bonesmoses/nagaconda

Python Nagios Integration

nagaconda's Issues

NagAconda install throwing error

C:\Python33>"c:\python33\scripts\pip.exe" install NagAconda
Downloading/unpacking NagAconda
Downloading NagAconda-0.2.1.tar.gz (235kB): 235kB downloaded
Running setup.py egg_info for package NagAconda
Traceback (most recent call last):
File "", line 16, in
File "c:\users\rajars~1\appdata\local\temp\pip_build_rajarshid\NagAconda\setup.py", line 4, in
import NagAconda
File ".\NagAconda__init__.py", line 19, in
from Plugin import *
ImportError: No module named 'Plugin'
Complete output from command python setup.py egg_info:
Traceback (most recent call last):

File "", line 16, in

File "c:\users\rajars~1\appdata\local\temp\pip_build_rajarshid\NagAconda\setup.py", line 4, in

import NagAconda

File ".\NagAconda__init__.py", line 19, in

from Plugin import *

ImportError: No module named 'Plugin'

---

Cleaning up...
Command python setup.py egg_info failed with error code 1 in c:\users\rajars~1\appdata\local\temp\pip_build_rajarshid\NagAconda
Storing complete log in C:\Users\rajarshid\pip\pip.log

C:\Python33>pause
Press any key to continue . . .

Upload version 0.2.0 to PyPi

Currently only NagAconda 0.1.4 is on PyPi:
https://pypi.python.org/pypi?%3Aaction=search&term=NagAconda

Could you upload a package for version 0.2.0?

Link to documentation in README on github

It would be nice to have the link to PyPi https://pypi.python.org/pypi/NagAconda/0.1.4 and the online documentation http://pythonhosted.org//NagAconda/ here.

Extra print statements cause problems when alerting to SMS

When I am sending Nagios alerts to an SMS recipient, the extra print statements that show the range, etc make it so that I lose much of the content of the alert. Would it be possible to set a flag for whether or not to send them? I can fork and upload my own modified copy of Plugin.py for reference if needed (I just commented out many of the print statements). Thanks!

Can't set range

If you run this code, you will have an error.

#!/usr/bin/python

from NagAconda import Plugin

plugin = Plugin('The plugin', '0.1')
plugin.add_option('p', 'parameter', 'The parameter.')
plugin.enable_status('warning')
plugin.enable_status('critical')
plugin.start()

#have to huck the code due to a bug. It sets defaults for warning and critical to None and then try to get its len :(
# plugin._Plugin__warning = []
# plugin._Plugin__critical = []
#end of huck

plugin.set_range('warning', 1)
plugin.set_range('critical', 2, 1)


plugin.set_status_message('the message')
plugin.set_value('state', 1)

plugin.finish()

Traceback (most recent call last):
  File "nagaconda_bug.py", line 16, in <module>
    plugin.set_range('warning', 1)
  File "/Library/Python/2.6/site-packages/NagAconda/Plugin.py", line 353, in set_range
    if range_num > len(my_range):
TypeError: object of type 'NoneType' has no len()

Due to info here you can set range after start. Uncomment "huck" code it make it run.

Plugin output is multiline

Is there any way to get just the last line (see below) from the plugin output? The check works fine and Icinga isn't complaining but "output of the check plugin" field only contains "check warning" instead of "Status Ok, Records in the last 50 minute(s): 18321011".

Here's my plugin output:

checking warning
range was warning
(0.0, 250000.0, True)
18321011.0:0.0:250000.0 = True
checking critical
range was critical
(0.0, 500000.0, True)
18321011.0:0.0:500000.0 = True
Status Ok, Records in the last 50 minute(s): 18321011|num_records=18321011.0;250000;500000;;

Ranges for performance data

I wrote a little nagios_plugin to check btrfs

#!/usr/bin/python
# Von https://pythonhosted.org/NagAconda/plugin.html
from NagAconda import Plugin
import os
import subprocess
from pyparsing import *

btrfs_check = Plugin("Plugin to show disk usage of btrfs.", "0.1")
btrfs_check.add_option('m', 'mountpoint', 'mountpoint for btrfs',
    required=True)

btrfs_check.enable_status('warning')
btrfs_check.enable_status('critical')
btrfs_check.start()

btrfs_output = subprocess.check_output(["btrfs", "fi", "df", btrfs_check.options.mountpoint])

# DEBUG: 
# print btrfs_output

# PyParsing definitions
#
# Output is something like:
# Data, RAID1: total=222.72GB, used=197.44GB
# System, RAID1: total=64.00MB, used=40.00KB
# System: total=4.00MB, used=0.00
# Metadata, RAID1: total=10.00GB, used=5.40GB

# Parse Byte values with units
byteDefs=["B", "KB", "MB", "GB", "TB", "PB", "EB"]
byteDef = oneOf(byteDefs)
fnumber = Word(nums+'.').setParseAction(lambda t: float(t[0]))
bnumber = Group(fnumber('num') + Optional(byteDef('unit'), default="B"))

# Definitions for each row
row_data = "Data, " + "RAID1: " + "total=" + bnumber('bytesize_total') + "," + "used=" + bnumber('bytesize_used')
row_system1 = "System, " + "RAID1: " + "total=" + bnumber('bytesize_total') + "," + "used=" + bnumber('bytesize_used')
row_system2 = "System: " + "total=" + bnumber('bytesize_total') + "," + "used=" + bnumber('bytesize_used')
row_metadata = "Metadata, " + "RAID1: " + "total=" + bnumber('bytesize_total') + "," + "used=" + bnumber('bytesize_used')

# The whole parsing term
btrfs_output_parser = row_data('data') + row_system1('system1') + row_system2('system2') + row_metadata('metadata')

# Parse btrfs output
btrfs_output_parsed = btrfs_output_parser.parseString(btrfs_output)

# calculate values
byteUnits=[("B", 1), ("KB", 1000), ("MB", 1000000), ("GB", 1000000000), ("TB", 1000000000000), ("PB", 1000000000000000), ("EB", 1000000000000000000)]
def calc_bnumber(bnum):
        """Get float number of bytes for something like 2GB"""
        for byteSize in byteUnits:
                if byteSize[0] == bnum.unit:
                        return byteSize[1] * bnum.num
        return None

data_bytes_used_float = calc_bnumber(btrfs_output_parsed.data.bytesize_used)
data_bytes_size_float = calc_bnumber(btrfs_output_parsed.data.bytesize_total)
data_usage_percentage_float = data_bytes_used_float / data_bytes_size_float * 100.0

system1_bytes_used_float = calc_bnumber(btrfs_output_parsed.system1.bytesize_used)
system1_bytes_size_float = calc_bnumber(btrfs_output_parsed.system1.bytesize_total)

system2_bytes_used_float = calc_bnumber(btrfs_output_parsed.system2.bytesize_used)
system2_bytes_size_float = calc_bnumber(btrfs_output_parsed.system2.bytesize_total)

metadata_bytes_used_float = calc_bnumber(btrfs_output_parsed.metadata.bytesize_used)

# DEBUG:
# print btrfs_output_parsed.dump()

# set nagios output
btrfs_check.set_range('warning', 100000000000000000000000, range_num=2)
btrfs_check.set_range('critical', 200000000000000000000000, range_num=2)


btrfs_check.set_value("data_used", btrfs_output_parsed.data.bytesize_used.num, scale=btrfs_output_parsed.data.bytesize_used.unit, threshold=2)
btrfs_check.set_value("data_total", btrfs_output_parsed.data.bytesize_total.num, scale=btrfs_output_parsed.data.bytesize_total.unit, threshold=2)
btrfs_check.set_value("data_ratio", data_usage_percentage_float, scale="%", threshold=1)

btrfs_check.set_value("system1_used", btrfs_output_parsed.system1.bytesize_used.num, scale=btrfs_output_parsed.system1.bytesize_used.unit, threshold=2)
btrfs_check.set_value("system1_total", btrfs_output_parsed.system1.bytesize_total.num, scale=btrfs_output_parsed.system1.bytesize_total.unit, threshold=2)

btrfs_check.set_value("system2_used", btrfs_output_parsed.system2.bytesize_used.num, scale=btrfs_output_parsed.system2.bytesize_used.unit, threshold=2)
btrfs_check.set_value("system2_total", btrfs_output_parsed.system2.bytesize_total.num, scale=btrfs_output_parsed.system2.bytesize_total.unit, threshold=2)

btrfs_check.set_value("metadata_used", btrfs_output_parsed.metadata.bytesize_used.num, scale=btrfs_output_parsed.metadata.bytesize_used.unit, threshold=2)
btrfs_check.set_value("metadata_total", btrfs_output_parsed.metadata.bytesize_total.num, scale=btrfs_output_parsed.metadata.bytesize_total.unit, threshold=2)

btrfs_check.finish()

I'd like to only check the data_ratio with the standard warning and critical values. This actually works when adding a second range that all the other performance datas are checked against.
But the output doesn't care about this.

When being called by

nagaconda_btrfs_df.py -m /mounts/BackupRAID -w0.9 -c0.95

Status Ok|metadata_used=5400000000.0B;0.9;0.95;; data_total=222.72GB;0.9;0.95;; data_ratio=0.886494252874;0.9;0.95;; data_used=197.44GB;0.9;0.95;; system2_used=0.0B;0.9;0.95;; metadata_total=10000000000.0B;0.9;0.95;; system1_used=40.0KB;0.9;0.95;; system2_total=4000000.0B;0.9;0.95;; system1_total=64.0MB;0.9;0.95;;

I'd expect the 2nd threshold not only in the tests but also in the output