1 /**
2 * collectd-nagios - src/collectd-nagios.c
3 * Copyright (C) 2008-2010 Florian octo Forster
4 *
5 * Permission is hereby granted, free of charge, to any person obtaining a
6 * copy of this software and associated documentation files (the "Software"),
7 * to deal in the Software without restriction, including without limitation
8 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
9 * and/or sell copies of the Software, and to permit persons to whom the
10 * Software is furnished to do so, subject to the following conditions:
11 *
12 * The above copyright notice and this permission notice shall be included in
13 * all copies or substantial portions of the Software.
14 *
15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
20 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
21 * DEALINGS IN THE SOFTWARE.
22 *
23 * Authors:
24 * Florian octo Forster <octo at collectd.org>
25 **/
27 #if HAVE_CONFIG_H
28 # include "config.h"
29 #endif
31 #if !defined(__GNUC__) || !__GNUC__
32 # define __attribute__(x) /**/
33 #endif
35 #include <stdlib.h>
36 #include <unistd.h>
37 #include <stdio.h>
38 #include <errno.h>
39 #include <string.h>
40 #include <strings.h>
41 #include <assert.h>
43 #if NAN_STATIC_DEFAULT
44 # include <math.h>
45 /* #endif NAN_STATIC_DEFAULT*/
46 #elif NAN_STATIC_ISOC
47 # ifndef __USE_ISOC99
48 # define DISABLE_ISOC99 1
49 # define __USE_ISOC99 1
50 # endif /* !defined(__USE_ISOC99) */
51 # include <math.h>
52 # if DISABLE_ISOC99
53 # undef DISABLE_ISOC99
54 # undef __USE_ISOC99
55 # endif /* DISABLE_ISOC99 */
56 /* #endif NAN_STATIC_ISOC */
57 #elif NAN_ZERO_ZERO
58 # include <math.h>
59 # ifdef NAN
60 # undef NAN
61 # endif
62 # define NAN (0.0 / 0.0)
63 # ifndef isnan
64 # define isnan(f) ((f) != (f))
65 # endif /* !defined(isnan) */
66 # ifndef isfinite
67 # define isfinite(f) (((f) - (f)) == 0.0)
68 # endif
69 # ifndef isinf
70 # define isinf(f) (!isfinite(f) && !isnan(f))
71 # endif
72 #endif /* NAN_ZERO_ZERO */
74 #include "libcollectdclient/collectd/client.h"
76 #define RET_OKAY 0
77 #define RET_WARNING 1
78 #define RET_CRITICAL 2
79 #define RET_UNKNOWN 3
81 #define CON_NONE 0
82 #define CON_AVERAGE 1
83 #define CON_SUM 2
84 #define CON_PERCENTAGE 3
86 struct range_s
87 {
88 double min;
89 double max;
90 int invert;
91 };
92 typedef struct range_s range_t;
94 extern char *optarg;
95 extern int optind, opterr, optopt;
97 static char *socket_file_g = NULL;
98 static char *value_string_g = NULL;
99 static char *hostname_g = NULL;
101 static range_t range_critical_g;
102 static range_t range_warning_g;
103 static int consolitation_g = CON_NONE;
104 static _Bool nan_is_error_g = 0;
106 static char **match_ds_g = NULL;
107 static int match_ds_num_g = 0;
109 /* `strdup' is an XSI extension. I don't want to pull in all of XSI just for
110 * that, so here's an own implementation.. It's easy enough. The GCC attributes
111 * are supposed to get good performance.. -octo */
112 __attribute__((malloc, nonnull (1)))
113 static char *cn_strdup (const char *str) /* {{{ */
114 {
115 size_t strsize;
116 char *ret;
118 strsize = strlen (str) + 1;
119 ret = (char *) malloc (strsize);
120 if (ret != NULL)
121 memcpy (ret, str, strsize);
122 return (ret);
123 } /* }}} char *cn_strdup */
125 static int filter_ds (size_t *values_num,
126 double **values, char ***values_names)
127 {
128 gauge_t *new_values;
129 char **new_names;
131 size_t i;
133 if (match_ds_g == NULL)
134 return (RET_OKAY);
136 new_values = (gauge_t *)calloc (match_ds_num_g, sizeof (*new_values));
137 if (new_values == NULL)
138 {
139 fprintf (stderr, "malloc failed: %s\n", strerror (errno));
140 return (RET_UNKNOWN);
141 }
143 new_names = (char **)calloc (match_ds_num_g, sizeof (*new_names));
144 if (new_names == NULL)
145 {
146 fprintf (stderr, "malloc failed: %s\n", strerror (errno));
147 free (new_values);
148 return (RET_UNKNOWN);
149 }
151 for (i = 0; i < (size_t) match_ds_num_g; i++)
152 {
153 size_t j;
155 /* match_ds_g keeps pointers into argv but the names will be freed */
156 new_names[i] = cn_strdup (match_ds_g[i]);
157 if (new_names[i] == NULL)
158 {
159 fprintf (stderr, "cn_strdup failed: %s\n", strerror (errno));
160 free (new_values);
161 for (j = 0; j < i; j++)
162 free (new_names[j]);
163 free (new_names);
164 return (RET_UNKNOWN);
165 }
167 for (j = 0; j < *values_num; j++)
168 if (strcasecmp (new_names[i], (*values_names)[j]) == 0)
169 break;
171 if (j == *values_num)
172 {
173 printf ("ERROR: DS `%s' is not available.\n", new_names[i]);
174 free (new_values);
175 for (j = 0; j <= i; j++)
176 free (new_names[j]);
177 free (new_names);
178 return (RET_CRITICAL);
179 }
181 new_values[i] = (*values)[j];
182 }
184 free (*values);
185 for (i = 0; i < *values_num; i++)
186 free ((*values_names)[i]);
187 free (*values_names);
189 *values = new_values;
190 *values_names = new_names;
191 *values_num = match_ds_num_g;
192 return (RET_OKAY);
193 } /* int filter_ds */
195 static void parse_range (char *string, range_t *range)
196 {
197 char *min_ptr;
198 char *max_ptr;
200 if (*string == '@')
201 {
202 range->invert = 1;
203 string++;
204 }
206 max_ptr = strchr (string, ':');
207 if (max_ptr == NULL)
208 {
209 min_ptr = NULL;
210 max_ptr = string;
211 }
212 else
213 {
214 min_ptr = string;
215 *max_ptr = '\0';
216 max_ptr++;
217 }
219 assert (max_ptr != NULL);
221 /* `10' == `0:10' */
222 if (min_ptr == NULL)
223 range->min = 0.0;
224 /* :10 == ~:10 == -inf:10 */
225 else if ((*min_ptr == '\0') || (*min_ptr == '~'))
226 range->min = NAN;
227 else
228 range->min = atof (min_ptr);
230 if ((*max_ptr == '\0') || (*max_ptr == '~'))
231 range->max = NAN;
232 else
233 range->max = atof (max_ptr);
234 } /* void parse_range */
236 static int match_range (range_t *range, double value)
237 {
238 int ret = 0;
240 if (!isnan (range->min) && (range->min > value))
241 ret = 1;
242 if (!isnan (range->max) && (range->max < value))
243 ret = 1;
245 return (((ret - range->invert) == 0) ? 0 : 1);
246 } /* int match_range */
248 static void usage (const char *name)
249 {
250 fprintf (stderr, "Usage: %s <-s socket> <-n value_spec> <-H hostname> [options]\n"
251 "\n"
252 "Valid options are:\n"
253 " -s <socket> Path to collectd's UNIX-socket.\n"
254 " -n <v_spec> Value specification to get from collectd.\n"
255 " Format: `plugin-instance/type-instance'\n"
256 " -d <ds> Select the DS to examine. May be repeated to examine multiple\n"
257 " DSes. By default all DSes are used.\n"
258 " -g <consol> Method to use to consolidate several DSes.\n"
259 " See below for a list of valid arguments.\n"
260 " -H <host> Hostname to query the values for.\n"
261 " -c <range> Critical range\n"
262 " -w <range> Warning range\n"
263 " -m Treat \"Not a Number\" (NaN) as critical (default: warning)\n"
264 "\n"
265 "Consolidation functions:\n"
266 " none: Apply the warning- and critical-ranges to each data-source\n"
267 " individually.\n"
268 " average: Calculate the average of all matching DSes and apply the\n"
269 " warning- and critical-ranges to the calculated average.\n"
270 " sum: Apply the ranges to the sum of all DSes.\n"
271 " percentage: Apply the ranges to the ratio (in percent) of the first value\n"
272 " and the sum of all values."
273 "\n", name);
274 exit (1);
275 } /* void usage */
277 static int do_listval (lcc_connection_t *connection)
278 {
279 lcc_identifier_t *ret_ident = NULL;
280 size_t ret_ident_num = 0;
282 char *hostname = NULL;
284 int status;
285 size_t i;
287 status = lcc_listval (connection, &ret_ident, &ret_ident_num);
288 if (status != 0) {
289 printf ("UNKNOWN: %s\n", lcc_strerror (connection));
290 if (ret_ident != NULL)
291 free (ret_ident);
292 return (RET_UNKNOWN);
293 }
295 status = lcc_sort_identifiers (connection, ret_ident, ret_ident_num);
296 if (status != 0) {
297 printf ("UNKNOWN: %s\n", lcc_strerror (connection));
298 if (ret_ident != NULL)
299 free (ret_ident);
300 return (RET_UNKNOWN);
301 }
303 for (i = 0; i < ret_ident_num; ++i) {
304 char id[1024];
306 if ((hostname_g != NULL) && (strcasecmp (hostname_g, ret_ident[i].host)))
307 continue;
309 if ((hostname == NULL) || strcasecmp (hostname, ret_ident[i].host))
310 {
311 if (hostname != NULL)
312 free (hostname);
313 hostname = strdup (ret_ident[i].host);
314 printf ("Host: %s\n", hostname);
315 }
317 /* empty hostname; not to be printed again */
318 ret_ident[i].host[0] = '\0';
320 status = lcc_identifier_to_string (connection,
321 id, sizeof (id), ret_ident + i);
322 if (status != 0) {
323 printf ("ERROR: listval: Failed to convert returned "
324 "identifier to a string: %s\n",
325 lcc_strerror (connection));
326 continue;
327 }
329 /* skip over the (empty) hostname and following '/' */
330 printf ("\t%s\n", id + 1);
331 }
333 if (ret_ident != NULL)
334 free (ret_ident);
335 return (RET_OKAY);
336 } /* int do_listval */
338 static int do_check_con_none (size_t values_num,
339 double *values, char **values_names)
340 {
341 int num_critical = 0;
342 int num_warning = 0;
343 int num_okay = 0;
344 const char *status_str = "UNKNOWN";
345 int status_code = RET_UNKNOWN;
346 size_t i;
348 for (i = 0; i < values_num; i++)
349 {
350 if (isnan (values[i]))
351 {
352 if (nan_is_error_g)
353 num_critical++;
354 else
355 num_warning++;
356 }
357 else if (match_range (&range_critical_g, values[i]) != 0)
358 num_critical++;
359 else if (match_range (&range_warning_g, values[i]) != 0)
360 num_warning++;
361 else
362 num_okay++;
363 }
365 if ((num_critical == 0) && (num_warning == 0) && (num_okay == 0))
366 {
367 printf ("WARNING: No defined values found\n");
368 return (RET_WARNING);
369 }
370 else if ((num_critical == 0) && (num_warning == 0))
371 {
372 status_str = "OKAY";
373 status_code = RET_OKAY;
374 }
375 else if (num_critical == 0)
376 {
377 status_str = "WARNING";
378 status_code = RET_WARNING;
379 }
380 else
381 {
382 status_str = "CRITICAL";
383 status_code = RET_CRITICAL;
384 }
386 printf ("%s: %i critical, %i warning, %i okay", status_str,
387 num_critical, num_warning, num_okay);
388 if (values_num > 0)
389 {
390 printf (" |");
391 for (i = 0; i < values_num; i++)
392 printf (" %s=%f;;;;", values_names[i], values[i]);
393 }
394 printf ("\n");
396 return (status_code);
397 } /* int do_check_con_none */
399 static int do_check_con_average (size_t values_num,
400 double *values, char **values_names)
401 {
402 size_t i;
403 double total;
404 int total_num;
405 double average;
406 const char *status_str = "UNKNOWN";
407 int status_code = RET_UNKNOWN;
409 total = 0.0;
410 total_num = 0;
411 for (i = 0; i < values_num; i++)
412 {
413 if (isnan (values[i]))
414 {
415 if (!nan_is_error_g)
416 continue;
418 printf ("CRITICAL: Data source \"%s\" is NaN\n",
419 values_names[i]);
420 return (RET_CRITICAL);
421 }
423 total += values[i];
424 total_num++;
425 }
427 if (total_num == 0)
428 {
429 printf ("WARNING: No defined values found\n");
430 return (RET_WARNING);
431 }
433 average = total / total_num;
435 if (match_range (&range_critical_g, average) != 0)
436 {
437 status_str = "CRITICAL";
438 status_code = RET_CRITICAL;
439 }
440 else if (match_range (&range_warning_g, average) != 0)
441 {
442 status_str = "WARNING";
443 status_code = RET_WARNING;
444 }
445 else
446 {
447 status_str = "OKAY";
448 status_code = RET_OKAY;
449 }
451 printf ("%s: %g average |", status_str, average);
452 for (i = 0; i < values_num; i++)
453 printf (" %s=%f;;;;", values_names[i], values[i]);
454 printf ("\n");
456 return (status_code);
457 } /* int do_check_con_average */
459 static int do_check_con_sum (size_t values_num,
460 double *values, char **values_names)
461 {
462 size_t i;
463 double total;
464 int total_num;
465 const char *status_str = "UNKNOWN";
466 int status_code = RET_UNKNOWN;
468 total = 0.0;
469 total_num = 0;
470 for (i = 0; i < values_num; i++)
471 {
472 if (isnan (values[i]))
473 {
474 if (!nan_is_error_g)
475 continue;
477 printf ("CRITICAL: Data source \"%s\" is NaN\n",
478 values_names[i]);
479 return (RET_CRITICAL);
480 }
482 total += values[i];
483 total_num++;
484 }
486 if (total_num == 0)
487 {
488 printf ("WARNING: No defined values found\n");
489 return (RET_WARNING);
490 }
492 if (match_range (&range_critical_g, total) != 0)
493 {
494 status_str = "CRITICAL";
495 status_code = RET_CRITICAL;
496 }
497 else if (match_range (&range_warning_g, total) != 0)
498 {
499 status_str = "WARNING";
500 status_code = RET_WARNING;
501 }
502 else
503 {
504 status_str = "OKAY";
505 status_code = RET_OKAY;
506 }
508 printf ("%s: %g sum |", status_str, total);
509 for (i = 0; i < values_num; i++)
510 printf (" %s=%f;;;;", values_names[i], values[i]);
511 printf ("\n");
513 return (status_code);
514 } /* int do_check_con_sum */
516 static int do_check_con_percentage (size_t values_num,
517 double *values, char **values_names)
518 {
519 size_t i;
520 double sum = 0.0;
521 double percentage;
523 const char *status_str = "UNKNOWN";
524 int status_code = RET_UNKNOWN;
526 if ((values_num < 1) || (isnan (values[0])))
527 {
528 printf ("WARNING: The first value is not defined\n");
529 return (RET_WARNING);
530 }
532 for (i = 0; i < values_num; i++)
533 {
534 if (isnan (values[i]))
535 {
536 if (!nan_is_error_g)
537 continue;
539 printf ("CRITICAL: Data source \"%s\" is NaN\n",
540 values_names[i]);
541 return (RET_CRITICAL);
542 }
544 sum += values[i];
545 }
547 if (sum == 0.0)
548 {
549 printf ("WARNING: Values sum up to zero\n");
550 return (RET_WARNING);
551 }
553 percentage = 100.0 * values[0] / sum;
555 if (match_range (&range_critical_g, percentage) != 0)
556 {
557 status_str = "CRITICAL";
558 status_code = RET_CRITICAL;
559 }
560 else if (match_range (&range_warning_g, percentage) != 0)
561 {
562 status_str = "WARNING";
563 status_code = RET_WARNING;
564 }
565 else
566 {
567 status_str = "OKAY";
568 status_code = RET_OKAY;
569 }
571 printf ("%s: %lf percent |", status_str, percentage);
572 for (i = 0; i < values_num; i++)
573 printf (" %s=%lf;;;;", values_names[i], values[i]);
574 return (status_code);
575 } /* int do_check_con_percentage */
577 static int do_check (lcc_connection_t *connection)
578 {
579 gauge_t *values;
580 char **values_names;
581 size_t values_num;
582 char ident_str[1024];
583 lcc_identifier_t ident;
584 size_t i;
585 int status;
587 snprintf (ident_str, sizeof (ident_str), "%s/%s",
588 hostname_g, value_string_g);
589 ident_str[sizeof (ident_str) - 1] = 0;
591 memset (&ident, 0, sizeof (ident));
592 status = lcc_string_to_identifier (connection, &ident, ident_str);
593 if (status != 0)
594 {
595 printf ("ERROR: Creating an identifier failed: %s.\n",
596 lcc_strerror (connection));
597 LCC_DESTROY (connection);
598 return (RET_CRITICAL);
599 }
601 status = lcc_getval (connection, &ident,
602 &values_num, &values, &values_names);
603 if (status != 0)
604 {
605 printf ("ERROR: Retrieving values from the daemon failed: %s.\n",
606 lcc_strerror (connection));
607 LCC_DESTROY (connection);
608 return (RET_CRITICAL);
609 }
611 LCC_DESTROY (connection);
613 status = filter_ds (&values_num, &values, &values_names);
614 if (status != RET_OKAY)
615 return (status);
617 status = RET_UNKNOWN;
618 if (consolitation_g == CON_NONE)
619 status = do_check_con_none (values_num, values, values_names);
620 else if (consolitation_g == CON_AVERAGE)
621 status = do_check_con_average (values_num, values, values_names);
622 else if (consolitation_g == CON_SUM)
623 status = do_check_con_sum (values_num, values, values_names);
624 else if (consolitation_g == CON_PERCENTAGE)
625 status = do_check_con_percentage (values_num, values, values_names);
627 free (values);
628 if (values_names != NULL)
629 for (i = 0; i < values_num; i++)
630 free (values_names[i]);
631 free (values_names);
633 return (status);
634 } /* int do_check */
636 int main (int argc, char **argv)
637 {
638 char address[1024];
639 lcc_connection_t *connection;
641 int status;
643 range_critical_g.min = NAN;
644 range_critical_g.max = NAN;
645 range_critical_g.invert = 0;
647 range_warning_g.min = NAN;
648 range_warning_g.max = NAN;
649 range_warning_g.invert = 0;
651 while (42)
652 {
653 int c;
655 c = getopt (argc, argv, "w:c:s:n:H:g:d:hm");
656 if (c < 0)
657 break;
659 switch (c)
660 {
661 case 'c':
662 parse_range (optarg, &range_critical_g);
663 break;
664 case 'w':
665 parse_range (optarg, &range_warning_g);
666 break;
667 case 's':
668 socket_file_g = optarg;
669 break;
670 case 'n':
671 value_string_g = optarg;
672 break;
673 case 'H':
674 hostname_g = optarg;
675 break;
676 case 'g':
677 if (strcasecmp (optarg, "none") == 0)
678 consolitation_g = CON_NONE;
679 else if (strcasecmp (optarg, "average") == 0)
680 consolitation_g = CON_AVERAGE;
681 else if (strcasecmp (optarg, "sum") == 0)
682 consolitation_g = CON_SUM;
683 else if (strcasecmp (optarg, "percentage") == 0)
684 consolitation_g = CON_PERCENTAGE;
685 else
686 {
687 fprintf (stderr, "Unknown consolidation function `%s'.\n",
688 optarg);
689 usage (argv[0]);
690 }
691 break;
692 case 'd':
693 {
694 char **tmp;
695 tmp = (char **) realloc (match_ds_g,
696 (match_ds_num_g + 1)
697 * sizeof (char *));
698 if (tmp == NULL)
699 {
700 fprintf (stderr, "realloc failed: %s\n",
701 strerror (errno));
702 return (RET_UNKNOWN);
703 }
704 match_ds_g = tmp;
705 match_ds_g[match_ds_num_g] = cn_strdup (optarg);
706 if (match_ds_g[match_ds_num_g] == NULL)
707 {
708 fprintf (stderr, "cn_strdup failed: %s\n",
709 strerror (errno));
710 return (RET_UNKNOWN);
711 }
712 match_ds_num_g++;
713 break;
714 }
715 case 'm':
716 nan_is_error_g = 1;
717 break;
718 default:
719 usage (argv[0]);
720 } /* switch (c) */
721 }
723 if ((socket_file_g == NULL) || (value_string_g == NULL)
724 || ((hostname_g == NULL) && (strcasecmp (value_string_g, "LIST"))))
725 {
726 fprintf (stderr, "Missing required arguments.\n");
727 usage (argv[0]);
728 }
730 snprintf (address, sizeof (address), "unix:%s", socket_file_g);
731 address[sizeof (address) - 1] = 0;
733 connection = NULL;
734 status = lcc_connect (address, &connection);
735 if (status != 0)
736 {
737 printf ("ERROR: Connecting to daemon at %s failed.\n",
738 socket_file_g);
739 return (RET_CRITICAL);
740 }
742 if (0 == strcasecmp (value_string_g, "LIST"))
743 return (do_listval (connection));
745 return (do_check (connection));
746 } /* int main */