def screen_return_codes(msg_on_hit, return_code_target, o, e, rc, command):
"""
Provides a central mechanism to screen return codes from executing smart
commands. This is required as some non zero return codes would otherwise
trigger a generic exception clause in our general purpose run_command.
If the target return code is seen then email root with the message
provided, otherwise raise a generic exception with the command information.
N.B. May be done better by acting as a SMART run_command wrapper (Future).
:param msg_on_hit: message used to email root
:param return_code_target: return code to screen for
:param o: the output from the command when it was run
:param e: the error from the command when it was run
:param rc: the return code from running the command
:param command: the command that produced the previous o, e, and rc params.
"""
# if our return code is our target then log with our message and email root
# with the same.
if rc == return_code_target:
logger.error(msg_on_hit)
email_root('S.M.A.R.T error', msg_on_hit)
# In all other non zero (error) instances we raise an exception as normal.
elif rc != 0:
e_msg = ('non-zero code(%d) returned by command: %s output: '
'%s error: %s' % (rc, command, o, e))
logger.error(e_msg)
raise CommandException(('%s' % command), o, e, rc)
def screen_return_codes(msg_on_hit, return_code_target, o, e, rc, command):
"""Provides a central mechanism to screen return codes from executing smart
commands. This is required as some non zero return codes would otherwise
trigger a generic exception clause in our general purpose run_command. If
the target return code is seen then email root with the message provided,
otherwise raise a generic exception with the command information. N.B. May
be done better by acting as a SMART run_command wrapper (Future).
:param msg_on_hit: message used to email root
:param return_code_target: return code to screen for
:param o: the output from the command when it was run
:param e: the error from the command when it was run
:param rc: the return code from running the command
:param command: the command that produced the previous o, e, and rc params.
"""
# if our return code is our target then log with our message and email root
# with the same.
if rc == return_code_target:
logger.error(msg_on_hit)
email_root('S.M.A.R.T error', msg_on_hit)
# In all other non zero (error) instances we raise an exception as normal.
elif rc != 0:
e_msg = ('non-zero code(%d) returned by command: %s output: '
'%s error: %s' % (rc, command, o, e))
logger.error(e_msg)
raise CommandException(('%s' % command), o, e, rc)
def generate_otp(username):
#Generate a random 6 chars text and sent to root mail
new_otp = ''.join(random.choice(string.letters + string.digits) for _ in range(6))
otp_subject = 'Received password reset request for uid 0 user'
otp_message = 'System has received a password reset request for user %s\n\n OTP string value is: %s' % (username, new_otp)
email_root(otp_subject, otp_message)
return new_otp
def generate_otp(username):
#Generate a random 6 chars text and sent to root mail
new_otp = ''.join(random.choice(string.letters + string.digits) for _ in range(6))
otp_subject = 'Received password reset request for uid 0 user'
otp_message = 'System has received a password reset request for user %s\n\n OTP string value is: %s' % (username, new_otp)
email_root(otp_subject, otp_message)
return new_otp
def error_logs(device):
o, e, rc = run_command(
[SMART, '-l', 'error', '/dev/%s' % device], throw=False)
# As we mute exceptions when calling the above command we should at least
# examine what we have as return code (rc); 64 has been seen when the error
# log contains errors but otherwise executes successfully so we catch this.
if rc == 64:
e_msg = 'Drive /dev/%s has logged S.M.A.R.T errors. Please view ' \
'the Error logs tab for this device.' % device
logger.error(e_msg)
email_root('S.M.A.R.T error', e_msg)
# In all other instances that are an error (non zero) we raise exception
# as normal.
elif rc != 0:
e_msg = ('non-zero code(%d) returned by command: %s -l error output: '
'%s error: %s' % (rc, SMART, o, e))
logger.error(e_msg)
raise CommandException(('%s -l error /dev/%s' % (SMART, device)), o, e,
rc)
ecode_map = {
'ABRT': 'Command ABoRTed',
'AMNF': 'Address Mark Not Found',
'CCTO': 'Command Completion Timed Out',
'EOM': 'End Of Media',
'ICRC': 'Interface Cyclic Redundancy Code (CRC) error',
'IDNF': 'IDentity Not Found',
'ILI': '(packet command-set specific)',
'MC': 'Media Changed',
'MCR': 'Media Change Request',
'NM': 'No Media',
'obs': 'obsolete',
'TK0NF': 'TracK 0 Not Found',
'UNC': 'UNCorrectable Error in Data',
'WP': 'Media is Write Protected',
}
summary = {}
log_l = []
for i in range(len(o)):
if (re.match('=== START OF READ SMART DATA SECTION ===', o[i])
is not None):
err_num = None
lifetime_hours = None
state = None
etype = None
details = None
for j in range(i + 1, len(o)):
log_l.append(o[j])
if (re.match('Error ', o[j]) is not None):
fields = o[j].split()
err_num = fields[1]
if ('lifetime:' in fields):
lifetime_hours = int(fields[fields.index('lifetime:') +
1])
if (re.match(
'When the command that caused the error occurred, the device was',
o[j].strip()) is not None):
state = o[j].strip().split('the device was ')[1]
if (re.search('Error: ', o[j]) is not None):
e_substr = o[j].split('Error: ')[1]
e_fields = e_substr.split()
etype = e_fields[0]
if (etype in ecode_map):
etype = ecode_map[etype]
details = ' '.join(
e_fields[1:]) if (len(e_fields) > 1) else None
summary[err_num] = list(
[lifetime_hours, state, etype, details])
err_num = lifetime_hours = state = etype = details = None
print('summary_d %s' % summary)
return (summary, log_l)
def error_logs(device):
o, e, rc = run_command([SMART, "-l", "error", "/dev/%s" % device], throw=False)
# As we mute exceptions when calling the above command we should at least
# examine what we have as return code (rc); 64 has been seen when the error
# log contains errors but otherwise executes successfully so we catch this.
if rc == 64:
e_msg = "Drive /dev/%s has logged S.M.A.R.T errors. Please view " "the Error logs tab for this device." % device
logger.error(e_msg)
email_root("S.M.A.R.T error", e_msg)
# In all other instances that are an error (non zero) we raise exception
# as normal.
elif rc != 0:
e_msg = "non-zero code(%d) returned by command: %s -l error output: " "%s error: %s" % (rc, SMART, o, e)
logger.error(e_msg)
raise CommandException(("%s -l error /dev/%s" % (SMART, device)), o, e, rc)
ecode_map = {
"ABRT": "Command ABoRTed",
"AMNF": "Address Mark Not Found",
"CCTO": "Command Completion Timed Out",
"EOM": "End Of Media",
"ICRC": "Interface Cyclic Redundancy Code (CRC) error",
"IDNF": "IDentity Not Found",
"ILI": "(packet command-set specific)",
"MC": "Media Changed",
"MCR": "Media Change Request",
"NM": "No Media",
"obs": "obsolete",
"TK0NF": "TracK 0 Not Found",
"UNC": "UNCorrectable Error in Data",
"WP": "Media is Write Protected",
}
summary = {}
log_l = []
for i in range(len(o)):
if re.match("=== START OF READ SMART DATA SECTION ===", o[i]) is not None:
err_num = None
lifetime_hours = None
state = None
etype = None
details = None
for j in range(i + 1, len(o)):
log_l.append(o[j])
if re.match("Error ", o[j]) is not None:
fields = o[j].split()
err_num = fields[1]
if "lifetime:" in fields:
lifetime_hours = int(fields[fields.index("lifetime:") + 1])
if (
re.match("When the command that caused the error occurred, the device was", o[j].strip())
is not None
):
state = o[j].strip().split("the device was ")[1]
if re.search("Error: ", o[j]) is not None:
e_substr = o[j].split("Error: ")[1]
e_fields = e_substr.split()
etype = e_fields[0]
if etype in ecode_map:
etype = ecode_map[etype]
details = " ".join(e_fields[1:]) if (len(e_fields) > 1) else None
summary[err_num] = list([lifetime_hours, state, etype, details])
err_num = lifetime_hours = state = etype = details = None
print("summary_d %s" % summary)
return (summary, log_l)
def error_logs(device, test_mode=TESTMODE):
"""
Retrieves a parsed list of SMART errors from the output of smartctl -l error
May be empty if no errors, also returns a raw output of the error log itself
:param device: disk device name
:param test_mode: Not True causes cat from file rather than smartctl command
:return: summary: dictionary of lists containing details of error. Index is
error number.
:return: log_l: A list containing each line in turn of the error log.
"""
if not test_mode:
o, e, rc = run_command([SMART, '-l', 'error', '/dev/%s' % device],
throw=False)
else:
o, e, rc = run_command([CAT, '/root/smartdumps/smart-l-error.out'],
throw=False)
# As we mute exceptions when calling the above command we should at least
# examine what we have as return code (rc); 64 has been seen when the error
# log contains errors but otherwise executes successfully so we catch this.
if rc == 64:
e_msg = 'Drive /dev/%s has logged S.M.A.R.T errors. Please view ' \
'the Error logs tab for this device.' % device
logger.error(e_msg)
email_root('S.M.A.R.T error', e_msg)
# In all other instances that are an error (non zero) we raise exception
# as normal.
elif rc != 0:
e_msg = ('non-zero code(%d) returned by command: %s -l error output: '
'%s error: %s' % (rc, SMART, o, e))
logger.error(e_msg)
raise CommandException(('%s -l error /dev/%s' % (SMART, device)), o, e,
rc)
ecode_map = {
'ABRT' : 'Command ABoRTed',
'AMNF' : 'Address Mark Not Found',
'CCTO' : 'Command Completion Timed Out',
'EOM' : 'End Of Media',
'ICRC' : 'Interface Cyclic Redundancy Code (CRC) error',
'IDNF' : 'IDentity Not Found',
'ILI' : '(packet command-set specific)',
'MC' : 'Media Changed',
'MCR' : 'Media Change Request',
'NM' : 'No Media',
'obs' : 'obsolete',
'TK0NF' : 'TracK 0 Not Found',
'UNC' : 'UNCorrectable Error in Data',
'WP' : 'Media is Write Protected',
}
summary = {}
log_l = []
for i in range(len(o)):
if (re.match('=== START OF READ SMART DATA SECTION ===',
o[i]) is not None):
err_num = None
lifetime_hours = None
state = None
etype = None
details = None
for j in range(i + 1, len(o)):
log_l.append(o[j])
if (re.match('Error ', o[j]) is not None):
fields = o[j].split()
err_num = fields[1]
if ('lifetime:' in fields):
lifetime_hours = int(fields[fields.index('lifetime:')+1])
if (re.match('When the command that caused the error occurred, the device was', o[j].strip()) is not None):
state = o[j].strip().split('the device was ')[1]
if (re.search('Error: ', o[j]) is not None):
e_substr = o[j].split('Error: ')[1]
e_fields = e_substr.split()
etype = e_fields[0]
if (etype in ecode_map):
etype = ecode_map[etype]
details = ' '.join(e_fields[1:]) if (len(e_fields) > 1) else None
summary[err_num] = list([lifetime_hours, state, etype, details])
err_num = lifetime_hours = state = etype = details = None
print ('summary_d %s' % summary)
return (summary, log_l)