1NVME-WDC-SMART-AD(1)              NVMe Manual             NVME-WDC-SMART-AD(1)
2
3
4

NAME

6       nvme-wdc-smart-add-log - Send NVMe WDC smart-add-log Vendor Unique
7       Command, return result
8

SYNOPSIS

10       nvme wdc smart-add-log <device> [--interval=<NUM>, -i <NUM>] [--output-format=<normal|json> -o <normal|json>]
11

DESCRIPTION

13       For the NVMe device given, send a Vendor Unique WDC smart-add-log
14       command and provide the additional smart log. The --interval option
15       will return performance statistics from the specified reporting
16       interval.
17
18       The <device> parameter is mandatory and may be either the NVMe
19       character device (ex: /dev/nvme0).
20
21       This will only work on WDC devices supporting this feature. Results for
22       any other device are undefined.
23
24       On success it returns 0, error code otherwise.
25

OPTIONS

27       -i <NUM>, --interval=<NUM>
28           Return the statistics from specific interval, defaults to 14
29
30       -o <format>, --output-format=<format>
31           Set the reporting format to normal, or json. Only one output format
32           can be used at a time. Default is normal.
33
34       Valid Interval values and description :-
35
36       ┌──────┬────────────────────────────┐
37Value Description                
38       ├──────┼────────────────────────────┤
39       │      │                            │
401     │ Most recent five (5)       │
41       │      │ minute accumulated set.    │
42       ├──────┼────────────────────────────┤
43       │      │                            │
442-12  │ Previous five (5) minute   │
45       │      │ accumulated sets.          │
46       ├──────┼────────────────────────────┤
47       │      │                            │
4813    │ The accumulated total of   │
49       │      │ sets 1 through 12 that     │
50       │      │ contain the previous hour  │
51       │      │ of accumulated statistics. │
52       ├──────┼────────────────────────────┤
53       │      │                            │
5414    │ The statistical set        │
55       │      │ accumulated since          │
56       │      │ power-up.                  │
57       ├──────┼────────────────────────────┤
58       │      │                            │
5915    │ The statistical set        │
60       │      │ accumulated during the     │
61       │      │ entire lifetime of the     │
62       │      │ device.                    │
63       └──────┴────────────────────────────┘
64

CA LOG PAGE DATA OUTPUT EXPLANATION

66       ┌───────────────────────────┬────────────────────────────┐
67Field                      Description                
68       ├───────────────────────────┼────────────────────────────┤
69       │                           │                            │
70Physical NAND bytes        │ The number of bytes        │
71written.                   │ written to NAND. 16 bytes  │
72       │                           │ - hi/lo                    │
73       ├───────────────────────────┼────────────────────────────┤
74       │                           │                            │
75Physical NAND bytes read   │ The number of bytes read   │
76       │                           │ from NAND. 16 bytes -      │
77       │                           │ hi/lo                      │
78       ├───────────────────────────┼────────────────────────────┤
79       │                           │                            │
80Bad NAND Block Count       │ Raw and normalized count   │
81       │                           │ of the number of NAND      │
82       │                           │ blocks that have been      │
83       │                           │ retired after the drives   │
84       │                           │ manufacturing tests (i.e.  │
85       │                           │ grown back blocks). 2      │
86       │                           │ bytes normalized, 6 bytes  │
87       │                           │ raw count                  │
88       ├───────────────────────────┼────────────────────────────┤
89       │                           │                            │
90Uncorrectable Read Error   │ Total count of NAND reads  │
91Count                      │ that were not correctable  │
92       │                           │ by read retries, all       │
93       │                           │ levels of ECC, or XOR (as  │
94       │                           │ applicable). 8 bytes       │
95       ├───────────────────────────┼────────────────────────────┤
96       │                           │                            │
97Soft ECC Error Count       │ Total count of NAND reads  │
98       │                           │ that were not correctable  │
99       │                           │ by read retries, or        │
100       │                           │ first-level ECC. 8 bytes   │
101       ├───────────────────────────┼────────────────────────────┤
102       │                           │                            │
103SSD End to End Detection   │ A count of the detected    │
104Count                      │ errors by the SSD end to   │
105       │                           │ end error correction which │
106       │                           │ includes DRAM, SRAM, or    │
107       │                           │ other storage element      │
108       │                           │ ECC/CRC protection         │
109       │                           │ mechanism (not NAND ECC).  │
110       │                           │ 4 bytes                    │
111       ├───────────────────────────┼────────────────────────────┤
112       │                           │                            │
113SSD End to End Correction  │ A count of the corrected   │
114Count                      │ errors by the SSD end to   │
115       │                           │ end error correction which │
116       │                           │ includes DRAM, SRAM, or    │
117       │                           │ other storage element      │
118       │                           │ ECC/CRC protection         │
119       │                           │ mechanism (not NAND ECC).  │
120       │                           │ 4 bytes                    │
121       ├───────────────────────────┼────────────────────────────┤
122       │                           │                            │
123System Data % Used         │ A normalized cumulative    │
124       │                           │ count of the number of     │
125       │                           │ erase cycles per block     │
126       │                           │ since leaving the factory  │
127       │                           │ for the system (FW and     │
128       │                           │ metadata) area. Starts at  │
129       │                           │ 0 and increments. 100      │
130       │                           │ indicates that the         │
131       │                           │ estimated endurance has    │
132       │                           │ been consumed.             │
133       ├───────────────────────────┼────────────────────────────┤
134       │                           │                            │
135User Data Max Erase Count  │ The maximum erase count    │
136       │                           │ across all NAND blocks in  │
137       │                           │ the drive. 4 bytes         │
138       ├───────────────────────────┼────────────────────────────┤
139       │                           │                            │
140User Data Min Erase Count  │ The minimum erase count    │
141       │                           │ across all NAND blocks in  │
142       │                           │ the drive. 4 bytes         │
143       ├───────────────────────────┼────────────────────────────┤
144       │                           │                            │
145Refresh Count              │ A count of the number of   │
146       │                           │ blocks that have been      │
147       │                           │ re-allocated due to        │
148       │                           │ background operations      │
149       │                           │ only. 8 bytes              │
150       ├───────────────────────────┼────────────────────────────┤
151       │                           │                            │
152Program Fail Count         │ Raw and normalized count   │
153       │                           │ of total program failures. │
154       │                           │ Normalized count starts at │
155       │                           │ 100 and shows the percent  │
156       │                           │ of remaining allowable     │
157       │                           │ failures. 2 bytes          │
158       │                           │ normalized, 6 bytes raw    │
159       │                           │ count                      │
160       ├───────────────────────────┼────────────────────────────┤
161       │                           │                            │
162User Data Erase Fail Count │ Raw and normalized count   │
163       │                           │ of total erase failures in │
164       │                           │ the user area. Normalized  │
165       │                           │ count starts at 100 and    │
166       │                           │ shows the percent of       │
167       │                           │ remaining allowable        │
168       │                           │ failures. 2 bytes          │
169       │                           │ normalized, 6 bytes raw    │
170       │                           │ count                      │
171       ├───────────────────────────┼────────────────────────────┤
172       │                           │                            │
173System Area Erase Fail     │ Raw and normalized count   │
174Count                      │ of total erase failures in │
175       │                           │ the system area.           │
176       │                           │ Normalized count starts at │
177       │                           │ 100 and shows the percent  │
178       │                           │ of remaining allowable     │
179       │                           │ failures. 2 bytes          │
180       │                           │ normalized, 6 bytes raw    │
181       │                           │ count                      │
182       ├───────────────────────────┼────────────────────────────┤
183       │                           │                            │
184Thermal Throttling Status  │ The current status of      │
185       │                           │ thermal throttling         │
186       │                           │ (enabled or disabled). 2   │
187       │                           │ bytes                      │
188       ├───────────────────────────┼────────────────────────────┤
189       │                           │                            │
190Thermal Throttling Count   │ A count of the number of   │
191       │                           │ thermal throttling events. │
192       │                           │ 2 bytes                    │
193       ├───────────────────────────┼────────────────────────────┤
194       │                           │                            │
195PCIe Correctable Error     │ Summation counter of all   │
196Count                      │ PCIe correctable errors    │
197       │                           │ (Bad TLP, Bad DLLP,        │
198       │                           │ Receiver error, Replay     │
199       │                           │ timeouts, Replay           │
200       │                           │ rollovers). 8 bytes        │
201       └───────────────────────────┴────────────────────────────┘
202

C1 LOG PAGE DATA OUTPUT EXPLANATION

204       ┌───────────────────────────┬────────────────────────────┐
205Field                      Description                
206       ├───────────────────────────┼────────────────────────────┤
207       │                           │                            │
208Host Read Commands         │ Number of host read        │
209       │                           │ commands received during   │
210       │                           │ the reporting period.      │
211       ├───────────────────────────┼────────────────────────────┤
212       │                           │                            │
213Host Read Blocks           │ Number of 512-byte blocks  │
214       │                           │ requested during the       │
215       │                           │ reporting period.          │
216       ├───────────────────────────┼────────────────────────────┤
217       │                           │                            │
218Average Read Size          │ Average Read size is       │
219       │                           │ calculated using (Host     │
220       │                           │ Read Blocks/Host Read      │
221       │                           │ Commands).                 │
222       ├───────────────────────────┼────────────────────────────┤
223       │                           │                            │
224Host Read Cache Hit        │ Number of host read        │
225Commands                   │ commands that serviced     │
226       │                           │ entirely from the on-board │
227       │                           │ read cache during the      │
228       │                           │ reporting period. No       │
229       │                           │ access to the NAND flash   │
230       │                           │ memory was required. This  │
231       │                           │ count is only updated if   │
232       │                           │ the entire command was     │
233       │                           │ serviced from the cache    │
234       │                           │ memory.                    │
235       ├───────────────────────────┼────────────────────────────┤
236       │                           │                            │
237Host Read Cache Hit        │ Percentage of host read    │
238Percentage                 │ commands satisfied from    │
239       │                           │ the cache.                 │
240       ├───────────────────────────┼────────────────────────────┤
241       │                           │                            │
242Host Read Cache Hit Blocks │ Number of 512-byte blocks  │
243       │                           │ of data that have been     │
244       │                           │ returned for Host Read     │
245       │                           │ Cache Hit Commands during  │
246       │                           │ the reporting period. This │
247       │                           │ count is only updated with │
248       │                           │ the blocks returned for    │
249       │                           │ host read commands that    │
250       │                           │ were serviced entirely     │
251       │                           │ from cache memory.         │
252       ├───────────────────────────┼────────────────────────────┤
253       │                           │                            │
254Average Read Cache Hit     │ Average size of read       │
255Size                       │ commands satisfied from    │
256       │                           │ the cache.                 │
257       ├───────────────────────────┼────────────────────────────┤
258       │                           │                            │
259Host Read Commands Stalled │ Number of host read        │
260       │                           │ commands that were stalled │
261       │                           │ due to a lack of resources │
262       │                           │ within the SSD during the  │
263       │                           │ reporting period (NAND     │
264       │                           │ flash command queue full,  │
265       │                           │ low cache page count,      │
266       │                           │ cache page contention,     │
267       │                           │ etc.). Commands are not    │
268       │                           │ considered stalled if the  │
269       │                           │ only reason for the delay  │
270       │                           │ was waiting for the data   │
271       │                           │ to be physically read from │
272       │                           │ the NAND flash. It is      │
273       │                           │ normal to expect this      │
274       │                           │ count to equal zero on     │
275       │                           │ heavily utilized systems.  │
276       ├───────────────────────────┼────────────────────────────┤
277       │                           │                            │
278Host Read Commands Stalled │ Percentage of read         │
279Percentage                 │ commands that were         │
280       │                           │ stalled. If the figure is  │
281       │                           │ consistently high, then    │
282       │                           │ consideration should be    │
283       │                           │ given to spreading the     │
284       │                           │ data across multiple SSDs. │
285       ├───────────────────────────┼────────────────────────────┤
286       │                           │                            │
287Host Write Commands        │ Number of host write       │
288       │                           │ commands received during   │
289       │                           │ the reporting period.      │
290       ├───────────────────────────┼────────────────────────────┤
291       │                           │                            │
292Host Write Blocks          │ Number of 512-byte blocks  │
293       │                           │ written during the         │
294       │                           │ reporting period.          │
295       ├───────────────────────────┼────────────────────────────┤
296       │                           │                            │
297Average Write Size         │ Average Write size         │
298       │                           │ calculated using (Host     │
299       │                           │ Write Blocks/Host Write    │
300       │                           │ Commands).                 │
301       ├───────────────────────────┼────────────────────────────┤
302       │                           │                            │
303Host Write Odd Start       │ Number of host write       │
304Commands                   │ commands that started on a │
305       │                           │ non-aligned boundary       │
306       │                           │ during the reporting       │
307       │                           │ period. The size of the    │
308       │                           │ boundary alignment is      │
309       │                           │ normally 4K; therefore     │
310       │                           │ this returns the number of │
311       │                           │ commands that started on a │
312       │                           │ non-4K aligned boundary.   │
313       │                           │ The SSD requires slightly  │
314       │                           │ more time to process       │
315       │                           │ non-aligned write commands │
316       │                           │ than it does to process    │
317       │                           │ aligned write commands.    │
318       ├───────────────────────────┼────────────────────────────┤
319       │                           │                            │
320Host Write Odd Start       │ Percentage of host write   │
321Commands Percentage        │ commands that started on a │
322       │                           │ non-aligned boundary. If   │
323       │                           │ this figure is equal to or │
324       │                           │ near 100%, and the NAND    │
325       │                           │ Read Before Write value is │
326       │                           │ also high, then the user   │
327       │                           │ should investigate the     │
328       │                           │ possibility of offsetting  │
329       │                           │ the file system. For       │
330       │                           │ Microsoft Windows systems, │
331       │                           │ the user can use Diskpart. │
332       │                           │ For Unix-based operating   │
333       │                           │ systems, there is normally │
334       │                           │ a method whereby file      │
335       │                           │ system partitions can be   │
336       │                           │ placed where required.     │
337       ├───────────────────────────┼────────────────────────────┤
338       │                           │                            │
339Host Write Odd End         │ Number of host write       │
340Commands                   │ commands that ended on a   │
341       │                           │ non-aligned boundary       │
342       │                           │ during the reporting       │
343       │                           │ period. The size of the    │
344       │                           │ boundary alignment is      │
345       │                           │ normally 4K; therefore     │
346       │                           │ this returns the number of │
347       │                           │ commands that ended on a   │
348       │                           │ non-4K aligned boundary.   │
349       ├───────────────────────────┼────────────────────────────┤
350       │                           │                            │
351Host Write Odd End         │ Percentage of host write   │
352Commands Percentage        │ commands that ended on a   │
353       │                           │ non-aligned boundary.      │
354       ├───────────────────────────┼────────────────────────────┤
355       │                           │                            │
356Host Write Commands        │ Number of host write       │
357Stalled                    │ commands that were stalled │
358       │                           │ due to a lack of resources │
359       │                           │ within the SSD during the  │
360       │                           │ reporting period. The most │
361       │                           │ likely cause is that the   │
362       │                           │ write data was being       │
363       │                           │ received faster than it    │
364       │                           │ could be saved to the NAND │
365       │                           │ flash memory. If there was │
366       │                           │ a large volume of read     │
367       │                           │ commands being processed   │
368       │                           │ simultaneously, then other │
369       │                           │ causes might include the   │
370       │                           │ NAND flash command queue   │
371       │                           │ being full, low cache page │
372       │                           │ count, or cache page       │
373       │                           │ contention, etc. It is     │
374       │                           │ normal to expect this      │
375       │                           │ count to be non-zero on    │
376       │                           │ heavily utilized systems.  │
377       ├───────────────────────────┼────────────────────────────┤
378       │                           │                            │
379Host Write Commands        │ Percentage of write        │
380Stalled Percentage         │ commands that were         │
381       │                           │ stalled. If the figure is  │
382       │                           │ consistently high, then    │
383       │                           │ consideration should be    │
384       │                           │ given to spreading the     │
385       │                           │ data across multiple SSDs. │
386       ├───────────────────────────┼────────────────────────────┤
387       │                           │                            │
388NAND Read Commands         │ Number of read commands    │
389       │                           │ issued to the NAND devices │
390       │                           │ during the reporting       │
391       │                           │ period. This figure will   │
392       │                           │ normally be much higher    │
393       │                           │ than the host read         │
394       │                           │ commands figure, as the    │
395       │                           │ data needed to satisfy a   │
396       │                           │ single host read command   │
397       │                           │ may be spread across       │
398       │                           │ several NAND flash         │
399       │                           │ devices.                   │
400       ├───────────────────────────┼────────────────────────────┤
401       │                           │                            │
402NAND Read Blocks           │ Number of 512-byte blocks  │
403       │                           │ requested from NAND flash  │
404       │                           │ devices during the         │
405       │                           │ reporting period. This     │
406       │                           │ figure would normally be   │
407       │                           │ about the same as the host │
408       │                           │ read blocks figure         │
409       ├───────────────────────────┼────────────────────────────┤
410       │                           │                            │
411Average NAND Read Size     │ Average size of NAND read  │
412       │                           │ commands.                  │
413       ├───────────────────────────┼────────────────────────────┤
414       │                           │                            │
415NAND Write Commands        │ Number of write commands   │
416       │                           │ issued to the NAND devices │
417       │                           │ during the reporting       │
418       │                           │ period. There is no real   │
419       │                           │ correlation between the    │
420       │                           │ number of host write       │
421       │                           │ commands issued and the    │
422       │                           │ number of NAND Write       │
423       │                           │ Commands.                  │
424       ├───────────────────────────┼────────────────────────────┤
425       │                           │                            │
426NAND Write Blocks          │ Number of 512-byte blocks  │
427       │                           │ written to the NAND flash  │
428       │                           │ devices during the         │
429       │                           │ reporting period. This     │
430       │                           │ figure would normally be   │
431       │                           │ about the same as the host │
432       │                           │ write blocks figure.       │
433       ├───────────────────────────┼────────────────────────────┤
434       │                           │                            │
435Average NAND Write Size    │ Average size of NAND write │
436       │                           │ commands. This figure      │
437       │                           │ should never be greater    │
438       │                           │ than 128K, as this is the  │
439       │                           │ maximum size write that is │
440       │                           │ ever issued to a NAND      │
441       │                           │ device.                    │
442       ├───────────────────────────┼────────────────────────────┤
443       │                           │                            │
444NAND Read Before Write     │ This is the number of read │
445       │                           │ before write operations    │
446       │                           │ that were required to      │
447       │                           │ process non-aligned host   │
448       │                           │ write commands during the  │
449       │                           │ reporting period. See Host │
450       │                           │ Write Odd Start Commands   │
451       │                           │ and Host Write Odd End     │
452       │                           │ Commands. NAND Read Before │
453       │                           │ Write operations have a    │
454       │                           │ detrimental effect on the  │
455       │                           │ overall performance of the │
456       │                           │ device.                    │
457       └───────────────────────────┴────────────────────────────┘
458

EXAMPLES

460       ·   Has the program issue WDC smart-add-log Vendor Unique Command with
461           default interval (14) :
462
463               # nvme wdc smart-add-log /dev/nvme0
464

NVME

466       Part of the nvme-user suite.
467
468
469
470NVMe                              01/08/2019              NVME-WDC-SMART-AD(1)
Impressum