forked from willixix/naglio-plugins
-
Notifications
You must be signed in to change notification settings - Fork 0
/
check_redis.pl
executable file
·2914 lines (2762 loc) · 138 KB
/
check_redis.pl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#!/usr/bin/perl -w
#
# ============================== SUMMARY =====================================
#
# Program : check_redis.pl
# Version : 0.73
# Date : Mar 23, 2013
# Author : William Leibzon - william@leibzon.org
# Licence : GPL - summary below, full text at http://www.fsf.org/licenses/gpl.txt
#
# =========================== PROGRAM LICENSE =================================
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
# ===================== INFORMATION ABOUT THIS PLUGIN =========================
#
# This is Redis Server Check plugin. It gets stats variables and allows to set
# thresholds on their value or their rate of change. It can measure response time,
# hitrate, memory utilization, check replication sync and more. It can also test
# data in a specified key (if necessary doing average or sum on range).
#
# Plugin returns stats variables as performance data for further nagios 2.0
# post-processing, you can find graph templates for PNP4Nagios at:
# http://william.leibzon.org/nagios/
#
# This program is written and maintained by:
# William Leibzon - william(at)leibzon.org
#
# ============================= SETUP NOTES ====================================
#
# Make sure to install Redis perl library from CPAN first.
#
# Next for help and to see what parameters this plugin accepts do:
# ./check_redis.pl --help
#
# This plugin checks Redis NoSQL database status variables, measures its response
# time and if specified allows to set thresholds on one or more key data. You can
# set thresholds for data in stats variables and some of them are also conveniently
# available as long options with special threshold syntax. Plugin also calculates
# statistics such as Hitrate (calculated as rate of change of hits/misses) and
# memory use and can check replication delay.
#
# All variables can be returned as performance data for graphing and pnp4nagios
# template should be available with this plugin on the site you downloaded it from.
# 1. Connection Parameters
#
# The connection parameters are "-H hostname", "-p port", "-D database" and
# "-C password_file" or "-x password". Specifying hostname is required, if you
# run locally specify it as -H 127.0.0.1. Everything else is optional and rarely
# needed. Default port is 6337. Database name (usually a numeric id) is probably
# only needed if you use --query option. Password can be passed on a command
# line with -x but its safer to read read it from a file or change in the code
# itself if you do use authentication.
#
# 2. Response Time, HitRate, Memory Utilization, Replication Delay
#
# To get response time you use "-T" or "--response_time=" option. By itself
# it will cause output of response time at the status line. You can also use
# it as "-T warn,crit" to specify warning and critical thresholds.
#
# To get hitrate the option is "-R" or "--hitrate=". If previous performance
# data is not feed to plugin (-P option, see below) the plugin calculates
# it as total hitrate over life of redis process. If -P is specified and
# previous performance data is fed back, the data is based on real hitrate
# (which can show spikes and downs) with lifelong info also given in paranthesis
# The data is based on keyspace_hits and keyspace_misses stats variables.
# As with -T you can specify -R by itself or with thresholds as -R warn,crit
#
# Memory utilization is percent of real memory used by Redis out of total
# memory on the system. To be able to calculate it plugin needs to known
# amount of memory your system has which you specify with "-M" or "--total_memory="
# option. Memory utilization option itself is lower "-m" or "--memory_utilization="
# and you can specify threshold for it as "-m warn,crit"
#
# Replication delay threshold option "-R" or "--replication_delay=" is used
# to check replication with data from "master_last_io_seconds_ago" stats and
# valid only on slave servers. Other variables maybe checked for this later
# with more complex functionality, so it was chosen to do this as separate
# option rather than directing people to check that variable.
#
# 3. Checks on Redis Status Variables
#
# All status variables from redis can be checked with the plugin. For some
# status variables separate long option is provided to specify threshold.
# i.e. --connected_clients=<thresholds>
#
# This is a new alternative to specifying all variables together with -a
# (--variables) option. For example:
# -a connected_clients,blocked_clients
# When you do above results are included in status output line and you
# are required to specify thresholds with -w or --warn and -c or --crit
# with exactly number of thresholds as a number of variables specified
# in -a. If you simply want variable values on status line without specifying
# any threshold, use ~ in place of threshold value or skip value but specify
# all appropriate commas. For example:
# -a connected_clients,blocked_clients -w ~,~ -c ~,~
# OR -a connected_clients,blocked_clients -w , -c ,
#
# If you use new syntax with a long option for specific stats variables, you
# can specify list of one or more threshold specifiers which can be any of:
# NAME:<string> - Overrides name for this variable for use in status and PERF output
# PATTERN:<regex> - Regular Expression that allows to match multiple data results
# WARN:threshold - warning alert threshold
# CRIT:threshold - critical alert threshold
# Threshold is a value (usually numeric) which may have the following prefix:
# > - warn if data is above this value (default for numeric values)
# < - warn if data is below this value (must be followed by number)
# = - warn if data is equal to this value (default for non-numeric values)
# ! - warn if data is not equal to this value
# Threshold can also be specified as a range in two forms:
# num1:num2 - warn if data is outside range i.e. if data<num1 or data>num2
# \@num1:num2 - warn if data is in range i.e. data>=num1 && data<=num2
# ABSENT:OK|WARNING|CRITICAL|UNKNOWN - Nagios alert (or lock of thereof) if data is absent
# ZERO:OK|WARNING|CRITICAL|UNKNOWN - Nagios alert (or lock of thereof) if result is 0
# DISPLAY:YES|NO - Specifies if data should be included in nagios status line output
# PERF:YES|NO - Output in performance data or not (always YES if -F option is used)
# UOM:<string> - Unit Of Measurement symbol to add to perf data - 'c','%','s','B'
# This is used by programs that graph perf data such as PNP
#
# These can be specified in any order separated by ",". For example:
# --connected_clients=CRIT:>100,WARN:>50,ABSENT:CRITICAL,ZERO:OK,DISPLAY:YES,PERF:YES
#
# Variables that are not known to plugin and don't have specific long option (or even if
# they do) can be specified using general long option --check or --option or -o
# (all are aliases for same option):
# --check=NAME:connected_clients,CRIT:>100,WARN:>50,ABSENT:CRITICAL,DISPLAY:YES,PERF:YES
#
# Then NAME is used to specify what to match and multiple data vars maybe matched
# with PATTERN regex option (and please only use PATTERN with --check and not confuse
# plugin by using it in a named long option). Either NAME or PATTERN are required.
#
# 4. Calculating and using Rate of Change for Variables
#
# If you want to check rate of change rather than actual value you can do this
# by specifying it as '&variable' such as "&total_connections_received" or
# as "variable_rate" which is "total_connections_received_rate" and is similar
# to 'connected_clients' variable. By default it would be reported in the output
# as 'variable_rate' though '&variable' is a format used internally by plugin.
#
# As an alternative you can specify how to label these with --rate_label
# option where you can specify prefix and/or suffix. For example '--rate_label=dt_'
# would have the output being "dt_total_connections_received' where as
# '--rate_label=,_rate' is plugin default giving 'total_connections_received_rate'.
# You can use these names with -a and -A such as:
# --rate_label=,_rate -a total_connections_received_rate -w 1000 -c ~
# Note that --rate_label will not work with new variable-named options, the
# only way to change default if you use that is to modify code and change
# $o_rprefix and $o_rsuffix variables default values.
#
# Now in order to be able to calculate rate of change, the plugin needs to
# know values of the variables from when it was run the last time. This
# is done by feeding it previous performance data with a -P option.
# In commands.cfg this would be specified as:
# -P "$SERVICEPERFDATA$"
# And don't forget the quotes, in this case they are not just for documentation.
#
# 5. Threshold Specification
#
# The plugin fully supports Nagios plug-in specification for specifying thresholds:
# http://nagiosplug.sourceforge.net/developer-guidelines.html#THRESHOLDFORMAT
#
# And it supports an easier format with the following one-letter prefix modifiers:
# >value : issue alert if data is above this value (default for numeric value)
# <value : issue alert if data is below this value (must be followed by number)
# =value : issue alert if data is equal to this value (default for non-numeric)
# !value : issue alert if data is NOT equal to this value
# (because > and < are interpreted by shell you may need to specify this in quotes)
# There are also two specifications of range formats as with other nagios plugins:
# number1:number2 issue alert if data is OUTSIDE of range [number1..number2]
# i.e. alert if data<$number1 or data>$number2
# @number1:number2 issue alert if data is WITHIN range [number1..number2]
# i.e. alert if data>=$number and $data<=$number2
#
# The plugin will attempt to check that WARNING value is less than CRITICAL
# (or greater for <). A special prefix modifier '^' can be used to disable these
# checks. A quick example of such special use is '--warn=^<100 --crit=>200' which
# means warning alert if value is < 100 and critical alert if its greater than 200.
#
# 6. Performance Data
#
# With '-f' option values of all variables you specified in -a as well as
# response time from -T (response time),
# hitrate from -R,
# memory utilization from -m
# and other data are reported back out as performance data for Nagios graphing programs.
#
# You may also directly specify which variables are to be return as performance data
# with '-A' option. If you use '-A' by itself and not specify any variables or use
# special value of '*' (as in '-A *') the plugin will output all variables which is useful
# for finding what data you can check with this plugin.
#
# The plugin will output threshold values as part of performance data as specified at
# http://nagiosplug.sourceforge.net/developer-guidelines.html#AEN201
# And don't worry about using non-standard >,<,=,~ prefixes, all of that would get
# converted into nagios threshold format for performance output
#
# The plugin is smart enough to add 'c' suffix for known COUNTER variables to
# values in performance data. Known variables are specified in an array you can
# find at the top of the code (further below) and plugin author does not claim
# to have identified all variables correctly. Please email if you find an error
# or want to add more variables.
#
# As noted above performance data is also used to calculate rate of change
# by feeding it back with -P option. In that regard even if you did not specify
# -f or -A but you have specified &variable, its actual data would be sent out
# in performance output. Additionally last time plugin was run is also in
# performance data as special _ptime variable.
#
# 7. Query Option and setting thresholds for data in Redis Database
#
# With -q (--query) option the plugin can retrieve data from Redis database
# which become new variables you can then check thresholds on. Currently it
# supports getting single key values with GET and getting range or values (or
# everything in list) with LRANGE and finding their Average or Min or Max or Sum.
# The option maybe repeated more than once. The format for this option is:
#
# -q, --query=query_type,key[:varname]<,list of threshold specifiers>
#
# query_type is one of:
# GET - get one string value
# LLEN - returns number of items in a list
# LRANGE:AVG:start:end - retrieve list and average results
# LRANGE:SUM:start:end - retrieve list and sum results
# LRANGE:MIN:start:end - retrieve list and return minimum
# LRANGE:MAX:start:end - retrieve list and return maximum
# HLEN - returns number of items in a hash [TODO]
# HGET:name - get specific hash key 'name' [TODO]
# HEXISTS:name - returns 0 or 1 depending on if specified hash key 'name' exists [TODO]
# SLEN - returns number of items in a set [TODO, SCARD redis opp]
# SEXISTS:name - returns 0 or 1 depending on if set member 'name' exists [SISMEMBER, TODO]
# ZLEN - returns number of items in a sorted set [TODO, ZCARD redis opp]
# ZCOUNT:min:max - counts number of items in sorted set with scores within the given values
# ZRANGE:AVG:min:max - retrieve sorted set members from min to max and average results
# ZRANGE:SUM:min:max - retrieve sorted set members from min to max and sum results
# ZRANGE:MIN:min:max - retrieve sorted set members from min to max list and return minimum
# ZRANGE:MAX:min:max- retrieve sorted set members from min to max and return maximum
# For LRANGE if you do not specify start and end, then start will be 0 and end
# is last value in the list pointed to by this key (found by using llen).
#
# Key is the Redis key name to be retrieved and optionally you can add ":varname"
# after it which specifies what to name plugin variable based on this data -
# based on what you specify here is how it will be displayed in the status
# line and performance data, default is same as Redis key name.
#
# After these key name you specify list of thresholds in the same format as
# variable-based long options described in section 3. Again the list of the
# possible specifiers are:
# WARN:threshold
# CRIT:threshold
# ABSENT:OK|WARNING|CRITICAL|UNKNOWN - what to do if data is not available
# ZERO:OK|WARNING|CRIICAL|UNKNOWN - what to do if data is 0 (rarely needed)
# DISPLAY:YES|NO - display on status line or not (default YES)
# PERF:YES|NO - output in perf data or not
#
# You can also optionally use -a, -w and -c to check data from the query instead
# of specifying thresholds as part of query option itself And remember that you if
# you need to check multiple keys you just repeat --query option more than once.
#
# 8. Example of Nagios Config Definitions
#
# Sample command and service definitions are below:
#
# define command {
# command_name check_redis_new
# command_line $USER1$/check_redis.pl -H $HOSTADDRESS$ -p $ARG1$ -T $ARG2$ -R -A -M $_HOSTSYSTEM_MEMORY$ -m $ARG3$ -a $ARG4$ -w $ARG5$ -c $ARG6$ -f -P "$SERVICEPERFDATA$"
# }
#
# Arguments and thresholds are:
# $ARG1 : Port
# $ARG2 : response time thresholds
# $ARG3 : memory utilization thresholds
# $ARG4 : additional variables to be checked
# $ARG5 : warning thresholds for those variables
# $ARG6 : critical thresholds for those variables
#
# define service {
# use prod-service
# hostgroups redishosts
# service_description Redis
# check_command check_redis_new!6379!"1,2"!"80,90"!blocked_clients,connected_clients!50,~!100,~
# }
#
# define host {
# use prod-server
# host_name redis.mynetwork
# address redis.mynetwork
# alias Redis Stat Server
# hostgroups linux,redishosts
# _SYSTEM_MEMORY '8G'
# }
#
# Example of command-line use:
# /usr/lib/nagios/plugins/check_redis.pl -H localhost -a 'connected_clients,blocked_clients' -w ~,~ -c ~,~ -m -M 4G -A -R -T -f -v
#
# In above the -v option means "verbose" and with it plugin will output some debugging information
# about what it is doing. The option is not intended to be used when plugin is called from nagios itself.
#
# Example of using query and variable-based long options with debug enabled as well (-v):
#
# ./check_redis.pl -H localhost -p 6379 -D 1 --query LRANGE:AVG:0:,MyColumn1:Q1,ABSENT:WARNING,WARN:300,CRIT:500,DISPLAY:YES,PERF:NO
# --query GET,MyKey:K1,ABSENT:CRITICAL "--connected_clients=WARN:<2,CRIT:>100,ZERO:OK,ABSENT:WARNING,DISPLAY:YES,PERF:YES"
#
# ======================= VERSION HISTORY and TODO ================================
#
# The plugins is written by reusing code my check_memcached.pl which itself is based
# on check_mysqld.pl. check_mysqld.pl has history going back to 2004.
#
# [0.4 - Mar 2012] First version of the code based on check_mysqld.pl 0.93
# and check_memcached.pl 0.6. Internal work, not released.
# Version 0.4 because its based on a well developed code base
# [0.41 - Apr 15, 2012] Added list of variables array and perf_ok regex.
# Still testing internally and not released yet.
# [0.42 - Apr 28, 2012] Added total_keys, total_expires, nice uptime_info
# and memory utilization
# [0.43 - May 31, 2012] Release candidate. More documentation added
# replacing check_memcached examples. Bugs fixed.
# Made "_rate" as default rate variables suffix in
# place of &delta. Changed -D option to -r.
#
# [0.5 - Jun 01, 2012] First official release will start with version 0.5
# Documentation changes, but no code updates.
# [0.51 - Jun 16, 2012] Added support to specify filename to '-v' option
# for debug output and '--debug' as alias to '--verbose'
# [0.52 - Jul 10, 2012] Patch by Jon Schulz to support credentials with -C
# (credentials file) and addition by me to support
# password as command argument.
# [0.53 - Jul 15, 2012] Adding special option to do query on one redis key and
# and do threshold checking of results if its numeric
#
# [0.6 - Jul 17, 2012] Rewrote parts of thresholds checking code and moved code
# that checks and parses thresholds from main into separate
# functions that are to become part of plugin library.
# Added support for variable thresholds specified as:
# option=WARN:threshold,CRIT:threshold,ABSENT:OK|WARNING|CRITICAL,ZERO:..
# which are to be used for stats-variable based long options such as
# --connected_clients=WARN:threshold,CRIT:threshold
# and added DISPLAY:YES|NO and PERF specifiers for above too.
# Added -D option to specify database needed for --query
# [0.61 - Aug 03, 2012] Added more types of key query for lists, sets, hashes
# and options to find number of elements in a list/set/hash.
# New options added are:
# LLEN,HLEN,SLEN,ZLEN,HGET,HEXISTS,SEXISTS,ZRANGE
#
# [0.7 - Aug 28, 2012] A lot of internal rewrites in the library. Its now not just a
# a set of functions, but a proper object library with internal
# variables hidden from outside. Support has also been added for
# regex matching with PATTERN specifier and for generalized
# --check option that can be used where specific long option is
# not available. For use with that option also added UOM specifier.
# Also added checkin 'master_last_io_seconds_ago' (when link is down)
# for when replication_delay info is requested.
# [0.71 - Sep 03, 2012] Fixed bug in a new library related to when data is missing
# [0.72 - Oct 05, 2012] Fixed bug reported by Matt McMillan in specified memory size
# when KB are used. Fixed bugs in adding performance data that
# results in keyspace_hits, keyspace_misses, memory_utilization
# having double 'c' or '%' in perfdata. Added contributors section.
# [0.73 - Mar 23, 2013] Fixed bug in parse_threshold function of embedded library
#
# TODO or consider for future:
#
# 1. Library Enhancements (will apply to multiple plugins that share common code)
# (a) Add '--extra-opts' to allow to read options from a file as specified
# at http://nagiosplugins.org/extra-opts. This is TODO for all my plugins
# (b) [DONE]
# In plans are to allow long options to specify thresholds for known variables.
# These would mean you specify '--connected_clients' in similar way to '--hitrate'
# Internally these would be converged into -A, -w, -c as appropriate and used
# together with these options. So in practice it will now allow to get any data
# just a different way to specify options for this plugin.
# (c) Allow regex when selecting variable name(s) with -a, this will be enabled with
# a special option and not be default
# [DONE]
#
# 2. REDIS Specific
# (a) Add option to check from master that slave is connected and working.
# (b) Look into replication delay from master and how it can be done. Look
# for into on replication_delay from slave as well
# (c) How to better calculate memory utilization and get max memory available
# without directly specifying it
# (d) Maybe special options to measure cpu use and set thresholds
#
# Others are welcome recommend a new feature to be added here. If so please email to
# william@leibzon.org.
# And don't worry, I'm not a company with some hidden agenda to use your idea
# but an actual person who you can easily get hold of by email, find on forums
# and on Nagios conferences. More info on my nagios work is at:
# http://william.leibzon.org/nagios/
# Above site should also have PNP4Nagios template for this and other plugins.
#
# ============================ LIST OF CONTRIBUTORS ===============================
#
# The following individuals have contributed code, patches, bug fixes and ideas to
# this plugin (listed in last-name alphabetical order):
#
# William Leibzon
# Matthew Litwin
# Matt McMillan
# Jon Schulz
# M Spiegle
#
# ============================ START OF PROGRAM CODE =============================
use strict;
use IO::Socket;
use Time::HiRes;
use Text::ParseWords;
use Getopt::Long qw(:config no_ignore_case);
use Redis;
# default hostname, port, database, user and password, see NOTES above
my $HOSTNAME= 'localhost';
my $PORT= 6379;
my $PASSWORD= undef;
my $DATABASE= undef;
# Add path to additional libraries if necessary
use lib '/usr/lib/nagios/plugins';
our $TIMEOUT;
our %ERRORS;
eval 'use utils qw(%ERRORS $TIMEOUT)';
if ($@) {
$TIMEOUT = 20;
%ERRORS = ('OK'=>0,'WARNING'=>1,'CRITICAL'=>2,'UNKNOWN'=>3,'DEPENDENT'=>4);
}
my $Version='0.73';
# This is a list of known stat and info variables including variables added by plugin,
# used in order to designate COUNTER variables with 'c' in perfout for graphing programs
# The format is:
# VAR_NAME => [ TYPE, PerfSuffix, DESCRIPTION]
# If option has description, the variable will also become available as a long option so for example
# you can specify "--connected_clients=WARN,CRIT" instead of specifying "-a connected_clients -w WARN -c CRIT'
my %KNOWN_STATUS_VARS = (
'memory_utilization' => [ 'status', 'GAUGE', '%' ], # calculated by plugin
'redis_version' => [ 'status', 'VERSION', '' ], # version string variable
'response_time' => [ 'status','GAUGE', 's' ], # measured by plugin
'hitrate' => [ 'status', 'GAUGE', '%' ], # calculated by plugin
'total_keys' => [ 'status','GAUGE', '', 'Total Number of Keys on the Server' ],
'total_expires' => [ 'status','GAUGE', '', 'Number of Expired Keys for All DBs' ],
'last_save_time' => [ 'status', 'GAUGE', 's' ],
'bgsave_in_progress' => [ 'status', 'BOOLEAN', '' ],
'vm_enabled' => [ 'status', 'BOOLEAN', '' ],
'uptime_in_seconds' => [ 'status', 'COUNTER', 'c' ],
'total_connections_received' => [ 'status', 'COUNTER', 'c', 'Total Connections Received' ],
'used_memory_rss' => [ 'status', 'GAUGE', 'B', 'Resident Set Size, Used Memory in Bytes' ], # RSS - Resident Set Size
'used_cpu_sys' => [ 'status', 'GAUGE', '', 'Main Process Used System CPU' ],
'redis_git_dirty' => [ 'status', 'BOOLEAN', '', 'Git Dirty Set Bit' ],
'loading' => [ 'status', 'BOOLEAN', '' ],
'latest_fork_usec' => [ 'status', 'GAUGE', '' ],
'connected_clients' => [ 'status', 'GAUGE', '', 'Total Number of Connected Clients' ],
'used_memory_peak_human' => [ 'status', 'GAUGE', '' ],
'mem_allocator' => [ 'status', 'TEXTINFO', '' ],
'uptime_in_days' => [ 'status', 'COUNTER', 'c', 'Total Uptime in Days' ],
'keyspace_hits' => [ 'status', 'COUNTER', 'c', 'Total Keyspace Hits' ],
'client_biggest_input_buf' => [ 'status', 'GAUGE', '' ],
'gcc_version' => [ 'status', 'TEXTINFO', '' ],
'changes_since_last_save' => [ 'status', 'COUNTER', 'c' ],
'arch_bits' => [ 'status', 'TEXTINFO', '' ],
'lru_clock' => [ 'status', 'GAUGE', '' ], # LRU is page replacement algorithm (least recently used), I'm unsure what this represents though
'role' => [ 'status', 'SETTING', '' ],
'multiplexing_api' => [ 'status', 'SETTING' , '' ],
'slave' => [ 'status', 'TEXTDATA', '' ],
'pubsub_channels' => [ 'status', 'GAUGE', '', 'Number of Pubsub Channels' ],
'redis_git_sha1' => [ 'status', 'TEXTDATA', '' ],
'used_cpu_user_children' => [ 'status', 'GAUGE', '', 'Child Processes Used User CPU' ],
'process_id' => [ 'status', 'GAUGE', '' ],
'used_memory_human' => [ 'status', 'GAUGE', '' ],
'keyspace_misses' => [ 'status', 'COUNTER', 'c', 'Keyspace Misses' ],
'used_cpu_user' => [ 'status', 'GAUGE', '', 'Main Process Used User CPU' ],
'total_commands_processed' => [ 'status', 'COUNTER', 'c', 'Total Number of Commands Processed from Start' ],
'mem_fragmentation_ratio' => [ 'status', 'GAUGE', '', 'Memory Fragmentation Ratio' ],
'client_longest_output_list' => [ 'status', 'GAUGE', '' ],
'blocked_clients' => [ 'status', 'GAUGE', '', 'Number of Currently Blocked Clients' ],
'aof_enabled' => [ 'status', 'BOOLEAN', '' ],
'evicted_keys' => [ 'status', 'COUNTER', 'c', 'Total Number of Evicted Keys' ],
'bgrewriteaof_in_progress' => [ 'status','BOOLEAN', '' ],
'expired_keys' => [ 'status', 'COUNTER', 'c', 'Total Number of Expired Keys' ],
'used_memory_peak' => [ 'status', 'GAUGE', 'B' ],
'connected_slaves' => [ 'status', 'GAUGE', '', 'Number of Connected Slaves' ],
'used_cpu_sys_children' => [ 'status', 'GAUGE', '', 'Child Processed Used System CPU' ],
'master_host' => [ 'status', 'TEXTINFO', '' ],
'master_port' => [ 'status', 'TEXTINFO', '' ],
'master_link_status' => [ 'status', 'TEXTINFO', '' ],
'slave0' => [ 'status', 'TEXTINFO', '' ],
'slave1' => [ 'status', 'TEXTINFO', '' ],
'slave2' => [ 'status', 'TEXTINFO', '' ],
'slave3' => [ 'status', 'TEXTINFO', '' ],
);
# Here you can also specify which variables should go into perf data,
# For right now it is 'GAUGE', 'COUNTER', 'DATA' (but not 'TEXTDATA'), and 'BOOLEAN'
# you may want to remove BOOLEAN if you don't want too much data
my $PERF_OK_STATUS_REGEX = 'GAUGE|COUNTER|^DATA$|BOOLEAN';
# ============= MAIN PROGRAM CODE - DO NOT MODIFY BELOW THIS LINE ==============
my $o_host= undef; # hostname
my $o_port= undef; # port
my $o_pwfile= undef; # password file
my $o_password= undef; # password as parameter
my $o_database= undef; # database name (usually a number)
my $o_help= undef; # help option
my $o_verb= undef; # verbose mode
my $o_version= undef; # version info option
my $o_variables=undef; # list of variables for warn and critical
my $o_perfvars= undef; # list of variables to include in performance data
my $o_warn= undef; # warning level option
my $o_crit= undef; # Critical level option
my $o_perf= undef; # Performance data option
my @o_check= (); # General check option that maybe repeated more than once
my $o_timeout= undef; # Timeout to use - note that normally timeout is from nagios
my $o_timecheck=undef; # threshold spec for connection time
my $o_memutilization=undef; # threshold spec for memory utilization%
my $o_totalmemory=undef; # total memory on a system
my $o_hitrate= undef; # threshold spec for hitrate%
my $o_repdelay=undef; # replication delay time
my @o_querykey=(); # query this key, this option maybe repeated so its an array
my $o_prevperf= undef; # performance data given with $SERVICEPERFDATA$ macro
my $o_prevtime= undef; # previous time plugin was run $LASTSERVICECHECK$ macro
my $o_ratelabel=undef; # prefix and suffix for creating rate variables
my $o_rsuffix='_rate'; # default suffix
my $o_rprefix='';
## Additional global variables
my $redis= undef; # DB connection object
my @query=(); # array of queries with each entry being keyed hash of processedoption data on howto query
sub p_version { print "check_redis.pl version : $Version\n"; }
sub print_usage_line {
print "Usage: $0 [-v [debugfilename]] -H <host> [-p <port>] [-x password | -C credentials_file] [-D <database>] [-a <statistics variables> -w <variables warning thresholds> -c <variables critical thresholds>] [-A <performance output variables>] [-T [conntime_warn,conntime_crit]] [-R [hitrate_warn,hitrate_crit]] [-m [mem_utilization_warn,mem_utilization_crit] [-M <maxmemory>[B|K|M|G]]] [-r replication_delay_time_warn,replication_delay_time_crit] [-f] [-T <timeout>] [-V] [-P <previous performance data in quoted string>] [-q (GET|LLEN|HLEN|SLEN|ZLEN|HGET:name|HEXISTS:name|SEXISTS:name|LRANGE:(AVG|SUM|MIN|MAX):start:end|ZRANGE:(AVG|SUM|MIN|MAX):start:end),query_type,query_key_name[:data_name][,ABSENT:WARNING|CRITICAL][,WARN:threshold,CRIT:threshold]] [-o <threshold specification with name or pattern>]\n";
}
sub print_usage {
print_usage_line();
print "For more details on options do: $0 --help\n";
}
sub help {
my $nlib = shift;
print "Redis Check for Nagios version ",$Version,"\n";
print " by William Leibzon - william(at)leibzon.org\n\n";
print "This is redis monitoring plugin to check its stats variables, replication, response time\n";
print "hitrate, memory utilization and other info. The plugin can also query and test key data\n";
print "against specified thresholds. All data is available as performance output for graphing.\n\n";
print_usage_line();
print "\n";
print <<EOT;
General and Server Connection Options:
-v, --verbose[=FILENAME], --debug[=FILENAME]
Print extra debugging information.
If filename is specified instead of STDOUT the debug data is written to that file.
-h, --help
Print this detailed help screen
-H, --hostname=ADDRESS
Hostname or IP Address to check
-p, --port=INTEGER
port number (default: 6379)
-D, --database=NAME
optional database name (usually a number), needed for --query but otherwise not needed
-x, --password=STRING
Password for Redis authentication. Safer alternative is to put them in a file and use -C
-C, --credentials=FILENAME
Credentials file to read for Redis authentication
-t, --timeout=NUMBER
Allows to set timeout for execution of this plugin. This overrides nagios default.
-V, --version
Prints version number
Variables and Thresholds Set as List:
-a, --variables=STRING[,STRING[,STRING...]]
List of variables from info data to do threshold checks on.
The default (if option is not used) is not to monitor any variable.
The variable name should be prefixed with '&' to chec its rate of
change over time rather than actual value.
-w, --warn=STR[,STR[,STR[..]]]
This option can only be used if '--variables' (or '-a') option above
is used and number of values listed here must exactly match number
of variables specified with '-a'. The values specify warning threshold
for when Nagios should send WARNING alert. These values are usually
numbers and can have the following prefix modifiers:
> - warn if data is above this value (default for numeric values)
< - warn if data is below this value (must be followed by number)
= - warn if data is equal to this value (default for non-numeric values)
! - warn if data is not equal to this value
~ - do not check this data (must not be followed by number or ':')
^ - for numeric values this disables check that warning < critical
Threshold values can also be specified as range in two forms:
num1:num2 - warn if data is outside range i.e. if data<num1 or data>num2
\@num1:num2 - warn if data is in range i.e. data>=num1 && data<=num2
-c, --crit=STR[,STR[,STR[..]]]
This option can only be used if '--variables' (or '-a') option above
is used and number of values listed here must exactly match number of
variables specified with '-a'. The values specify critical threshold
for when Nagios should send CRITICAL alert. The format is exactly same
as with -w option except no '^' prefix.
Performance Data Processing Options:
-f, --perfparse
This should only be used with '-a' and causes variable data not only as part of
main status line but also as perfparse compatible output (for graphing, etc).
-A, --perfvars=[STRING[,STRING[,STRING...]]]
This allows to list variables which values will go only into perfparse
output (and not for threshold checking). The option by itself (emply value)
is same as a special value '*' and specify to output all variables.
-P, --prev_perfdata
Previous performance data (normally put '-P \$SERVICEPERFDATA\$' in nagios
command definition). This is used to calculate rate of change for counter
statistics variables and for proper calculation of hitrate.
--rate_label=[PREFIX_STRING[,SUFFIX_STRING]]
Prefix or Suffix label used to create a new variable which has rate of change
of another base variable. You can specify PREFIX or SUFFIX or both. Default
if not specified is suffix '_rate' i.e. --rate_label=,_rate
Key Data Query Option (maybe repeated more than once):
-q, --query=query_type,key[:varname][,ABSENT:OK|WARNING|CRITICAL,WARN:threshold,CRIT:threshold]
query_type is one of:
GET - get one data value
LLEN - number of items in a list
LRANGE:AVG:start:end - retrieve list and average results
LRANGE:SUM:start:end - retrieve list and sum results
LRANGE:MIN:start:end - retrieve list and return minimum
LRANGE:MAX:start:end - retrieve list and return maximum
HLEN - returns number of items in a hash
HGET:name - get specific hash key 'name'
HEXISTS:name - returns 0 or 1 depending on if specified hash key 'name' exists
SLEN - returns number of items in a set
SEXISTS:name - returns 0 or 1 depending on if set member 'name' exists
ZLEN - returns number of items in a sorted set
ZCOUNT:min:max - counts items in sorted set with scores within the given values
ZRANGE:AVG:min:max - retrieve sorted set members from min to max and average results
ZRANGE:SUM:min:max - retrieve sorted set members from min to max and sum results
ZRANGE:MIN:min:max - retrieve sorted set members from min to max list and return minimum
ZRANGE:MAX:min:max - retrieve sorted set memers from min to max and return maximum
Option specifies key to query and optional variable name to assign the results to after :
(if not specified it would be same as key). If key is not available the plugin can issue
either warning or critical alert depending on what you specified after ABSENT.
Numeric results are calculated for ranges and can be checked with specified thresholds
or you can do it together with standard with redis stats variables and -a option.
General Check Option (all 3 forms equivalent, can be repated more than once):
-o <list of specifiers>, --option=<list of specifiers>, --check=<list of specifiers>
where specifiers are separated by , and must include NAME or PATTERN:
NAME:<string> - Default name for this variable as you'd have specified with -v
PATTERN:<regex> - Regular Expression that allows to match multiple data results
WARN:threshold - warning alert threshold
CRIT:threshold - critical alert threshold
Threshold is a value (usually numeric) which may have the following prefix:
> - warn if data is above this value (default for numeric values)
< - warn if data is below this value (must be followed by number)
= - warn if data is equal to this value (default for non-numeric values)
! - warn if data is not equal to this value
Threshold can also be specified as a range in two forms:
num1:num2 - warn if data is outside range i.e. if data<num1 or data>num2
\@num1:num2 - warn if data is in range i.e. data>=num1 && data<=num2
ABSENT:OK|WARNING|CRITICAL|UNKNOWN - Nagios alert (or lock of thereof) if data is absent
ZERO:OK|WARNING|CRITICAL|UNKNOWN - Nagios alert (or lock of thereof) if result is 0
DISPLAY:YES|NO - Specifies if data should be included in nagios status line output
PERF:YES|NO - Output results as performance data or not (always YES if asked for rate)
UOM:<string> - Unit Of Measurement symbol to add to perf data - 'c','%','s','B'
Measured/Calculated Data:
-T, --response_time=[WARN,CRIT]
If this is used as just -T the plugin will measure and output connection
response time in seconds. With -f this would also be provided on perf variables.
You can also specify values for this parameter, these are interprted as
WARNING and CRITICAL thresholds (separated by ',').
-R, --hitrate=[WARN,CRIT]
Calculates Hitrate %: cache_miss/(cache_hits+cache_miss). If this is used
as just -R then this info just goes to output line. With '-R -f' these
go as performance data. You can also specify values for this parameter,
these are interprted as WARNING and CRITICAL thresholds (separated by ',').
The format for WARN and CRIT is same as what you would use in -w and -c.
-m, --memory_utilization=[WARN,CRIT]
This calculates percent of total memory on system used by redis, which is
utilization=redis_memory_rss/total_memory*100.
Total_memory on server must be specified with -M since Redis does not report
it and can use maximum memory unless you enabled virtual memory and set a limit
(I plan to test this case and see if it gets reported then).
If you specify -m by itself, the plugin will just output this info,
with '-f' it will also include this in performance data. You can also specify
parameter values which are interpreted as WARNING and CRITICAL thresholds.
-M, --total_memory=NUM[B|K|M|G]
Amount of memory on a system for memory utilization calculations above.
If it does not end with K,M,G then its assumed to be B (bytes)
-r, --replication_delay=WARN,CRIT
Allows to set threshold on replication delay info. Only valid if this is a slave!
The threshold value is in seconds and fractions are acceptable.
EOT
if (defined($nlib) && $nlib->{'enable_long_options'} == 1) {
my $long_opt_help = $nlib->additional_options_help();
if ($long_opt_help) {
print "Stats Variable Options (this is alternative to specifying them as list with -a):\n";
print $long_opt_help;
print "\n";
}
}
}
############################ START OF THE LIBRARY FUNCTIONS #####################################
#
# THIS IS WORK IN PROGRESS, THE LIBRARY HAS NOT BEEN RELEASED YET AND INTERFACES MAY CHANGE
#
# ====================================== SUMMARY ================================================
#
# Name : Naglio Perl Library For Developing Nagios Plugins
# Version : 0.2
# Date : Aug 28, 2012
# Author : William Leibzon - william@leibzon.org
# Licence : LGPL - full text at http://www.fsf.org/licenses/lgpl.txt
#
# ============================= LIBRARY HISTORY AND VERSIONS ====================================
#
# Note: you may safely skip this section if you're looking at documentation about this library or plugin
#
# [2006-2008] The history of this library goes back to plugins such as check_snmp_temperature.pl,
# check_mysqld,pl and others released as early as 2006 with common functions to
# support prefixes "<,>,=,!" for specifying thresholds and checking data against
# these thresholds. Several of my plugins had common architecture supporting multiple
# variables or attributes to be checked using -a/--attributes/--variables option and
# --warn and --crit options with list of thresholds for these attributes and --perfvars
# specifying variables whose data would only go as PERFOUT for graphing.
#
# [2008-2011] Threshold parsing and check code had been rewritten and support added for specifying
# range per plugin guidelines: http://nagiosplug.sourceforge.net/developer-guidelines.html
# Internal structures had been changing and becoming more complex to various cases.
# In 2010-2012 plugins started to get support for ;warn;crit output of thresholds in perf,
# as specified in the guidelines.
#
# [Early 2012] Code from check_memcached had been used as a base for check_memcached and then
# check_redis plugins with some of the latest threshold code from check_netstat
# with more updates. Starting with check_redis the code from check_options() and
# from main part of plugin that was very similar across my plugins were separated
# into their own functions. KNOWN_STATS_VARS array was introduced as well to be
# able to properly add UOM symbol ('c', '%', 's', 'ms', 'B', 'KB') to perfout.
# check_memcached and check_redis also included support for calculating rate of
# variables in a similar way to how its been done in check_snmp_netint
#
# [0.1 - July 17, 2012] In 0.6 release of check_redis.pl support had been added for long options
# with special threshold line syntax:
# --option=WARN:threshold,CRIT:threshold,ABSENT:OK|WARNING|CRITICAL|UNKNOWN,DISPLAY:YES|NO,PERF:YES|NO
# This was extension from just doing --option=WARN,CRIT to have a more universal
# and extendable way to specify and alike parameters for checking. check_redis 0.6
# also introduced support automatically adding long options with above syntax based
# on description in KNOWN_STATS_VARS. The functions for the library were all separated
# into their own section of the code. When inported to check_memcached global variables
# were added to that section and accessor functions written for some of them.
# This is considered 0.1 version of the library
#
# [0.2 - Aug 28, 2012] In August the library code in check_memcached had been re-written from
# just functions to object-oriented perl interface. All variables were hidden from
# direct access with accessor functions written. Documentation header had been added
# to each library function and the header for the library itself. This was major work
# taking over a week to do although functions and mainly sllllame as in 0.1. They are
# not stabilized and so library is only to be included within plugins. Support was
# also added for regex matching with PATTERN option spec. Also added NAME spec.
# License changed to LGPL from GPL for this code.
# [0.21 - Sep 3, 2012] Fix bug in handling absent data
# [0.22 - Mar 23, 2013] Fix bug in parse_threshold functon
#
# ================================== LIBRARY TODO =================================================
#
# (a) Add library function to support '--extra-opts' to read plugin options from a file
# This is being to be compatible with http://nagiosplugins.org/extra-opts
# (b) Support regex matching and allowing multiple data for same threshold definition.
# [DONE]
# (c) Support for expressions in places of numeric values for thresholds. The idea is to allow
# to refer to another variable or to special macro. I know at least one person has extended
# my check_mysqld to support using mysql variables (not same as status data) for thresholds.
# I also previouslyhad planned such support with experimental check_snmp_attributes plugin
# library/base. The idea was also floated around on nagios-devel list.
# (d) Support specifying variables as expressions. This is straight out of check_snmp_atributes
# and maybe part of it can be reused for this
# (e) Add common SNMP functions into library as so many of my plugins use it#
# (f) Add more functions to make this library easier to use and stabilize its interfaces.
# Port my plugins to this library.
# (f) Add support for functions in Nagios-Plugins perl library. While its interfaces are
# different, I believe, it'd be possible to add "shim" code to support them too.
# (h) Write proper Perl-style documentation as well as web documentation (much of above maybe
# moved to web documentation) and move library to separate GITHUB project. Release it.
# (i) Port this library to Python and write one or two example plugins
#
# ================================================================================================
{
package Naglio;
use fields qw();
use Text::ParseWords;
my %ERRORS = ('OK'=>0,'WARNING'=>1,'CRITICAL'=>2,'UNKNOWN'=>3,'DEPENDENT'=>4);
my $DEFAULT_PERF_OK_STATUS_REGEX = 'GAUGE|COUNTER|^DATA$|BOOLEAN';
# @DESCRIPTION : Library object constructor
# @LAST CHANGED : 08-27-12 by WL
# @INPUT : Hash array of named config settings. All parameters are optiona. Currently supported are:
# plugin_name => string - short name of the plugin
# plugin_description => string - plugin longer description
# plugin_authors => string - list of plugin authors
# knownStatsVars => reference to hash - hash array defining known variables, what type they are, their description
# usage_function => &ref - function that would display helpful text in case of error with options for this plugin
# verbose => 1 or "" or "filename" - set to 1 or "" if verbose/debug or to filename to send data to (may not be called "0" or "1")
# output_comparison_symbols => 0 or 1 - 1 means library output in case threshold is met can use "<", ">", "="
# 0 means output is something like "less than or equal", "more than", etc.
# all_variables_perf => 0 or 1 - 1 means data for all variables would go to PERF. This is what '-A *' or just -A do
# enable_long_options => 0 or 1 - 1 enables long options generated based on knownStatsVars. This is automatically enabled (from 0
# to 1) when plugin references additional_options_list() unless this is set to -1 at library init
# enable_rate_of_change => 0 or 1 - enables support for calculating rate of change based on previously saved data, default is 1
# enable_regex_match => 0 or 1 - when set to 1 each threshold-specified var name is treated as regex and can match
# to multiple collected data. this can also be enabled per-variable with PATTERN spec
# @RETURNS : Reference representing object instance of this library
# @PRIVACY & USE : PUBLIC, To be used when initializing the library
sub lib_init {
my $invocant = shift;
my $class = ref($invocant) || $invocant;
my %other_args = @_;
# These used to be global variables, now these are object local variables in self with accessor
my @allVars = (); # all variables after options processing
my @perfVars = (); # performance variables list [renamed from @o_perfVarsL in earlier code]
my %thresholds=(); # hash array of thresholds for above variables, [this replaced @o_warnL and @o_critL in earlier code]
my %dataresults= (); # This is where data is loaded. It is a hash with variable names as keys and array array for value:
# $dataresults{$var}[0] - undef of value of this variable
# $dataresults{$var}[1] - 0 if variable not printed out to status line yet, 1 or more otherwise
# $dataresults{$var}[2] - 0 if variable data not yet put into PERF output, -1 if PERF output is preset, 1 after output
# $dataresults{$var}[3] - string, '' to start with, holds ready performance data output for this variable
# $dataresults{$var}[4] - only for regex matches. name of match var (which should be key in thresholds), otherwise undef
my %dataVars = (); # keys are variables from allVars and perfVars, values is array of data that matched i.e. keys in dataresults
my @ar_warnLv = (); # used during options processing
my @ar_critLv = (); # used during options processing
my @ar_varsL= (); # used during options processing
my @prev_time= (); # timestamps if more then one set of previois performance data
my $self = { # library and nagios versions
_NaglioLibraryVersion => 0.2, # this library's version
_NagiosVersion => 3, # assume nagios core 3.x unless known otherwise
# library internal data structures
_allVars => \@allVars,
_perfVars => \@perfVars,
_thresholds => \%thresholds,
_dataresults => \%dataresults,
_datavars => \%dataVars,
_ar_warnLv => \@ar_warnLv,
_ar_critLv => \@ar_critLv,
_ar_varsL => \@ar_varsL,
_prevTime => \@prev_time,
_prevPerf => {}, # array that is populated with previous performance data
_checkTime => undef, # time when data was last checked
_statuscode => "OK", # final status code
_statusinfo => "", # if there is an error, this has human info about what it is
_statusdata => "", # if there is no error but we want some data in status line, this var gets it
_perfdata => "", # this variable collects performance data line
_saveddata => "", # collects saved data (for next plugin re-run, not implimented yet)
_init_args => \%other_args,
# copy of data from plugin option variables
o_variables => undef, # List of variables for warn and critical checks
o_crit => undef, # Comma-separated list of critical thresholds for each checked variable
o_warn => undef, # Comma-separated list of warning thresholds for each checked variable
o_perf => undef, # defined or undef. perf option means all data from variables also goes as PERFDATA
o_perfvars => undef, # List of variables only for PERFDATA
o_prevperf => undef, # previously saved performance data coming from $SERVICEPERFDATA$ macro
# library special input variables (similar to options)
o_rprefix => '', # prefix used to distinguish rate variables
o_rsuffix => '_rate', # suffix used to distinguish rate variables
knownStatusVars => {}, # Special HASH ARRAY with names and description of known variables
perfOKStatusRegex => $DEFAULT_PERF_OK_STATUS_REGEX,
verbose => 0, # verbose, same as debug, same as o_verb
plugin_name => '', # next 3 parameters are variables are currently not used
plugin_description => '', # but its still better if these are provided
plugin_authors => '', # in the future these maybe used for help & usage functions
# library setting variables
debug_file => "", # instead of setting file name in verbose, can also set it here
output_comparison_symbols => 1, # should plugin output >,<.=,! for threshold match
# if 0, it will say it in human form, i.e. "less"
all_variables_perf => 0, # should we all variables go to PERF (even those not listed in o_variables and o_perfvars)
# this is the option set to 1 when --perfvars '*' is used
enable_long_options => 0, # enable support for long options generated based on knownStatusVars description
enable_rate_of_change => 1, # enables support for calculatin rate of chane and for rate of change long options
enable_regex_match => 0, # 0 is not enabled, 1 means variables in o_variables and o_perfvars are considered regex to match actual data
# a value of 2 means its enabled, but for options with PATTERN specifier (this is not configurale value)
};
# bless to create an object
bless $self, $class;
# deal with arguments that maybe passed to library when initalizing
if (exists($other_args{'KNOWN_STATUS_VARS'})) {
$self->{'knownStatusVars'} = $other_args{'KNOWN_STATUS_VARS'};
}
$self->{'plugin_name'} = $other_args{'plugin_name'} if exists($other_args{'plugin_name'});
$self->{'plugin_description'} = $other_args{'plugin_description'} if exists($other_args{'plugin_description'});
$self->{'plugin_authors'} = $other_args{'plugin_authors'} if exists($other_args{'plugin_authors'});
$self->{'usage_function'} = $other_args{'usage_gunction'} if exists($other_args{'usage_function'});
$self->configure(%other_args);
# return self object
return $self;
}
# This is just an alias for object constructor lib_init function
sub new {
return lib_init(@_);
}
# @DESCRIPTION : Allows to confiure some settings after initialization (all these can also be done as part of lib_init)
# @LAST CHANGED : 08-27-12 by WL
# @INPUT : Hash array of named config settings. All parameters are optiona. Currently supported are:
# verbose => 1 or "" or "filename" - set to 1 or "" if verbose/debug or to filename to send data to (may not be called "0" or "1")
# output_comparison_symbols => 0 or 1 - 1 means library output in case threshold is met can use "<", ">", "="
# 0 means output is something like "less than or equal", "more than", etc.
# all_variables_perf => 0 or 1 - 1 means data for all variables would go to PERF. This is what '-A *' or just -A do
# enable_long_options => 0 or 1 - 1 enables long options generated based on knownStatsVars. This is automatically enabled (from 0
# to 1) when plugin references additional_options_list() unless this is set to -1 at library init
# enable_rate_of_change => 0 or 1 - enables support for calculating rate of change based on previously saved data, default is 1
# enable_regex_match => 0 or 1 - when set to 1 each threshold-specified var name is treated as regex and can match
# to multiple collected data. this can also be enabled per-variable with PATTERN spec
# @RETURNS : nothing (future: 1 on success, 0 on error)
# @PRIVACY & USE : PUBLIC, Must be used as an object instance function.
sub configure {
my $self = shift;
my %args = @_;
if (exists($args{'verbose'}) || exists($args{'debug'})) {
$self->{'verbose'} = 1;
if (exists($args{'verbose'}) && $args{'verbose'}) {
$self->{'debug_file'} = $args{'verbose'};
}
if (exists($args{'debug_log_filename'})) {
$self->{'debug_file'} = $args{'debug_log_filename'};
}
}
$self->{'all_variables_perf'} = $args{'all_variables_perf'} if exists($args{'all_variables_perf'});
$self->{'enable_long_options'} = $args{'enable_long_options'} if exists($args{'enable_long_options'});
$self->{'enable_rate_of_change'} = $args{'enable_rate_of_change'} if exists($args{'enable_rate_of_change'});
$self->{'enable_regex_match'} = 1 if exists($args{'enable_regex_match'}) && $args{'enable_regex_match'}!=0;
$self->{'output_comparison_symbols'} = $args{'output_comparison_symbols'} if exists($args{'output_comparison_symbols'});
}
# @DESCRIPTION : Allows functions to take be used both directly and as object referenced functions
# In the 2nd case they get $self as 1st argument, in 1st they don't. this just adds
# $self if its if its not there so their argument list is known.
# Functions that allow both should still check if $self is defined
# @LAST CHANGED : 08-20-12 by WL
# @INPUT : arbitrary list of arguments
# @RETURNS : arbitrary list of arguments with 1st being object hash or undef
# @PRIVACY & USE : PRIVATE
sub _self_args {
return @_ if ref($_[0]) && exists($_[0]->{'_NaglioLibraryVersion'});
unshift @_,undef;
return @_;
}
# @DESCRIPTION : Sets function to be called to display help text on using plugin in case of error
# @LAST CHANGED : 08-22-12 by WL
# @INPUT : reference to usage function
# @RETURNS : nothing
# @PRIVACY & USE : PUBLIC, Must be used as an object instance function :
sub set_usage_function {
my ($self, $usage_function) = @_;
$self->{'usage_function'} = $usage_function;
}
# @DESCRIPTION : Usage function. For right now it just calls usage function given as a parameter
# In the future if it is not available, it'll print something standard.
# @LAST CHANGED : 08-22-12 by WL
# @INPUT : none
# @RETURNS : nothing
# @PRIVACY & USE : PUBLIC, But primary for internal use. Must be used as an object instance function.
sub usage {
my $self = shift;
if (defined($self) && defined($self->{'usage_function'})) { &{$self->{'usage_function'}}(); }
}
# @DESCRIPTION : This function converts uptime in seconds to nice & short output format
# @LAST_CHANGED : 08-20-12 by WL
# @INPUT : ARG1 - uptime in seconds
# @RETURNS : string of uptime for human consumption
# @PRIVACY & USE : PUBLIC, Maybe used directly or as object instance function :
sub uptime_info {
my ($self,$uptime_seconds) = _self_args(@_);
my $upinfo = "";
my ($secs,$mins,$hrs,$days) = (undef,undef,undef,undef);