def getSmbStatus(volInfo):
status, msg, error = utils.execCmd(_checkSmbCmd)
if status == utils.PluginStatusCode.OK:
return status, "Process smb is running"
smbConfigStat, msg, error = utils.execCmd(_chkConfigSmb)
if smbConfigStat != utils.PluginStatusCode.OK:
return utils.PluginStatusCode.OK, "OK: SMB not configured"
# if smb is not running and any of the volume uses smb
# then its required to alert the user
for volume, volumeInfo in volInfo.iteritems():
if volumeInfo['volumeStatus'] == glustercli.VolumeStatus.OFFLINE:
continue
cifsStatus = volumeInfo.get('options', {}).get('user.cifs', 'enable')
smbStatus = volumeInfo.get('options', {}).get('user.smb', 'enable')
if cifsStatus == 'enable' and smbStatus == 'enable':
# SMB configured, service not running
msg = "CRITICAL: Process smb is not running"
status = utils.PluginStatusCode.CRITICAL
break
else:
# No volumes available or Non of the volumes are CIFS enabled
msg = "OK: No gluster volume uses smb"
status = utils.PluginStatusCode.OK
return status, msg
def _verifyNagiosConfig():
(rc, out, err) = utils.execCmd(
[server_utils.nagiosCmdPath.cmd, '-v', NAGIOS_CONFIG_FILE])
if rc == 0:
return True
else:
return False
def getLvs():
lvmCommand = [
"lvm", "vgs", "--unquoted", "--noheading", "--nameprefixes",
"--separator", "$", "--nosuffix", "--units", "m", "-o",
"lv_uuid,lv_name,data_percent,pool_lv,lv_attr,"
"lv_size,lv_path,lv_metadata_size,"
"metadata_percent,vg_name"
]
rc, out, err = utils.execCmd(lvmCommand)
if rc != 0:
logging.error(
"lvm command failed.\nCommand=%s\nrc=%s\nout=%s\nerr=%s" %
(lvmCommand, rc, out, err))
return None
l = map(
lambda x: dict(x),
map(lambda x: [e.split('=') for e in x],
map(lambda x: x.strip().split('$'), out)))
d = {}
for i in l:
if i['LVM2_LV_ATTR'][0] == 't':
k = "%s/%s" % (i['LVM2_VG_NAME'], i['LVM2_LV_NAME'])
else:
k = os.path.realpath(i['LVM2_LV_PATH'])
d.update({k: i})
return d
def _verifyNagiosConfig():
(rc, out, err) = utils.execCmd([server_utils.nagiosCmdPath.cmd, '-v',
NAGIOS_CONFIG_FILE])
if rc == 0:
return True
else:
return False
def execNRPECommand(host,
command,
arguments=None,
timeout=None,
json_output=True):
nrpeCmd = getNRPEBaseCommand(host, timeout).split()
nrpeCmd.append(command)
if arguments:
nrpeCmd.append('-a')
nrpeCmd.extend(arguments)
(returncode, outputStr, err) = utils.execCmd(nrpeCmd, raw=True)
if returncode == 0:
if json_output:
try:
resultDict = json.loads(outputStr)
except Exception as e:
e.args += (outputStr,)
raise
return resultDict
else:
return outputStr
else:
print "Failed to execute NRPE command '%s' in host '%s' " \
"\nError : %s" \
"Make sure NRPE server in host '%s' is configured to accept " \
"requests from Nagios server" % (command, host, outputStr, host)
sys.exit(utils.PluginStatusCode.CRITICAL)
def execNRPECommand(host,
command,
arguments=None,
timeout=None,
json_output=True):
nrpeCmd = getNRPEBaseCommand(host, timeout).split()
nrpeCmd.append(command)
if arguments:
nrpeCmd.append('-a')
nrpeCmd.extend(arguments)
(returncode, outputStr, err) = utils.execCmd(nrpeCmd, raw=True)
if returncode == 0:
if json_output:
try:
resultDict = json.loads(outputStr)
except Exception as e:
e.args += (outputStr, )
raise
return resultDict
else:
return outputStr
else:
print "Failed to execute NRPE command '%s' in host '%s' " \
"\nError : %s" \
"Make sure NRPE server in host '%s' is configured to accept " \
"requests from Nagios server" % (command, host, outputStr, host)
sys.exit(utils.PluginStatusCode.CRITICAL)
def getLvs():
lvmCommand = ["lvm", "vgs", "--unquoted", "--noheading",
"--nameprefixes", "--separator", "$",
"--nosuffix", "--units", "m", "-o",
"lv_uuid,lv_name,data_percent,pool_lv,lv_attr,"
"lv_size,lv_path,lv_metadata_size,"
"metadata_percent,vg_name"]
rc, out, err = utils.execCmd(lvmCommand)
if rc != 0:
logging.error(
"lvm command failed.\nCommand=%s\nrc=%s\nout=%s\nerr=%s"
% (lvmCommand, rc, out, err)
)
return None
l = map(lambda x: dict(x),
map(lambda x: [e.split('=') for e in x],
map(lambda x: x.strip().split('$'), out)))
d = {}
for i in l:
if i['LVM2_LV_ATTR'][0] == 't':
k = "%s/%s" % (i['LVM2_VG_NAME'], i['LVM2_LV_NAME'])
else:
k = os.path.realpath(i['LVM2_LV_PATH'])
d.update({k: i})
return d
def getGlusterdStatus():
retCode, msg, error = utils.execCmd(_checkGlusterService)
if retCode == 0:
return utils.PluginStatusCode.OK, "Process glusterd is running"
elif retCode == 3:
return utils.PluginStatusCode.CRITICAL, \
"Process glusterd is not running"
return utils.PluginStatusCode.CRITICAL, msg[0] if len(msg) > 0 else ""
def getCtdbStatus(smbStatus, nfsStatus):
# If SMB/NFS is not running, then skip ctdb check
if smbStatus != utils.PluginStatusCode.OK and \
nfsStatus != utils.PluginStatusCode.OK:
return (utils.PluginStatusCode.OK,
"CTDB ignored as SMB and NFS are not running")
status, msg, error = utils.execCmd(_checkCtdbCmd)
if status != utils.PluginStatusCode.OK:
status, msg, error = utils.execCmd(_chkConfigCtdb)
if status == utils.PluginStatusCode.OK:
return (utils.PluginStatusCode.CRITICAL,
"CTDB process is not running")
return utils.PluginStatusCode.UNKNOWN, "CTDB not configured"
# 'cdtb nodestatus' command will return the output in following format
#
# pnn:0 host_ip_address OK (THIS NODE)
#
# Possible states are -
# Ok,Disconnected,Banned,Disabled,Unhealthy,Stopped,Inactive,
# PartiallyOnline
# And combinations of them like
# pnn:0 host_ip_address BANNED|INACTIVE(THIS NODE)
#
# UNHEALTHY/DISABLED/PARTIALLYONLINE - node is partially operational
# Any other state - node in not operational
status, msg, error = utils.execCmd(['ctdb', 'nodestatus'])
if len(msg) > 0:
message = msg[0].split()
if len(message) >= 2:
msg = "Node status: %s" % message[2]
if CtdbNodeStatus.OK in message[2]:
status = utils.PluginStatusCode.OK
elif (CtdbNodeStatus.UNHEALTHY in message[2] or
CtdbNodeStatus.PARTIALLYONLINE in message[2] or
CtdbNodeStatus.DISABLED in message[2]):
status = utils.PluginStatusCode.WARNING
else:
status = utils.PluginStatusCode.CRITICAL
else:
status = utils.PluginStatusCode.UNKNOWN
return status, msg
def sadfExecCmd(sadfCmd):
try:
(rc, out, err) = utils.execCmd(sadfCmd, raw=True)
except (OSError, ValueError) as e:
raise SadfCmdExecFailedException(err=[str(e)])
if rc != 0:
raise SadfCmdExecFailedException(rc, [out], [err])
try:
return etree.fromstring(out)
except _etreeExceptions:
raise SadfXmlErrorException(err=[out])
def sadfExecCmd(sadfCmd):
try:
(rc, out, err) = utils.execCmd(sadfCmd, raw=True)
except (OSError, ValueError) as e:
raise SadfCmdExecFailedException(err=[str(e)])
if rc != 0:
raise SadfCmdExecFailedException(rc, [out], [err])
try:
return etree.fromstring(out)
except _etreeExceptions:
raise SadfXmlErrorException(err=[out])
def getShdStatus(volInfo):
status, msg, error = utils.execCmd(_checkShdCmd)
if status == utils.PluginStatusCode.OK:
return status, "Gluster Self Heal Daemon is running"
hostUuid = glustercli.hostUUIDGet()
for volumeName, volumeInfo in volInfo.iteritems():
if volumeInfo['volumeStatus'] == glustercli.VolumeStatus.OFFLINE:
continue
if hasBricks(hostUuid, volumeInfo['bricksInfo']) and \
int(volumeInfo['replicaCount']) > 1:
status = utils.PluginStatusCode.CRITICAL
msg = "CRITICAL: Gluster Self Heal Daemon not running"
break
else:
msg = "OK: Process Gluster Self Heal Daemon"
status = utils.PluginStatusCode.OK
return status, msg
def getNfsStatus(volInfo):
# if nfs is already running we need not to check further
status, msg, error = utils.execCmd(_checkNfsCmd)
if status == utils.PluginStatusCode.OK:
return status, "Process glusterfs-nfs is running"
# if nfs is not running and any of the volume uses nfs
# then its required to alert the user
for volume, volumeInfo in volInfo.iteritems():
if volumeInfo['volumeStatus'] == glustercli.VolumeStatus.OFFLINE:
continue
nfsStatus = volumeInfo.get('options', {}).get('nfs.disable', 'off')
if nfsStatus == 'off':
msg = "CRITICAL: Process glusterfs-nfs is not running"
status = utils.PluginStatusCode.CRITICAL
break
else:
msg = "OK: No gluster volume uses nfs"
status = utils.PluginStatusCode.OK
return status, msg
def getQuotadStatus(volInfo):
# if quota is already running we need not to check further
status, msg, error = utils.execCmd(_checkQuotaCmd)
if status == utils.PluginStatusCode.OK:
return status, "Process quotad is running"
# if quota is not running and any of the volume uses quota
# then the quotad process should be running in the host
for volume, volumeInfo in volInfo.iteritems():
if volumeInfo['volumeStatus'] == glustercli.VolumeStatus.OFFLINE:
continue
quotadStatus = volumeInfo.get('options', {}).get('features.quota', '')
if quotadStatus == 'on':
msg = "CRITICAL: Process quotad is not running"
utils.PluginStatusCode.CRITICAL
break
else:
msg = "OK: Quota not enabled"
status = utils.PluginStatusCode.OK
return status, msg
def getBrickStatus(volumeName, brickName):
status = None
brickPath = brickName.split(':')[1]
pidFile = brickName.replace(":/", "-").replace("/", "-") + ".pid"
try:
with open("%s/%s/run/%s" % (
_glusterVolPath, volumeName, pidFile)) as f:
try:
if _pidExists(int(f.read().strip())):
status = utils.PluginStatusCode.OK
brickDevice = storage.getBrickDeviceName(brickPath)
disk = storage.getDisksForBrick(brickDevice)
cmd = [checkIdeSmartCmdPath.cmd, "-d", disk, "-n"]
rc, out, err = utils.execCmd(cmd)
if rc == utils.PluginStatusCode.CRITICAL and \
"tests failed" in out[0]:
status = utils.PluginStatusCode.WARNING
msg = "WARNING: Brick %s: %s" % (brickPath, out[0])
else:
status = utils.PluginStatusCode.CRITICAL
except ValueError as e:
status = utils.PluginStatusCode.CRITICAL
msg = "Invalid pid of brick %s: %s" % (brickPath,
str(e))
return status, msg
except IOError as e:
if e.errno == errno.ENOENT:
status = utils.PluginStatusCode.CRITICAL
else:
status = utils.PluginStatusCode.UNKNOWN
msg = "UNKNOWN: Brick %s: %s" % (brickPath, str(e))
finally:
if status == utils.PluginStatusCode.OK:
msg = "OK: Brick %s is up" % brickPath
elif status == utils.PluginStatusCode.CRITICAL:
msg = "CRITICAL: Brick %s is down" % brickPath
return status, msg
def isNagiosRunning():
(rc, out, err) = utils.execCmd([serviceCmdPath.cmd, 'nagios', 'status'])
if rc == 0:
return True
else:
return False
def isNagiosRunning():
(rc, out, err) = utils.execCmd([serviceCmdPath.cmd, 'nagios', 'status'])
if rc == 0:
return True
else:
return False
def configureParam(paramName, value):
sed_pattern = "/%s=.*/c\\%s=%s" % (paramName, paramName, value)
command_sed = [__SED_CMD_PATH.cmd, "-i", sed_pattern, nscautils.__NAGIOSSERVER_CONF]
utils.execCmd(command_sed)
def execNRPECommand(command):
status, output, err = utils.execCmd(command.split(), raw=True)
return status, output
def _runUnknown():
(rc, out, err) = utils.execCmd(["unknown"])
def configureParam(paramName, value):
sed_pattern = '/%s=.*/c\\%s=%s' % (paramName, paramName, value)
command_sed = [
__SED_CMD_PATH.cmd, '-i', sed_pattern, nscautils.__NAGIOSSERVER_CONF
]
utils.execCmd(command_sed)
def execNRPECommand(command):
status, output, err = utils.execCmd(command.split(), raw=True)
return status, output
def testSuccess(self):
(rc, out, err) = utils.execCmd(["true"])
self.assertEquals(rc, 0)
def testFailure(self):
(rc, out, err) = utils.execCmd(["false"])
self.assertEquals(rc, 1)
