def _detect_adaptec(self, disklist, request):
# map disk name to disk mount path
adaptec_disks = {}
diskset = set(disklist)
for disk in diskset:
cmd = nagios.rootify(self._get_smartctl(request) + " -a %s" % disk)
output = commands.getoutput(cmd)
results = re.findall(r"Device: Adaptec\s+(\S+)", output)
if len(results) == 1:
adaptec_disks[results[0]] = disk
disklist.remove(disk)
# map to disknum to disk name (with disknum)
cmd = nagios.rootify("/usr/StorMan/arcconf getconfig 1")
output = commands.getoutput(cmd)
results = re.findall(r"Logical device number(?:.*\n)+?\n", output,
re.M)
diskdict = {}
for result in results:
name = re.findall(r"Logical device name\s*:\s*(.*)", result)[0]
for diskid in re.findall(r"Present \(\D*\d+,[^)]*?(\d+)\)",
result):
diskdict[diskid] = "%s-%s" % (adaptec_disks[name], diskid)
return diskdict
def check_health_status(self, request, disklist):
message = "overall test results"
status_code = nagios.Status.OK
for device_with_type in disklist:
if device_with_type:
disk = device_with_type.split()[0]
else:
continue
cmd = nagios.rootify(
self._get_smartctl(request) + " -H %s" % device_with_type)
# print "get health command: ", cmd
output = commands.getoutput(cmd)
if not self._validate_output(request, output):
continue
if "SMART support is: Unavailable" in output:
message += " %s=NOTSUPPORT" % disk
continue
test_result = re.findall(
r"(?<=SMART overall-health self-assessment test result: )(\w+)",
output)
if not test_result:
test_result = re.findall(r"(?<=SMART Health Status: )(\w+)",
output)
if not test_result:
continue
message += " %s=%s" % (disk, test_result[0])
if test_result[0] != "PASSED" and test_result[0] != "OK":
status_code = nagios.Status.CRITICAL
return status_code, message
def retrieve_batch_status(self, request):
devicelist = self._get_disks(request)
# print "devicelist: ", devicelist
disklist = [d.split()[0] for d in devicelist if d]
# print "disklist: ", disklist
stats = {}
# load the SMART info of adaptec raid controller
if request.raid == "adaptec":
stats.update(self.retrieve_adaptec_status(request, disklist))
return stats
# load the SMART info of the rest disks.
for device_with_type in devicelist:
if device_with_type:
disk = device_with_type.split()[0]
else:
continue
cmd = nagios.rootify(
self._get_smartctl(request) + (" -A %s" % device_with_type))
# print "smartctl cmd: ", cmd
output = commands.getoutput(cmd)
if not self._validate_output(request, output):
continue
for attrid, attribute in self._parse_output(request, output):
# stats.setdefault(attrid, {})[disk] = attribute
stats.setdefault(attrid, {})[device_with_type] = attribute
# print "stats: ", stats
return stats
def check_health_status(self, request, disklist):
message = "overall test results"
status_code = nagios.Status.OK
for device_with_type in disklist:
if device_with_type:
disk = device_with_type.split()[0]
else:
continue
cmd = nagios.rootify(self._get_smartctl(request) + " -H %s" % device_with_type)
# print "get health command: ", cmd
output = commands.getoutput(cmd)
if not self._validate_output(request, output):
continue
if "SMART support is: Unavailable" in output:
message += " %s=NOTSUPPORT" % disk
continue
test_result = re.findall(r"(?<=SMART overall-health self-assessment test result: )(\w+)", output)
if not test_result:
test_result = re.findall(r"(?<=SMART Health Status: )(\w+)", output)
if not test_result:
continue
message += " %s=%s" % (disk, test_result[0])
if test_result[0] != "PASSED" and test_result[0] != "OK":
status_code = nagios.Status.CRITICAL
return status_code, message
def retrieve_batch_status(self, request):
devicelist = self._get_disks(request)
# print "devicelist: ", devicelist
disklist = [d.split()[0] for d in devicelist if d]
# print "disklist: ", disklist
stats = {}
# load the SMART info of adaptec raid controller
if request.raid == "adaptec":
stats.update(self.retrieve_adaptec_status(request, disklist))
return stats
# load the SMART info of the rest disks.
for device_with_type in devicelist:
if device_with_type:
disk = device_with_type.split()[0]
else:
continue
cmd = nagios.rootify(self._get_smartctl(request) + (" -A %s" % device_with_type))
# print "smartctl cmd: ", cmd
output = commands.getoutput(cmd)
if not self._validate_output(request, output):
continue
for attrid, attribute in self._parse_output(request, output):
# stats.setdefault(attrid, {})[disk] = attribute
stats.setdefault(attrid, {})[device_with_type] = attribute
# print "stats: ", stats
return stats
def _get_disks(self, request):
if request.disk:
disklist = [request.disk]
elif request.raid == "megaraid":
# get list of OS device names (/dev/sda1, etc)
devlist = []
cmd = nagios.rootify("/sbin/fdisk -l")
output = commands.getoutput(cmd)
devlist = re.findall(r"(?<=Disk )((?:/[\w-]+)+)(?=:)", output)
#print "devlist: ", devlist
disklist = []
# get list of raid controller device ids
cmd = nagios.rootify(
'/usr/sbin/MegaCli -PDList -aALL | grep "Device Id"')
output = commands.getoutput(cmd)
#print "MegaCli Output: ", output
for line in output.split('\n'):
did = line.split(":")[1].strip()
# note that it does not matter which 'dev' you specify, as long as it's valid. There is no mapping from 'dev' to 'device ID'
disk = "%s -d sat+megaraid,%s" % (devlist[0], did)
disklist.append(disk)
else:
output = commands.getoutput(
self._get_smartctl(request) + " --scan")
if self._validate_scan_output(request, output):
disklist = []
for line in output.split("\n"):
if line:
d = line.split("#")[0].strip()
disklist.append(d)
elif sys.platform != "win32":
cmd = nagios.rootify("/sbin/fdisk -l")
output = commands.getoutput(cmd)
disklist = re.findall(r"(?<=Disk )((?:/[\w-]+)+)(?=:)", output)
else:
nagios.StatusUnknownError(request, "Can't get disk list")
return disklist
def _get_disks(self, request):
if request.disk:
disklist = [request.disk]
elif request.raid == "megaraid":
# get list of OS device names (/dev/sda1, etc)
devlist = []
cmd = nagios.rootify("/sbin/fdisk -l")
output = commands.getoutput(cmd)
devlist = re.findall(r"(?<=Disk )((?:/[\w-]+)+)(?=:)", output)
#print "devlist: ", devlist
disklist = []
# get list of raid controller device ids
cmd = nagios.rootify('/usr/sbin/MegaCli -PDList -aALL | grep "Device Id"')
output = commands.getoutput(cmd)
#print "MegaCli Output: ", output
for line in output.split('\n'):
did = line.split(":")[1].strip()
# note that it does not matter which 'dev' you specify, as long as it's valid. There is no mapping from 'dev' to 'device ID'
disk = "%s -d sat+megaraid,%s" % (devlist[0], did)
disklist.append(disk)
else:
output = commands.getoutput(self._get_smartctl(request) + " --scan")
if self._validate_scan_output(request, output):
disklist = []
for line in output.split("\n"):
if line:
d = line.split("#")[0].strip()
disklist.append(d)
elif sys.platform != "win32":
cmd = nagios.rootify("/sbin/fdisk -l")
output = commands.getoutput(cmd)
disklist = re.findall(r"(?<=Disk )((?:/[\w-]+)+)(?=:)", output)
else:
nagios.StatusUnknownError(request, "Can't get disk list")
return disklist
def _detect_adaptec(self, disklist, request):
# map disk name to disk mount path
adaptec_disks = {}
diskset = set(disklist)
for disk in diskset:
cmd = nagios.rootify(self._get_smartctl(request) + " -a %s" % disk)
output = commands.getoutput(cmd)
results = re.findall(r"Device: Adaptec\s+(\S+)", output)
if len(results) == 1:
adaptec_disks[results[0]] = disk
disklist.remove(disk)
# map to disknum to disk name (with disknum)
cmd = nagios.rootify("/usr/StorMan/arcconf getconfig 1")
output = commands.getoutput(cmd)
results = re.findall(r"Logical device number(?:.*\n)+?\n", output, re.M)
diskdict = {}
for result in results:
name = re.findall(r"Logical device name\s*:\s*(.*)", result)[0]
for diskid in re.findall(r"Present \(\D*\d+,[^)]*?(\d+)\)", result):
diskdict[diskid] = "%s-%s" % (adaptec_disks[name], diskid)
return diskdict;
def retrieve_adaptec_status(self, request, disklist):
stats = {}
diskdict = self._detect_adaptec(disklist, request)
if not diskdict:
return stats
cmd = nagios.rootify("/usr/StorMan/arcconf getsmartstats 1")
output = commands.getoutput(cmd)
xml = output[output.index("<SmartStats"):output.index("</SmartStats>")+13]
dom = parseString(xml)
for disknode in dom.getElementsByTagName("PhysicalDriveSmartStats"):
disk = diskdict[disknode.getAttribute("id")]
for attrnode in disknode.getElementsByTagName("Attribute"):
attribute = SmartAttribute()
attribute.value = attrnode.getAttribute("normalizedCurrent")
attribute.worst = attrnode.getAttribute("normalizedWorst")
attribute.raw_value = attrnode.getAttribute("rawValue")
attrid = str(int(attrnode.getAttribute("id"), 16))
stats.setdefault(attrid, {})[disk] = attribute
return stats
def get_adaptec_health(self, request):
if sys.platform == "win32":
raise nagios.StatusUnknownError(
request, "Adaptec Health only supported on linux.")
disklist = self._get_disks(request)
diskdict = self._detect_adaptec(disklist, request)
if not diskdict:
raise nagios.StatusUnknownError(
request, "No Adaptec Raid Controller detected.")
message = ""
cmd = nagios.rootify("/usr/StorMan/arcconf getlogs 1 stats")
output = commands.getoutput(cmd)
if not self._validate_arcconf_output(request, output):
return
xml = output[output.index("<ControllerLog"
):output.index("</ControllerLog>") + 16]
dom = parseString(xml)
status_code = nagios.Status.OK
if not request.warn:
warn = 1
else:
warn = request.warn
sub_perfs = []
for statsnodee in dom.getElementsByTagName("physicaldrivestats"):
value = int(statsnodee.getAttribute("smartWarnCnt"))
disk = diskdict[statsnodee.getAttribute("id")]
if value > 0:
if message == "":
message = "smart warnings:"
message += " %s=%s" % (disk, value)
status_code = self.superimpose(status_code, value, warn,
request.crit)
sub_perfs.append((disk, value))
if message == "":
message = "no smart warning"
r = nagios.Result(request.option, status_code, message,
request.appname)
for disk, value in sub_perfs:
r.add_performance_data(disk, value, warn=warn, crit=request.crit)
return r
def retrieve_adaptec_status(self, request, disklist):
stats = {}
diskdict = self._detect_adaptec(disklist, request)
if not diskdict:
return stats
cmd = nagios.rootify("/usr/StorMan/arcconf getsmartstats 1")
output = commands.getoutput(cmd)
xml = output[output.index("<SmartStats"
):output.index("</SmartStats>") + 13]
dom = parseString(xml)
for disknode in dom.getElementsByTagName("PhysicalDriveSmartStats"):
disk = diskdict[disknode.getAttribute("id")]
for attrnode in disknode.getElementsByTagName("Attribute"):
attribute = SmartAttribute()
attribute.value = attrnode.getAttribute("normalizedCurrent")
attribute.worst = attrnode.getAttribute("normalizedWorst")
attribute.raw_value = attrnode.getAttribute("rawValue")
attrid = str(int(attrnode.getAttribute("id"), 16))
stats.setdefault(attrid, {})[disk] = attribute
return stats
def get_adaptec_health(self, request):
if sys.platform == "win32":
raise nagios.StatusUnknownError(request, "Adaptec Health only supported on linux.")
disklist = self._get_disks(request)
diskdict = self._detect_adaptec(disklist, request)
if not diskdict:
raise nagios.StatusUnknownError(request, "No Adaptec Raid Controller detected.")
message = ""
cmd = nagios.rootify("/usr/StorMan/arcconf getlogs 1 stats")
output = commands.getoutput(cmd)
if not self._validate_arcconf_output(request, output):
return
xml = output[output.index("<ControllerLog"):output.index("</ControllerLog>")+16]
dom = parseString(xml)
status_code = nagios.Status.OK
if not request.warn:
warn = 1
else:
warn = request.warn
sub_perfs = []
for statsnodee in dom.getElementsByTagName("physicaldrivestats"):
value = int(statsnodee.getAttribute("smartWarnCnt"))
disk = diskdict[statsnodee.getAttribute("id")]
if value > 0:
if message == "":
message = "smart warnings:"
message += " %s=%s" % (disk, value)
status_code = self.superimpose(status_code, value, warn, request.crit)
sub_perfs.append((disk, value))
if message == "":
message = "no smart warning"
r = nagios.Result(request.option, status_code, message, request.appname)
for disk, value in sub_perfs:
r.add_performance_data(disk, value, warn=warn, crit=request.crit)
return r
