def getClusterQuorumStatus():
exitstatus = 0
message = ""
try:
volumes = glustercli.volumeInfo()
except glustercli.GlusterLockedException as e:
out = ("UNKNOWN: temporary error. %s" % '.'.join(e.err))
return utils.PluginStatusCode.UNKNOWN, out
except glustercli.GlusterCmdFailedException as e:
out = ("Quorum status could not be determined. %s"
% '.'.join(e.err))
return utils.PluginStatusCode.WARNING, out
quorumVolumes = []
for volumename, volume in volumes.iteritems():
if (volume.get('options') and
volume.get('options').get('cluster.server-quorum-type')
== "server"):
quorumVolumes.append(volumename)
if not quorumVolumes:
exitstatus = utils.PluginStatusCode.UNKNOWN
message = "Server quorum not turned on for any volume"
else:
exitstatus = utils.PluginStatusCode.OK
message = ("Server quorum turned on for %s"
% (','.join(quorumVolumes)))
return exitstatus, message
def getClusterQuorumStatus():
exitstatus = 0
message = ""
try:
volumes = glustercli.volumeInfo()
except glustercli.GlusterLockedException as e:
out = ("UNKNOWN: temporary error. %s" % '.'.join(e.err))
return utils.PluginStatusCode.UNKNOWN, out
except glustercli.GlusterCmdFailedException as e:
out = ("Quorum status could not be determined. %s" % '.'.join(e.err))
return utils.PluginStatusCode.WARNING, out
quorumVolumes = []
for volumename, volume in volumes.iteritems():
if (volume.get('options')
and volume.get('options').get('cluster.server-quorum-type')
== "server"):
quorumVolumes.append(volumename)
if not quorumVolumes:
exitstatus = utils.PluginStatusCode.UNKNOWN
message = "Server quorum not turned on for any volume"
else:
exitstatus = utils.PluginStatusCode.OK
message = ("Server quorum turned on for %s" %
(','.join(quorumVolumes)))
return exitstatus, message
def discoverVolumes(volumeName, list):
"""
This method helps to discover volumes list and volume info
Parameters
----------
list: Flag used for getting volume info. If the flag is 'True' then the
method will return only list of volume names with volume Type and doesn't
include the brick details. . If list is 'False' then returns the volume
details with brick information.
volumeName: Fetch information only for the given volume.
Note: glustercli.volumeInfo(volName) command accept a volumeName. But if
the volume name is not passed then it returns details about all the volumes
in the cluster.
Returns
---------
Returns volume details in the following dictionary format
{
'vol-name' : {vol-details}
'vol-name' : {vol-details}
'vol-name' : {vol-details}
...
}
"""
resultlist = {}
try:
volumes = glustercli.volumeInfo(volumeName)
except glustercli.GlusterLockedException as e:
resultString = ("UNKNOWN: temporary error. %s" % '.'.join(e.err))
return utils.PluginStatusCode.UNKNOWN, resultString
except glustercli.GlusterCmdFailedException as e:
resultString = ("UNKNOWN: Failed to get the volume Information. "
"%s" % '.'.join(e.err))
return utils.PluginStatusCode.UNKNOWN, resultString
for key, volume in volumes.iteritems():
volDict = {}
volDict['name'] = key
volDict['type'] = volume['volumeType']
if not list:
volOptions = volume.get('options')
if volOptions:
quotaStatus = volOptions.get('features.quota')
if quotaStatus == "on":
volDict['quota'] = quotaStatus
geoRepStatus = volOptions.get('geo-replication.indexing')
if geoRepStatus == "on":
volDict['geo-rep'] = geoRepStatus
volDict['replicaCount'] = volume['replicaCount']
volDict['bricks'] = []
volDict['disperseCount'] = volume['disperseCount']
volDict['redundancyCount'] = volume['redundancyCount']
for brick in volume['bricksInfo']:
brickproplist = brick['name'].split(':')
volDict['bricks'].append({'brickaddress': brickproplist[0],
'brickpath': brickproplist[1],
'hostUuid': brick['hostUuid']})
resultlist[key] = volDict
resultString = json.dumps(resultlist)
return utils.PluginStatusCode.OK, resultString
def getVolumeStatus(args):
exitstatus = 0
message = ""
try:
volumes = glustercli.volumeInfo(args.volume)
if volumes.get(args.volume) is None:
exitstatus = utils.PluginStatusCode.CRITICAL
message = "CRITICAL: Volume not found"
return exitstatus, message
elif volumes[args.volume]["volumeStatus"] == (glustercli.
VolumeStatus.ONLINE):
exitstatus = utils.PluginStatusCode.OK
message = "OK: Volume : %s type - Volume is up" % \
(volumes[args.volume]["volumeType"])
elif volumes[args.volume]["volumeStatus"] == (glustercli.
VolumeStatus.OFFLINE):
exitstatus = utils.PluginStatusCode.CRITICAL
message = "CRITICAL: Volume : %s type is stopped" % \
(volumes[args.volume]["volumeType"])
except glustercli.GlusterLockedException as e:
out = ("UNKNOWN: Glusterd cannot be queried. %s" % '.'.join(e.err))
return utils.PluginStatusCode.UNKNOWN, out
except glustercli.GlusterCmdFailedException as e:
out = ("WARNING: Command execution failed. %s" % '.'.join(e.err))
return utils.PluginStatusCode.WARNING, out
return exitstatus, message
def getVolumeStatus(args):
exitstatus = 0
message = ""
try:
volumes = glustercli.volumeInfo(args.volume)
if volumes.get(args.volume) is None:
exitstatus = utils.PluginStatusCode.CRITICAL
message = "CRITICAL: Volume not found"
return exitstatus, message
elif volumes[args.volume]["volumeStatus"] == (
glustercli.VolumeStatus.ONLINE):
exitstatus = utils.PluginStatusCode.OK
message = "OK: Volume : %s type - Volume is up" % \
(volumes[args.volume]["volumeType"])
elif volumes[args.volume]["volumeStatus"] == (
glustercli.VolumeStatus.OFFLINE):
exitstatus = utils.PluginStatusCode.CRITICAL
message = "CRITICAL: Volume : %s type is stopped" % \
(volumes[args.volume]["volumeType"])
except glustercli.GlusterLockedException as e:
out = ("UNKNOWN: Glusterd cannot be queried. %s" % '.'.join(e.err))
return utils.PluginStatusCode.UNKNOWN, out
except glustercli.GlusterCmdFailedException as e:
out = ("WARNING: Command execution failed. %s" % '.'.join(e.err))
return utils.PluginStatusCode.WARNING, out
return exitstatus, message
def check_volume_status(volume):
try:
volumes = glustercli.volumeInfo(volume)
if volumes.get(volume) is None:
sys.stdout.write("CRITICAL: Volume not found\n")
sys.exit(utils.PluginStatusCode.CRITICAL)
elif volumes[volume]["volumeStatus"] == \
glustercli.VolumeStatus.OFFLINE:
sys.stdout.write("CRITICAL: Volume is stopped\n")
sys.exit(utils.PluginStatusCode.CRITICAL)
except glustercli.GlusterCmdFailedException:
sys.stdout.write("UNKNOWN: Failed to get the "
"Volume Utilization Data\n")
sys.exit(utils.PluginStatusCode.UNKNOWN)
def check_volume_status(volume):
try:
volumes = glustercli.volumeInfo(volume)
if volumes.get(volume) is None:
sys.stdout.write("CRITICAL: Volume not found\n")
sys.exit(utils.PluginStatusCode.CRITICAL)
elif volumes[volume]["volumeStatus"] == \
glustercli.VolumeStatus.OFFLINE:
sys.stdout.write("CRITICAL: Volume is stopped\n")
sys.exit(utils.PluginStatusCode.CRITICAL)
except glustercli.GlusterCmdFailedException:
sys.stdout.write("UNKNOWN: Failed to get the "
"Volume Utilization Data\n")
sys.exit(utils.PluginStatusCode.UNKNOWN)
def run(self):
hostName = nscautils.getCurrentHostNameInNagiosServer()
sleepTime = int(nscautils.getProcessMonitorSleepTime())
glusterdStatus = Status()
nfsStatus = Status()
smbStatus = Status()
shdStatus = Status()
quotaStatus = Status()
ctdbStatus = Status()
brickStatus = {}
while True:
if not hostName:
hostName = nscautils.getCurrentHostNameInNagiosServer()
if not hostName:
logger.warn("'hostname_in_nagios' is not configured "
"in %s" % nagios_server_conf_path)
time.sleep(sleepTime)
continue
status, msg = check_proc_util.getGlusterdStatus()
if glusterdStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _glusterdService, status, msg)
# Get the volume status only if glusterfs is running to avoid
# unusual delay
if status != utils.PluginStatusCode.OK:
logger.warn("Glusterd is not running")
time.sleep(sleepTime)
continue
try:
volInfo = glustercli.volumeInfo()
except glusternagios.glustercli.GlusterCmdFailedException:
logger.error("failed to find volume info")
time.sleep(sleepTime)
continue
status, msg = check_proc_util.getNfsStatus(volInfo)
if nfsStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _nfsService, status, msg)
status, msg = check_proc_util.getSmbStatus(volInfo)
if smbStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _smbService, status, msg)
status, msg = check_proc_util.getCtdbStatus(
smbStatus.code, nfsStatus.code)
if ctdbStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _ctdbdService, status, msg)
status, msg = check_proc_util.getShdStatus(volInfo)
if shdStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _shdService, status, msg)
status, msg = check_proc_util.getQuotadStatus(volInfo)
if quotaStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _quotadService, status, msg)
brick = getBrickStatus(volInfo)
# brickInfo contains status, and message
for brickService, brickInfo in brick.iteritems():
if brickInfo != brickStatus.get(brickService, [None]) \
or brickInfo[0] == utils.PluginStatusCode.CRITICAL:
brickStatus[brickService] = brickInfo
nscautils.send_to_nsca(hostName, brickService,
brickInfo[0], brickInfo[1])
time.sleep(sleepTime)
def _findBrickName(volInfo, brickPath):
hostUuid = glustercli.hostUUIDGet()
for volumeName, volumeInfo in volInfo.iteritems():
for brick in volumeInfo['bricksInfo']:
if brick.get('hostUuid') == hostUuid \
and brick['name'].split(':')[1] == brickPath:
return brick['name']
if __name__ == '__main__':
args = parse_input()
status, msg = check_proc_util.getGlusterdStatus()
if status == utils.PluginStatusCode.OK:
if args.type == _NFS:
status, msg = check_proc_util.getNfsStatus(glustercli.volumeInfo())
elif args.type == _SMB:
status, msg = check_proc_util.getSmbStatus(glustercli.volumeInfo())
elif args.type == _SHD:
status, msg = check_proc_util.getShdStatus(glustercli.volumeInfo())
elif args.type == _QUOTA:
status, msg = check_proc_util.getQuotadStatus(
glustercli.volumeInfo())
elif args.type == _CTDB:
volInfo = glustercli.volumeInfo()
nfsStatus, nfsMsg = check_proc_util.getNfsStatus(volInfo)
smbStatus, smbMsg = check_proc_util.getSmbStatus(volInfo)
status, msg = check_proc_util.getCtdbStatus(smbStatus, nfsStatus)
elif args.type == _BRICK:
volInfo = glustercli.volumeInfo(args.volume)
brickName = _findBrickName(volInfo, args.brickPath)
def run(self):
hostName = nscautils.getCurrentHostNameInNagiosServer()
sleepTime = int(nscautils.getProcessMonitorSleepTime())
glusterdStatus = Status()
nfsStatus = Status()
smbStatus = Status()
shdStatus = Status()
quotaStatus = Status()
ctdbStatus = Status()
brickStatus = {}
while True:
if not hostName:
hostName = nscautils.getCurrentHostNameInNagiosServer()
if not hostName:
logger.warn("'hostname_in_nagios' is not configured "
"in %s" % nagios_server_conf_path)
time.sleep(sleepTime)
continue
status, msg = check_proc_util.getGlusterdStatus()
if glusterdStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _glusterdService, status, msg)
# Get the volume status only if glusterfs is running to avoid
# unusual delay
if status != utils.PluginStatusCode.OK:
logger.warn("Glusterd is not running")
time.sleep(sleepTime)
continue
try:
volInfo = glustercli.volumeInfo()
except glusternagios.glustercli.GlusterCmdFailedException:
logger.error("failed to find volume info")
time.sleep(sleepTime)
continue
status, msg = check_proc_util.getNfsStatus(volInfo)
if nfsStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _nfsService, status, msg)
status, msg = check_proc_util.getSmbStatus(volInfo)
if smbStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _smbService, status, msg)
status, msg = check_proc_util.getCtdbStatus(smbStatus.code,
nfsStatus.code)
if ctdbStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _ctdbdService, status, msg)
status, msg = check_proc_util.getShdStatus(volInfo)
if shdStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _shdService, status, msg)
status, msg = check_proc_util.getQuotadStatus(volInfo)
if quotaStatus.isStatusChanged(status, msg):
nscautils.send_to_nsca(hostName, _quotadService, status, msg)
brick = getBrickStatus(volInfo)
# brickInfo contains status, and message
for brickService, brickInfo in brick.iteritems():
if brickInfo != brickStatus.get(brickService, [None]) \
or brickInfo[0] == utils.PluginStatusCode.CRITICAL:
brickStatus[brickService] = brickInfo
nscautils.send_to_nsca(hostName, brickService,
brickInfo[0], brickInfo[1])
time.sleep(sleepTime)
