summaryrefslogtreecommitdiff
path: root/roles/monitoring/prometheus/exporter/node/templates/textfile-collector-scripts/smartmon.j2
blob: 829383e272458a4f06ee232faaf5a5cc332a33b3 (plain) (blame)
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
#!/usr/bin/env {{ python_basename }}

import argparse
import collections
import csv
import re
import shlex
import subprocess
import sys
from prometheus_client import CollectorRegistry, Gauge, generate_latest

device_info_re = re.compile(r'^(?P<k>[^:]+?)(?:(?:\sis|):)\s*(?P<v>.*)$')

ata_error_count_re = re.compile(
    r'^Error (\d+) \[\d+\] occurred', re.MULTILINE)

self_test_re = re.compile(r'^SMART.*(PASSED|OK)$', re.MULTILINE)

device_info_map = {
    'Vendor': 'vendor',
    'Product': 'product',
    'Revision': 'revision',
    'Logical Unit id': 'lun_id',
    'Model Family': 'model_family',
    'Device Model': 'device_model',
    'Serial Number': 'serial_number',
    'Serial number': 'serial_number',
    'Firmware Version': 'firmware_version',
}

smart_attributes_whitelist = (
    'airflow_temperature_cel',
    'command_timeout',
    'current_pending_sector',
    'end_to_end_error',
    'erase_fail_count_total',
    'g_sense_error_rate',
    'hardware_ecc_recovered',
    'host_reads_mib',
    'host_reads_32mib',
    'host_writes_mib',
    'host_writes_32mib',
    'load_cycle_count',
    'media_wearout_indicator',
    'wear_leveling_count',
    'nand_writes_1gib',
    'offline_uncorrectable',
    'power_cycle_count',
    'power_on_hours',
    'program_fail_count',
    'raw_read_error_rate',
    'reallocated_event_count',
    'reallocated_sector_ct',
    'reported_uncorrect',
    'sata_downshift_count',
    'seek_error_rate',
    'spin_retry_count',
    'spin_up_time',
    'start_stop_count',
    'temperature_case',
    'temperature_celsius',
    'temperature_internal',
    'total_lbas_read',
    'total_lbas_written',
    'udma_crc_error_count',
    'unsafe_shutdown_count',
    'workld_host_reads_perc',
    'workld_media_wear_indic',
    'workload_minutes',
)

registry = CollectorRegistry()
namespace = "smartmon"

metrics = {
    "smartctl_version": Gauge(
        "smartctl_version",
        "SMART metric smartctl_version",
        ["version"],
        namespace=namespace,
        registry=registry,
    ),
    "device_active": Gauge(
        "device_active",
        "SMART metric device_active",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "device_info": Gauge(
        "device_info",
        "SMART metric device_info",
        [
            "device",
            "disk",
            "vendor",
            "product",
            "revision",
            "lun_id",
            "model_family",
            "device_model",
            "serial_number",
            "firmware_version",
        ],
        namespace=namespace,
        registry=registry,
    ),
    "device_smart_available": Gauge(
        "device_smart_available",
        "SMART metric device_smart_available",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "device_smart_enabled": Gauge(
        "device_smart_enabled",
        "SMART metric device_smart_enabled",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "device_smart_healthy": Gauge(
        "device_smart_healthy",
        "SMART metric device_smart_healthy",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),

    # SMART attributes - ATA disks only
    "attr_value": Gauge(
        "attr_value",
        "SMART metric attr_value",
        ["device", "disk", "name"],
        namespace=namespace,
        registry=registry,
    ),
    "attr_worst": Gauge(
        "attr_worst",
        "SMART metric attr_worst",
        ["device", "disk", "name"],
        namespace=namespace,
        registry=registry,
    ),
    "attr_threshold": Gauge(
        "attr_threshold",
        "SMART metric attr_threshold",
        ["device", "disk", "name"],
        namespace=namespace,
        registry=registry,
    ),
    "attr_raw_value": Gauge(
        "attr_raw_value",
        "SMART metric attr_raw_value",
        ["device", "disk", "name"],
        namespace=namespace,
        registry=registry,
    ),
    "device_errors": Gauge(
        "device_errors",
        "SMART metric device_errors",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "available_spare_ratio": Gauge(
        "available_spare_ratio",
        "SMART metric available_spare_ratio",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "available_spare_threshold_ratio": Gauge(
        "available_spare_threshold_ratio",
        "SMART metric available_spare_threshold_ratio",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "percentage_used_ratio": Gauge(
        "percentage_used_ratio",
        "SMART metric percentage_used_ratio",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "power_cycles_total": Gauge(
        "power_cycles_total",
        "SMART metric power_cycles_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "power_on_hours_total": Gauge(
        "power_on_hours_total",
        "SMART metric power_on_hours_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "temperature_celcius": Gauge(
        "temperature_celcius",
        "SMART metric temperature_celcius",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "unsafe_shutdowns_total": Gauge(
        "unsafe_shutdowns_total",
        "SMART metric unsafe_shutdowns_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "media_errors_total": Gauge(
        "media_errors_total",
        "SMART metric media_errors_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "num_err_log_entries_total": Gauge(
        "num_err_log_entries_total",
        "SMART metric num_err_log_entries_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "warning_temperature_time_total": Gauge(
        "warning_temperature_time_total",
        "SMART metric warning_temperature_time_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
    "critical_temperature_time_total": Gauge(
        "critical_temperature_time_total",
        "SMART metric critical_temperature_time_total",
        ["device", "disk"],
        namespace=namespace,
        registry=registry,
    ),
}

SmartAttribute = collections.namedtuple('SmartAttribute', [
    'id', 'name', 'flag', 'value', 'worst', 'threshold', 'type', 'updated',
    'when_failed', 'raw_value',
])


class Device(collections.namedtuple('DeviceBase', 'path opts')):
    """Representation of a device as found by smartctl --scan output."""

    @property
    def type(self):
        return self.opts.type

    @property
    def base_labels(self):
        return {'device': self.path, 'disk': self.type.partition('+')[2] or '0'}

    def smartctl_select(self):
        return ['--device', self.type, self.path]


def smart_ctl(*args, check=True):
    """Wrapper around invoking the smartctl binary.

    Returns:
        (str) Data piped to stdout by the smartctl subprocess.
    """
    return subprocess.run(
        ['smartctl', *args], stdout=subprocess.PIPE, check=check
    ).stdout.decode('utf-8')


def smart_ctl_version():
    return smart_ctl('-V').split('\n')[0].split()[1]


def find_devices(by_id):
    """Find SMART devices.

    Yields:
        (Device) Single device found by smartctl.
    """
    parser = argparse.ArgumentParser()
    parser.add_argument('-d', '--device', dest='type')

    args = ['--scan-open']
    if by_id:
        args.extend(['-d', 'by-id'])
    devices = smart_ctl(*args)

    for device in devices.split('\n'):
        device = device.strip()
        if not device:
            continue

        tokens = shlex.split(device, comments=True)
        if not tokens:
            continue

        yield Device(tokens[0], parser.parse_args(tokens[1:]))


def device_is_active(device):
    """Returns whenever the given device is currently active or not.

    Args:
        device: (Device) Device in question.

    Returns:
        (bool) True if the device is active and False otherwise.
    """
    try:
        smart_ctl('--nocheck', 'standby', *device.smartctl_select())
    except subprocess.CalledProcessError:
        return False

    return True


def device_info(device):
    """Query device for basic model information.

    Args:
        device: (Device) Device in question.

    Returns:
        (generator): Generator yielding:

            key (str): Key describing the value.
            value (str): Actual value.
    """
    info_lines = smart_ctl(
        '--info', *device.smartctl_select()
    ).strip().split('\n')[3:]

    matches = (device_info_re.match(line) for line in info_lines)
    return (m.groups() for m in matches if m is not None)


def device_smart_capabilities(device):
    """Returns SMART capabilities of the given device.

    Args:
        device: (Device) Device in question.

    Returns:
        (tuple): tuple containing:

            (bool): True whenever SMART is available, False otherwise.
            (bool): True whenever SMART is enabled, False otherwise.
    """

    # NVME devices are SMART capable
    if device.type == 'nvme':
        return True, True

    groups = device_info(device)

    state = {
        g[1].split(' ', 1)[0]
        for g in groups if g[0] == 'SMART support'}

    smart_available = 'Available' in state
    smart_enabled = 'Enabled' in state

    return smart_available, smart_enabled


def collect_device_info(device):
    """Collect basic device information.

    Args:
        device: (Device) Device in question.
    """
    values = dict(device_info(device))
    metrics["device_info"].labels(
        device.base_labels["device"],
        device.base_labels["disk"],
        values.get("Vendor", ""),
        values.get("Product", ""),
        values.get("Revision", ""),
        values.get("Logical Unit id", ""),
        values.get("Model Family", ""),
        values.get("Device Model", ""),
        values.get("Serial Number", ""),
        values.get("Firmware Version", ""),
    ).set(1)


def collect_device_health_self_assessment(device):
    """Collect metric about the device health self assessment.

    Args:
        device: (Device) Device in question.
    """
    out = smart_ctl('--health', *device.smartctl_select(), check=False)

    self_assessment_passed = bool(self_test_re.search(out))
    metrics["device_smart_healthy"].labels(
        device.base_labels["device"], device.base_labels["disk"]
    ).set(self_assessment_passed)


def collect_ata_metrics(device):
    # Fetch SMART attributes for the given device.
    attributes = smart_ctl(
        '--attributes', *device.smartctl_select()
    )

    # replace multiple occurrences of whitespace with a single whitespace
    # so that the CSV Parser recognizes individual columns properly.
    attributes = re.sub(r'[\t\x20]+', ' ', attributes)

    # Turn smartctl output into a list of lines and skip to the table of
    # SMART attributes.
    attribute_lines = attributes.strip().split('\n')[7:]

    # Some attributes have multiple IDs but have the same name.  Don't
    # yield attributes that already have been reported before.
    seen = set()

    reader = csv.DictReader(
        (line.strip() for line in attribute_lines),
        fieldnames=SmartAttribute._fields[:-1],
        restkey=SmartAttribute._fields[-1], delimiter=' ')
    for entry in reader:
        # We're only interested in the SMART attributes that are
        # whitelisted here.
        entry['name'] = entry['name'].lower()
        if entry['name'] not in smart_attributes_whitelist:
            continue

        # Ensure that only the numeric parts are fetched from the raw_value.
        # Attributes such as 194 Temperature_Celsius reported by my SSD
        # are in the format of "36 (Min/Max 24/40)" which can't be expressed
        # properly as a prometheus metric.
        m = re.match(r'^(\d+)', ' '.join(entry['raw_value']))
        if not m:
            continue
        entry['raw_value'] = m.group(1)

        # Some device models report "---" in the threshold value where most
        # devices would report "000". We do the substitution here because
        # downstream code expects values to be convertable to integer.
        if entry['threshold'] == '---':
            entry['threshold'] = '0'

        if entry['name'] in smart_attributes_whitelist and entry['name'] not in seen:
            for col in 'value', 'worst', 'threshold', 'raw_value':
                metrics["attr_" + col].labels(
                    device.base_labels["device"],
                    device.base_labels["disk"],
                    entry["name"],
                ).set(entry[col])

            seen.add(entry['name'])


def collect_ata_error_count(device):
    """Inspect the device error log and report the amount of entries.

    Args:
        device: (Device) Device in question.
    """
    error_log = smart_ctl(
        '-l', 'xerror,1', *device.smartctl_select(), check=False)

    m = ata_error_count_re.search(error_log)

    error_count = m.group(1) if m is not None else 0
    metrics["device_errors"].labels(
        device.base_labels["device"], device.base_labels["disk"]
    ).set(error_count)


def collect_nvme_metrics(device):
    # Fetch NVME metrics
    attributes = smart_ctl(
        '--attributes', *device.smartctl_select()
    )

    # replace multiple occurrences of whitespaces with a single whitespace
    attributes = re.sub(r'[\t\x20]+', ' ', attributes)

    # Turn smartctl output into a list of lines and skip to the table of
    # SMART attributes.
    attribute_lines = attributes.strip().split('\n')[6:]
    for line in attribute_lines:
        label, value = line.split(':')
        if label == 'Available Spare':
            metrics['available_spare_ratio'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value[0:-1])
        elif label == 'Available Spare Threshold':
            metrics['available_spare_threshold_ratio'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value[0:-1])
        elif label == 'Percentage Used':
            metrics['percentage_used_ratio'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value[0:-1])
        elif label == 'Power Cycle':
            metrics['power_cycles_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value)
        elif label == 'Power On Hours':
            metrics['power_on_hours_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value.replace(',', ''))
        elif label == 'Temperature':
            metrics['temperature_celcius'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value.replace(' Celsius', ''))
        elif label == 'Unsafe Shutdowns':
            metrics['unsafe_shutdowns_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value)
        elif label == 'Media and Data Integrity Errors':
            metrics['media_errors_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value)
        elif label == 'Error Information Log Entries':
            metrics['num_err_log_entries_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value)
        elif label == 'Warning Comp. Temperature Time':
            metrics['warning_temperature_time_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value)
        elif label == 'Critical Comp. Temperature Time':
            metrics['critical_temperature_time_total'].labels(device.base_labels["device"], device.base_labels["disk"]).set(value)


def collect_disks_smart_metrics(wakeup_disks, by_id, include_nvme):
    for device in find_devices(by_id):
        is_active = device_is_active(device)
        metrics["device_active"].labels(
            device.base_labels["device"], device.base_labels["disk"],
        ).set(is_active)

        # Skip further metrics collection to prevent the disk from spinning up.
        if not is_active and not wakeup_disks:
            continue

        collect_device_info(device)

        smart_available, smart_enabled = device_smart_capabilities(device)

        metrics["device_smart_available"].labels(
            device.base_labels["device"], device.base_labels["disk"]
        ).set(smart_available)

        metrics["device_smart_enabled"].labels(
            device.base_labels["device"], device.base_labels["disk"]
        ).set(smart_enabled)

        # Skip further metrics collection here if SMART is disabled on the device. Further smartctl
        # invocations would fail anyway.
        if not smart_available:
            continue

        collect_device_health_self_assessment(device)

        if device.type.startswith('sat'):
            collect_ata_metrics(device)
            collect_ata_error_count(device)

        if include_nvme and device.type == 'nvme':
            collect_nvme_metrics(device)


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument('-s', '--wakeup-disks', dest='wakeup_disks', action='store_true',
                        help="Wake up disks to collect live stats")
    parser.add_argument('--by-id', dest='by_id', action='store_true',
                        help="Use /dev/disk/by-id/X instead of /dev/sdX to index devices")
    parser.add_argument('--include-nvme', dest='include_nvme', action='store_true',
                        help="Include metrics for NVMe drives")
    args = parser.parse_args(sys.argv[1:])

    metrics["smartctl_version"].labels(smart_ctl_version()).set(1)

    collect_disks_smart_metrics(args.wakeup_disks, args.by_id, args.include_nvme)
    print(generate_latest(registry).decode(), end="")


if __name__ == '__main__':
    main()