugins.append(Plugin(path))
except Exception as e:
logger.error(str(e))
else:
logger.warning(f"{p} plugin does not exist")
else:
path = os.path.join(default_path, "report", "plugins")
if not os.path.exists(path):
os.mkdir(path)
logger.error("Plugins folder is empty")
else:
for file in os.listdir(path):
try:
plugins.append(Plugin(os.path.join(path, file)))
except Exception as e:
logger.error(str(e))
engine.execute(plugins, output_directory)

View File

@ -3,3 +3,4 @@ etc/collect.d/* /etc/collect.d
usr/local/sbin/* /usr/local/sbin
usr/local/bin/collect /usr/local/bin
usr/sbin/collect /usr/sbin
/usr/local/bin/report/* /usr/local/bin/report

View File

@ -13,8 +13,10 @@ override_dh_auto_install:
install -m 755 -d $(SYSCONFDIR)/collect.d
install -m 755 -d $(SYSCONFDIR)/collect
install -m 755 -d $(SYSCONFDIR)/collect/plugins # Report Tool
install -m 755 -d $(ROOT)/usr/local/sbin
install -m 755 -d $(ROOT)/usr/local/bin
install -m 755 -d $(ROOT)/usr/local/bin/report/tool # Report Tool
install -m 755 -d $(SBINDIR)
install -m 755 -p collect $(ROOT)/usr/local/sbin/collect
@ -26,6 +28,24 @@ override_dh_auto_install:
install -m 755 -p expect_done $(ROOT)/usr/local/sbin/expect_done
install -m 755 -p mariadb-cli.sh $(ROOT)/usr/local/sbin/mariadb-cli
# Report Tool
install -m 755 -p report/report.py $(ROOT)/usr/local/bin/report/tool/report.py
install -m 755 -p report/execution_engine.py $(ROOT)/usr/local/bin/report/tool/execution_engine.py
install -m 755 -p report/algorithms.py $(ROOT)/usr/local/bin/report/tool/algorithms.py
install -m 755 -p report/plugin.py $(ROOT)/usr/local/bin/report/tool/plugin.py
install -m 755 -p report/correlator.py $(ROOT)/usr/local/bin/report/tool/correlator.py
install -m 755 -p report/README $(ROOT)/usr/local/bin/report/tool/README
install -m 755 -p report/plugins/alarm $(SYSCONFDIR)/collect/plugins/alarm
install -m 755 -p report/plugins/daemon_failures $(SYSCONFDIR)/collect/plugins/daemon_failures
install -m 755 -p report/plugins/heartbeat_loss $(SYSCONFDIR)/collect/plugins/heartbeat_loss
install -m 755 -p report/plugins/maintenance_errors $(SYSCONFDIR)/collect/plugins/maintenance_errors
install -m 755 -p report/plugins/process_failures $(SYSCONFDIR)/collect/plugins/process_failures
install -m 755 -p report/plugins/puppet_errors $(SYSCONFDIR)/collect/plugins/puppet_errors
install -m 755 -p report/plugins/state_changes $(SYSCONFDIR)/collect/plugins/state_changes
install -m 755 -p report/plugins/substring $(SYSCONFDIR)/collect/plugins/substring
install -m 755 -p report/plugins/swact_activity $(SYSCONFDIR)/collect/plugins/swact_activity
install -m 755 -p report/plugins/system_info $(SYSCONFDIR)/collect/plugins/system_info
install -m 755 -p collect_sysinv.sh $(SYSCONFDIR)/collect.d/collect_sysinv
install -m 755 -p collect_psqldb.sh $(SYSCONFDIR)/collect.d/collect_psqldb
install -m 755 -p collect_mariadb.sh $(SYSCONFDIR)/collect.d/collect_mariadb

View File

@ -1,62 +0,0 @@
Refer to report.py file header for a description of the tool
Example:
Consider the following collect bundle structure
SELECT_NODES_20220527.193605
├── controller-0_20220527.193605
│  ├── etc
│  ├── root
│  └── var
├── controller-1_20220527.193605
│  ├── etc
│  ├── root
│  └── var
├── plugins (where the plugin files will be placed)
│  ├── alarm_plugin_example
│  └── substring_plugin_example
├── report
└── tool (where the tool will be placed)
└── output (where the output files will be placed)
> cat plugins/alarm_plugin_example
algorithm=alarm
alarm_ids=400.,401.
entity_ids = host=controller-0
> cat plugins/substring_plugin_example
algorithm=substring
files=var/log/mtcAgent.log
hosts=controllers
substring=operation failed
> report/tool/report.py --start 20220501 --end 20220530
Running the command above will populate the report folder with output files.
The tool also provides default values, more details are in 'report.py -h'.
The substring algorithm creates an output file for every host of the
specified host type. The files will contain log events within the
provided date range containing the substring 'operation failed'.
The alarm algorithm creates two output file: 'log' and 'alarm'
'log' contains customer log messages created within the provided date range,
and 'alarm' contains system alarms created within the provided date range.
For more detailed information about an algorithm use 'report.py <algorithm> -h'.
Here is the report directory after running the above command
report
├── output
│ └── 20220815.140008 (time in utc when tool was ran)
│ ├── alarm
│ ├── controller-0_substring_plugin_example_substring
│ ├── controller-1_substring_plugin_example_substring
│ ├── report.log (log file for report tool)
│ └── log
└── tool (where the report tool is)

View File

@ -1,546 +0,0 @@
########################################################################
#
# Copyright (c) 2022 Wind River Systems, Inc.
#
# SPDX-License-Identifier: Apache-2.0
#
########################################################################
#
# This file contains the ExecutionEngine class.
# The ExecutionEngine class contains all the available algorithms.
#
# The ExecutionEngine class runs plugins and gathers relevant logs and
# information, creating output files in the report directory.
#
########################################################################
from datetime import datetime
import gzip
import logging
import os
import re
import shutil
import subprocess
import algorithms
logger = logging.getLogger(__name__)
class ExecutionEngine:
def __init__(self, opts):
"""Constructor for the ExecutionEngine class
Parameters:
opts (dictionary): Options from command line
"""
self.opts = opts
self.hosts = {"controllers": {}, "workers": {}, "storages": {}}
self.active_controller_directory = None
for folder in (f.path for f in os.scandir(self.opts.directory)):
database_path = os.path.join(folder, "var", "extra", "database")
host_info_path = os.path.join(folder, "var", "extra", "host.info")
if os.path.isdir(database_path) and os.listdir(database_path):
self.active_controller_directory = folder
if os.path.exists(host_info_path):
hostname, subfunction = self._extract_subfunction(host_info_path)
if "controller" in subfunction:
self.hosts["controllers"][hostname] = folder
elif "worker" in subfunction:
self.hosts["workers"][hostname] = folder
elif "storage" in subfunction:
self.hosts["storages"][hostname] = folder
if not self.active_controller_directory:
raise ValueError("Active controller not found")
def execute(self, plugins, output_directory):
"""Run a list of plugins
Parameters:
plugins (Plugin list): List of plugins to run
Errors:
FileNotFoundError
"""
for plugin in plugins:
logger.info(f"Processing plugin: {os.path.basename(plugin.file)}")
hosts = {}
if (
plugin.state["hosts"] and len(plugin.state["hosts"]) >= 1
): # if host list is given
for h in plugin.state["hosts"]:
if h == "all":
hosts.update(self.hosts["workers"])
hosts.update(self.hosts["storages"])
hosts.update(self.hosts["controllers"])
else:
hosts.update(self.hosts[h])
for hostname, folderpath in hosts.items():
events = []
if plugin.state["algorithm"] == algorithms.SUBSTRING:
try:
events = self.substring(
plugin.state["substring"],
[
os.path.join(folderpath, file)
for file in plugin.state["files"]
],
)
except FileNotFoundError as e:
logger.error(e)
continue
# creating output file
output_file = os.path.join(
output_directory,
f"{hostname}_{os.path.basename(plugin.file)}_{plugin.state['algorithm']}",
)
logger.info("output at " + output_file)
with open(output_file, "w") as file:
file.write(
f"Date range: {self.opts.start} until {self.opts.end}\n"
)
file.write(
f"substrings: {' '.join(plugin.state['substring'])}\n"
)
for line in events:
file.write(line + "\n")
else:
if plugin.state["algorithm"] == algorithms.SYSTEM_INFO:
info = self.system_info()
system_info_output = os.path.join(output_directory, "system_info")
with open(system_info_output, "w") as file:
for i in info:
file.write(i + "\n")
for k, v in self.hosts.items():
file.write(f"{k}: {','.join(v.keys())}\n")
logger.info("output at " + system_info_output)
elif plugin.state["algorithm"] == algorithms.AUDIT:
hosts = {}
hosts.update(self.hosts["workers"])
hosts.update(self.hosts["storages"])
hosts.update(self.hosts["controllers"])
for hostname, folderpath in hosts.items():
self._create_output_file(
f"{hostname}_audit",
output_directory,
self.audit(
plugin.state["start"],
plugin.state["end"],
os.path.join(
folderpath, "var", "log", "dcmanager", "audit.log"
),
),
)
elif plugin.state["algorithm"] == algorithms.SWACT:
self._create_output_file(
"swact_activity", output_directory, self.swact()
)
elif plugin.state["algorithm"] == algorithms.PUPPET:
self._create_output_file(
"puppet_errors", output_directory, self.puppet()
)
elif plugin.state["algorithm"] == algorithms.PROCESS_FAILURE:
self._create_output_file(
"process_failures", output_directory, self.process_failure()
)
elif plugin.state["algorithm"] == algorithms.ALARM:
alarms, logs = self.alarm(
plugin.state["alarm_ids"], plugin.state["entity_ids"]
)
alarm_output = os.path.join(output_directory, "alarm")
log_output = os.path.join(output_directory, "log")
os.makedirs(os.path.dirname(log_output), exist_ok=True)
# creating output alarm file
with open(alarm_output, "w") as file:
for k, v in alarms.items():
file.write(f"{k} {v['count']}\n")
file.write("\n")
for k, v in alarms.items():
file.write(f"{k}\n")
for date in v["dates"]:
file.write(f" {date}\n")
# creating output log file
with open(log_output, "w") as file:
for k, v in logs.items():
file.write(f"{k} {v['count']}\n")
file.write("\n")
for k, v in logs.items():
file.write(f"{k}\n")
for date in v["dates"]:
file.write(f" {date}\n")
logger.info("output at " + alarm_output)
logger.info("output at " + log_output)
# Built-in algorithms ------------------------------
def alarm(self, alarm_ids=[], entity_ids=[]):
"""Alarm algorithm
Gathers list of alarms and customer logs
Parameters:
alarm_ids (string list) : List of alarm id patterns to search for
entity_ids (string list): List of entity id patterns to search for
"""
alarm_data = {}
log_data = {}
with open(
os.path.join(
self.active_controller_directory,
"var",
"extra",
"database",
"fm.db.sql.txt",
)
) as file:
start = False
for line in file:
# start of event log
if "COPY event_log" in line:
start = True
elif start and line == "\\.\n":
break
elif start:
entry = re.split(r"\t", line)
INDEX_ALARM_ID = 5
INDEX_ACTION = 6
INDEX_ENTITY_ID = 8
INDEX_ALARM_DATE = 9
INDEX_SEVERITY = 10
alarm_id = entry[INDEX_ALARM_ID]
entity_id = entry[INDEX_ENTITY_ID]
action = entry[INDEX_ACTION]
severity = entry[INDEX_SEVERITY]
alarm_date = entry[INDEX_ALARM_DATE]
entry_date = alarm_date.replace(
" ", "T"
) # making time format of alarm the same
if self.opts.start <= entry_date and entry_date <= self.opts.end:
# if the alarm is not in the user specified list of alarm or entity ids
for id in alarm_ids:
if id in alarm_id:
break
else:
if len(alarm_ids) > 0:
continue
for entity in entity_ids:
if entity in entity_id:
break
else:
if len(entity_ids) > 0:
continue
try:
if action == "log":
log_info = log_data[
f"{alarm_id} {entity_id} {severity}"
]
log_info["count"] += 1
log_info["dates"].append(alarm_date)
else:
alarm_info = alarm_data[
f"{alarm_id} {entity_id} {severity}"
]
alarm_info["count"] += 1
alarm_info["dates"].append(f"{alarm_date} {action}")
except KeyError:
if entry[6] != "log":
alarm_data[f"{alarm_id} {entity_id} {severity}"] = {
"count": 1,
"dates": [f"{alarm_date} {action}"],
}
else:
log_data[f"{alarm_id} {entity_id} {severity}"] = {
"count": 1,
"dates": [alarm_date],
}
for _, v in alarm_data.items():
v["dates"] = sorted(v["dates"])
for _, v in log_data.items():
v["dates"] = sorted(v["dates"])
return alarm_data, log_data
def substring(self, substr, files):
"""Substring algorithm
Looks for substrings within files
Parameters:
substr (string list): List of substrings to look for
files (string list): List of absolute filepaths to search in
Errors:
FileNotFoundError
"""
CONTINUE_CURRENT = 0 # don't analyze older files, continue with current file
CONTINUE_CURRENT_OLD = 1 # analyze older files, continue with current file
data = []
for file in files:
if not os.path.exists(file):
raise FileNotFoundError(f"File not found: {file}")
cont = True
# Searching through file
command = f"""grep -Ea "{'|'.join(s for s in substr)}" {file}"""
status = self._continue(file)
if (
status == CONTINUE_CURRENT or status == CONTINUE_CURRENT_OLD
): # continue with current file
if status == CONTINUE_CURRENT:
cont = False
self._evaluate_substring(data, command)
# Searching through rotated log files
n = 1
while os.path.exists(f"{file}.{n}.gz") and cont:
command = f"""zgrep -E "{'|'.join(s for s in substr)}" {file}.{n}.gz"""
status = self._continue(f"{file}.{n}.gz", compressed=True)
if status == CONTINUE_CURRENT or status == CONTINUE_CURRENT_OLD:
if status == CONTINUE_CURRENT:
cont = False
self._evaluate_substring(data, command)
n += 1
return sorted(data)
def system_info(self):
"""System info algorithm
Presents basic information about the system
"""
data = []
with open(
os.path.join(
self.active_controller_directory, "etc", "platform", "platform.conf"
)
) as file:
for line in file:
if "system_mode" in line:
data.append(
f"System Mode: {re.match('^system_mode=(.*)', line).group(1)}"
)
elif "system_type" in line:
data.append(
f"System Type: {re.match('^system_type=(.*)', line).group(1)}"
)
elif "distributed_cloud_role" in line:
data.append(
f"Distributed cloud role: {re.match('^distributed_cloud_role=(.*)', line).group(1)}"
)
elif "sw_version" in line:
data.append(
f"SW Version: {re.match('^sw_version=(.*)', line).group(1)}"
)
with open(
os.path.join(self.active_controller_directory, "etc", "build.info")
) as file:
for line in file:
if "BUILD_TYPE" in line:
data.append(
f"Build Type: {re.match('^BUILD_TYPE=(.*)', line).group(1)}"
)
elif re.match("^OS=(.*)", line):
data.append(f"OS: {re.match('^OS=(.*)', line).group(1)}")
return data
def swact(self):
"""Swact activity algorithm
Presents all swacting activity in the system
"""
data = []
sm_files = []
sm_customer_files = []
swact_start = None
swact_in_progress = False
swact_end = None
for _, folder in self.hosts["controllers"].items():
sm_path = os.path.join(folder, "var", "log", "sm.log")
sm_files.append(sm_path)
sm_substrings = ["Swact has started,", "Swact update"]
data = self.substring(sm_substrings, sm_files)
for i, line in enumerate(data):
if "Swact has started," in line and not swact_in_progress:
swact_in_progress = True
swact_start = datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
elif "Swact update" in line and swact_in_progress:
swact_in_progress = False
swact_end = datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
line += f" SWACT TOOK {swact_end - swact_start} \n"
data[i] = line
for _, folder in self.hosts["controllers"].items():
sm_customer_path = os.path.join(folder, "var", "log", "sm-customer.log")
sm_customer_files.append(sm_customer_path)
sm_customer_substrings = ["swact"]
data += self.substring(sm_customer_substrings, sm_customer_files)
return sorted(data)
def puppet(self):
"""Puppet error algorithm
Presents log errors from puppet logs
"""
data = []
for _, folder in self.hosts["controllers"].items():
puppet_folder = os.path.join(folder, "var", "log", "puppet")
command = f"grep -rh 'Error:' {puppet_folder}"
self._evaluate_substring(data, command)
return sorted(data)
def process_failure(self):
"""Process failure algorithm
Presents log errors from pmond
"""
data = []
files = []
for host_type in self.hosts.keys():
for _, folder in self.hosts[host_type].items():
pmond = os.path.join(folder, "var", "log", "pmond.log")
files.append(pmond)
data = self.substring(["Error :"], files)
return data
def audit(self, start, end, audit_log_path):
"""Counts audit events in dcmanager within a specified date range
Parameters:
start (string) : start date in YYYY-MM-DD HH:MM:SS format
end (string) : end date in YYYY-MM-DD HH:MM:SS format
audit_log_path (string) : absolute path of augit log file
"""
if not shutil.which("lnav"):
raise ValueError("Lnav program not found")
SECONDS_PER_HOUR = 3600
fmt = "%Y-%m-%d %H:%M:%S"
d1 = datetime.strptime(start, fmt)
d2 = datetime.strptime(end, fmt)
seconds = (d2 - d1).total_seconds()
log_texts = [
"Triggered subcloud audit%",
"Trigger patch audit%",
"Trigger load audit%",
"Triggered firmware audit%",
"Triggered kubernetes audit%",
# Counts sum of audits from all subclouds
]
INDEX_MIDDLE_WORD = 1
data = ["These rates and totals represent the sum of audits from all subclouds"]
def command(text):
return (
f'lnav -R -n -c ";SELECT count(log_body) AS {text.split(" ")[INDEX_MIDDLE_WORD]}_total'
f' from openstack_log WHERE (log_time > \\"{start}\\" AND not log_time > \\"{end}\\")'
f' AND log_body like \\"{text}\\"" "{audit_log_path}"'
)
for text in log_texts:
p = subprocess.Popen(command(text), shell=True, stdout=subprocess.PIPE)
for line in p.stdout:
line = line.decode("utf-8").strip()
if line.isnumeric():
data.append(
f"rate {round((int(line)/seconds * SECONDS_PER_HOUR), 3)} per hour. total: {line}"
)
else:
data.append(line)
return data
# -----------------------------------
def _continue(self, file, compressed=False):
CONTINUE_CURRENT = 0 # don't analyze older files, continue with current file
CONTINUE_CURRENT_OLD = 1 # analyze older files, continue with current file
CONTINUE_OLD = 2 # don't analyze current file, continue to older files
# check date of first log event and compare with provided start end dates
first = ""
if not compressed:
with open(file) as f:
line = f.readline()
first = line[0:19]
else:
with gzip.open(file, "rb") as f:
line = f.readline().decode("utf-8")
first = line[0:19]
try:
datetime.strptime(line[0:19], "%Y-%m-%dT%H:%M:%S")
first = line[0:19]
except ValueError:
return CONTINUE_CURRENT_OLD
if first < self.opts.start:
return CONTINUE_CURRENT
elif first < self.opts.end and first > self.opts.start:
return CONTINUE_CURRENT_OLD
elif first > self.opts.end:
return CONTINUE_OLD
def _evaluate_substring(self, data, command):
p = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE)
for line in p.stdout:
line = line.decode("utf-8")
dates = [line[0:19], line[2:21]] # different date locations for log events
for date in dates:
try:
datetime.strptime(date, "%Y-%m-%dT%H:%M:%S")
if date > self.opts.start and date < self.opts.end:
if line[0] == "|": # sm-customer.log edge case
line = line.replace("|", "").strip()
line = re.sub("\s+", " ", line)
data.append(line)
break
except ValueError:
if date == dates[-1]:
data.append(line)
def _extract_subfunction(self, host_info_path):
GROUP_ONE = 1
with open(host_info_path) as file:
for line in file:
hostname_match = re.match("^hostname => (.+)", line)
subfunction_match = re.match("^subfunction => (.+)", line)
if subfunction_match:
subfunction = subfunction_match.group(GROUP_ONE)
if hostname_match:
hostname = hostname_match.group(GROUP_ONE)
return hostname, subfunction
def _create_output_file(self, filename, directory, events):
with open(os.path.join(directory, filename), "w") as file:
for i in events:
file.write(i + "\n")
logger.info("output at " + os.path.join(directory, filename))

View File

@ -1,257 +0,0 @@
#!/usr/bin/env python3
########################################################################
#
# Copyright (c) 2022 Wind River Systems, Inc.
#
# SPDX-License-Identifier: Apache-2.0
#
########################################################################
#
# Description: The Report tool is used to gather relevant log events
# and information about the system from a collect bundle.
#
# The report tool allows user created plugins which decides relevance
# for log events. Plugins contain an algorithm label which instructs the
# tool what information to search and how to search for it.
#
# The report tool requires the collect bundle and host tarballs to be
# untarred.
#
# The report tool reads user plugins from a plugins directory in the
# top level of the collect bundle, and outputs files containing
# relevant logs to a report directory in the top level as well.
#
# Typical Usage:
# command line functionality
# ------------------------------- ----------------------------------
# > report.py - Run all plugins in directory
# > report.py [plugin ...] - Run only specified plugins
# > report.py <algorithm> [labels] - Run algorithm with labels
# > report.py --help - help message
# > report.py <algorithm> --help - algorithm specific help
#
# See --help output for a complete list of full and abbreviated
# command line options and examples of plugins.
#
# Refer to README file for more usage and output examples
#######################################################################
import argparse
from cmath import log
from datetime import datetime
from datetime import timezone
import logging
import os
import time
from execution_engine import ExecutionEngine
from plugin import Plugin
now = datetime.now(timezone.utc)
base_dir = os.path.realpath(__file__)
default_path = os.path.join(os.path.dirname(base_dir), "..", "..")
plugins = []
parser = argparse.ArgumentParser(
description="Log Event Reporter",
epilog="Place plugins in 'plugins' directory at top level of collect bundle. Output files will be placed in 'report' directory."
"\nThis tool will create a report.log file along with other output files",
)
parser.add_argument(
"-s",
"--start",
default="20000101",
help="Specify a start date in YYYYMMDD format for analysis (default:20000101)",
)
parser.add_argument(
"-e",
"--end",
default=datetime.strftime(now, "%Y%m%d"),
help="Specify an end date in YYYYMMDD format for analysis (default: current date)",
)
parser.add_argument(
"-p",
"--plugin",
default=None,
nargs="*",
help="Specify what plugins to run (default: runs every plugin in plugins folder)",
)
parser.add_argument(
"-d",
"--directory",
default=default_path,
help="Specify top level of collect bundle to analyze (default: two levels above current location)",
)
subparsers = parser.add_subparsers(help="algorithms", dest="algorithm")
# substring algorithm arguments
parser_substring = subparsers.add_parser(
"substring",
formatter_class=argparse.RawTextHelpFormatter,
help="""Searches through specified files for lines containing specified substring.
There will be an output file for each host of the host type specified.""",
epilog="Plugin file example:\n"
" algorithm=substring\n"
" files=mtcAgent.log, sm.log\n"
" hosts=controllers, workers\n"
" substring=Swact in progress\n"
" substring=Swact update",
)
substring_required = parser_substring.add_argument_group("required arguments")
substring_required.add_argument(
"--files",
required=True,
nargs="+",
help="Files to perform substring analysis on (required)",
)
substring_required.add_argument(
"--substring", nargs="+", required=True, help="Substrings to search for (required)"
)
substring_required.add_argument(
"--hosts",
choices=["controllers", "workers", "storages", "all"],
required=True,
nargs="+",
help="Host types to perform analysis on (required)",
)
# alarm algorithm arguments
parser_alarm = subparsers.add_parser(
"alarm",
formatter_class=argparse.RawTextHelpFormatter,
help="Searches through fm.db.sql.txt for alarms and logs. There are 2 output files: 'alarm', and 'log'",
epilog="Plugin file example:\n"
" algorithm=alarm\n"
" alarm_ids=400.005,200.004\n"
" entity_ids= host=controller-0,host=controller-1\n",
)
parser_alarm.add_argument(
"--alarm_ids",
nargs="+",
required=False,
default=[],
help="Alarm id patterns to search for (not required)",
)
parser_alarm.add_argument(
"--entity_ids",
nargs="+",
required=False,
default=[],
help="Entity id patterns to search for (not required)",
)
# system info algorithm
parser_system_info = subparsers.add_parser(
"system_info",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents information about the system",
epilog="Plugin file example:\n" " algorithm=system_info\n",
)
# swact activity algorithm
parser_swact = subparsers.add_parser(
"swact",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents system swacting activity",
epilog="Plugin file example:\n" " algorithm=swact\n",
)
# puppet errors algorithm
parser_puppet = subparsers.add_parser(
"puppet",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any puppet errors",
epilog="Plugin file example:\n" " algorithm=puppet\n",
)
# process failure algorithm
parser_process_failure = subparsers.add_parser(
"process_failure",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents any process failures from pmond.log",
epilog="Plugin file example:\n" " algorithm=process_failure\n",
)
# audit algorithm
parser_audit = subparsers.add_parser(
"audit",
formatter_class=argparse.RawTextHelpFormatter,
help="Presents information about audit events in dcmanager.\n"
"The rates and totals represents the sum of audits on all subclouds ",
epilog="Plugin file example:\n"
" algorithm=audit\n"
" start=2022-06-01 10:00:00\n"
" end=2022-06-01 04:00:00\n",
)
parser_audit_required = parser_audit.add_argument_group("required arguments")
parser_audit_required.add_argument("--start", required=True)
parser_audit_required.add_argument(
"--end",
required=True,
)
args = parser.parse_args()
args.start = datetime.strptime(args.start, "%Y%m%d").strftime("%Y-%m-%dT%H:%M:%S")
args.end = datetime.strptime(args.end, "%Y%m%d").strftime("%Y-%m-%dT%H:%M:%S")
output_directory = os.path.join(
args.directory, "report", "output", now.strftime("%Y%m%d.%H%M%S")
)
# creating report log
os.makedirs(output_directory)
open(os.path.join(output_directory, "report.log"), "w").close()
# setting up logger
formatter = logging.Formatter("%(message)s")
logger = logging.getLogger()
logging.basicConfig(
filename=os.path.join(output_directory, "report.log"),
level=logging.INFO,
format="%(asctime)s %(levelname)s: %(message)s",
datefmt="%Y-%m-%dT%H:%M:%S",
)
logging.Formatter.converter = time.gmtime
ch = logging.StreamHandler()
ch.setLevel(logging.INFO)
ch.setFormatter(formatter)
logger.addHandler(ch)
try:
engine = ExecutionEngine(args)
except ValueError as e:
logger.error(str(e))
if args.algorithm:
plugins.append(Plugin(opts=vars(args)))
else:
if args.plugin:
for p in args.plugin:
path = os.path.join(args.directory, "plugins", p)
if os.path.exists(path):
try:
plugins.append(Plugin(path))
except Exception as e:
logger.error(str(e))
else:
logger.warning(f"{p} plugin does not exist")
else:
path = os.path.join(args.directory, "plugins")
if not os.path.exists(path):
os.mkdir(path)
logger.error("Plugins folder is empty")
else:
for file in os.listdir(path):
try:
plugins.append(Plugin(os.path.join(path, file)))
except Exception as e:
logger.error(str(e))
engine.execute(plugins, output_directory)