* What to do about Offline_Uncorrectable and Pending_Sector in RAID1
@ 2016-11-13 18:46 Bruce Merry
2016-11-13 20:18 ` Anthony Youngman
0 siblings, 1 reply; 17+ messages in thread
From: Bruce Merry @ 2016-11-13 18:46 UTC (permalink / raw)
To: linux-raid
[-- Attachment #1: Type: text/plain, Size: 1423 bytes --]
Hi
I'm running software RAID1 across two drives in my home machine (LVM
on LUKS on RAID1). I've just installed smartmontools and run short
tests, and promptly received emails to tell me that one of the drives
has 4 offline uncorrectable sectors and 3 current pending sectors.
I've attached smartctl --xall output for sda (good) and sdb (bad).
These drives are pretty old (over 5 years) so I'm going to replace
them as soon as I have time (and yes, I have backups), but in the
meantime I'd like advice on:
1. What exactly this means. My understanding is that some data has
been lost (or may have been lost) on the drive, but the drive still
has spare sectors to remap things once the failed sectors are written
to. Is that correct?
2. How can I tell which sectors are problematic? If it's in the swap
partition I'm far less worried. I can see two LBAs for offline
uncorrectable errors in the --xall output, but that still leaves
another two at large.
3. Assuming my understanding is correct, and that the sector falls
within the RAID1 partition on the drive, is there some way I can
recover the sectors from the other drive in the RAID1? As a last
resort I imagine I could wipe the suspect drive and then rebuild it
from the good one, but I'm hoping there's something less risky I can
do.
Thanks in advance
Bruce
--
Dr Bruce Merry
bmerry <@> gmail <.> com
http://www.brucemerry.org.za/
http://blog.brucemerry.org.za/
[-- Attachment #2: sda.txt --]
[-- Type: text/plain, Size: 16559 bytes --]
smartctl 6.2 2013-07-26 r3841 [x86_64-linux-4.4.0-47-generic] (local build)
Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org
=== START OF INFORMATION SECTION ===
Model Family: Western Digital Caviar Green (AF, SATA 6Gb/s)
Device Model: WDC WD20EARX-00PASB0
Serial Number: WD-WCAZA9626479
LU WWN Device Id: 5 0014ee 2b0e3fa4c
Firmware Version: 51.0AB51
User Capacity: 2,000,398,934,016 bytes [2.00 TB]
Sector Sizes: 512 bytes logical, 4096 bytes physical
Device is: In smartctl database [for details use: -P show]
ATA Version is: ATA8-ACS (minor revision not indicated)
SATA Version is: SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is: Sun Nov 13 20:30:04 2016 SAST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
AAM feature is: Unavailable
APM feature is: Unavailable
Rd look-ahead is: Enabled
Write cache is: Enabled
ATA Security is: Disabled, frozen [SEC2]
Wt Cache Reorder: Enabled
=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED
General SMART Values:
Offline data collection status: (0x84) Offline data collection activity
was suspended by an interrupting command from host.
Auto Offline Data Collection: Enabled.
Self-test execution status: ( 0) The previous self-test routine completed
without error or no self-test has ever
been run.
Total time to complete Offline
data collection: (36780) seconds.
Offline data collection
capabilities: (0x7b) SMART execute Offline immediate.
Auto Offline data collection on/off support.
Suspend Offline collection upon new
command.
Offline surface scan supported.
Self-test supported.
Conveyance Self-test supported.
Selective Self-test supported.
SMART capabilities: (0x0003) Saves SMART data before entering
power-saving mode.
Supports SMART auto save timer.
Error logging capability: (0x01) Error logging supported.
General Purpose Logging supported.
Short self-test routine
recommended polling time: ( 2) minutes.
Extended self-test routine
recommended polling time: ( 355) minutes.
Conveyance self-test routine
recommended polling time: ( 5) minutes.
SCT capabilities: (0x3035) SCT Status supported.
SCT Feature Control supported.
SCT Data Table supported.
SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAGS VALUE WORST THRESH FAIL RAW_VALUE
1 Raw_Read_Error_Rate POSR-K 200 200 051 - 0
3 Spin_Up_Time POS--K 172 162 021 - 6400
4 Start_Stop_Count -O--CK 097 097 000 - 3688
5 Reallocated_Sector_Ct PO--CK 200 200 140 - 0
7 Seek_Error_Rate -OSR-K 200 200 000 - 0
9 Power_On_Hours -O--CK 081 081 000 - 13891
10 Spin_Retry_Count -O--CK 100 100 000 - 0
11 Calibration_Retry_Count -O--CK 100 100 000 - 0
12 Power_Cycle_Count -O--CK 097 097 000 - 3683
192 Power-Off_Retract_Count -O--CK 200 200 000 - 50
193 Load_Cycle_Count -O--CK 001 001 000 - 912124
194 Temperature_Celsius -O---K 120 109 000 - 30
196 Reallocated_Event_Count -O--CK 200 200 000 - 0
197 Current_Pending_Sector -O--CK 200 200 000 - 0
198 Offline_Uncorrectable ----CK 200 200 000 - 0
199 UDMA_CRC_Error_Count -O--CK 200 200 000 - 0
200 Multi_Zone_Error_Rate ---R-- 200 200 000 - 0
||||||_ K auto-keep
|||||__ C event count
||||___ R error rate
|||____ S speed/performance
||_____ O updated online
|______ P prefailure warning
General Purpose Log Directory Version 1
SMART Log Directory Version 1 [multi-sector log support]
Address Access R/W Size Description
0x00 GPL,SL R/O 1 Log Directory
0x01 SL R/O 1 Summary SMART error log
0x02 SL R/O 5 Comprehensive SMART error log
0x03 GPL R/O 6 Ext. Comprehensive SMART error log
0x06 SL R/O 1 SMART self-test log
0x07 GPL R/O 1 Extended self-test log
0x09 SL R/W 1 Selective self-test log
0x10 GPL R/O 1 NCQ Command Error log
0x11 GPL R/O 1 SATA Phy Event Counters
0x80-0x9f GPL,SL R/W 16 Host vendor specific log
0xa0-0xa7 GPL,SL VS 16 Device vendor specific log
0xa8-0xb7 GPL,SL VS 1 Device vendor specific log
0xbd GPL,SL VS 1 Device vendor specific log
0xc0 GPL,SL VS 1 Device vendor specific log
0xc1 GPL VS 93 Device vendor specific log
0xe0 GPL,SL R/W 1 SCT Command/Status
0xe1 GPL,SL R/W 1 SCT Data Transfer
SMART Extended Comprehensive Error Log Version: 1 (6 sectors)
No Errors Logged
SMART Extended Self-test Log Version: 1 (1 sectors)
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Short offline Completed without error 00% 13888 -
SMART Selective self-test log data structure revision number 1
SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS
1 0 0 Not_testing
2 0 0 Not_testing
3 0 0 Not_testing
4 0 0 Not_testing
5 0 0 Not_testing
Selective self-test flags (0x0):
After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.
SCT Status Version: 3
SCT Version (vendor specific): 258 (0x0102)
SCT Support Level: 1
Device State: Active (0)
Current Temperature: 30 Celsius
Power Cycle Min/Max Temperature: 22/30 Celsius
Lifetime Min/Max Temperature: 22/41 Celsius
Under/Over Temperature Limit Count: 0/0
SCT Temperature History Version: 2
Temperature Sampling Period: 1 minute
Temperature Logging Interval: 1 minute
Min/Max recommended Temperature: 0/60 Celsius
Min/Max Temperature Limit: -41/85 Celsius
Temperature History Size (Index): 478 (462)
Index Estimated Time Temperature Celsius
463 2016-11-13 12:33 31 ************
... ..( 16 skipped). .. ************
2 2016-11-13 12:50 31 ************
3 2016-11-13 12:51 30 ***********
4 2016-11-13 12:52 30 ***********
5 2016-11-13 12:53 30 ***********
6 2016-11-13 12:54 31 ************
7 2016-11-13 12:55 30 ***********
... ..( 21 skipped). .. ***********
29 2016-11-13 13:17 30 ***********
30 2016-11-13 13:18 31 ************
... ..( 14 skipped). .. ************
45 2016-11-13 13:33 31 ************
46 2016-11-13 13:34 ? -
47 2016-11-13 13:35 22 ***
48 2016-11-13 13:36 22 ***
49 2016-11-13 13:37 22 ***
50 2016-11-13 13:38 23 ****
51 2016-11-13 13:39 23 ****
52 2016-11-13 13:40 24 *****
53 2016-11-13 13:41 24 *****
54 2016-11-13 13:42 25 ******
55 2016-11-13 13:43 25 ******
56 2016-11-13 13:44 26 *******
... ..( 2 skipped). .. *******
59 2016-11-13 13:47 26 *******
60 2016-11-13 13:48 27 ********
... ..( 4 skipped). .. ********
65 2016-11-13 13:53 27 ********
66 2016-11-13 13:54 28 *********
... ..( 4 skipped). .. *********
71 2016-11-13 13:59 28 *********
72 2016-11-13 14:00 29 **********
... ..( 19 skipped). .. **********
92 2016-11-13 14:20 29 **********
93 2016-11-13 14:21 30 ***********
... ..( 12 skipped). .. ***********
106 2016-11-13 14:34 30 ***********
107 2016-11-13 14:35 31 ************
... ..( 2 skipped). .. ************
110 2016-11-13 14:38 31 ************
111 2016-11-13 14:39 32 *************
112 2016-11-13 14:40 31 ************
... ..( 18 skipped). .. ************
131 2016-11-13 14:59 31 ************
132 2016-11-13 15:00 32 *************
133 2016-11-13 15:01 32 *************
134 2016-11-13 15:02 31 ************
135 2016-11-13 15:03 32 *************
136 2016-11-13 15:04 31 ************
... ..( 10 skipped). .. ************
147 2016-11-13 15:15 31 ************
148 2016-11-13 15:16 32 *************
149 2016-11-13 15:17 31 ************
150 2016-11-13 15:18 31 ************
151 2016-11-13 15:19 32 *************
152 2016-11-13 15:20 31 ************
... ..( 10 skipped). .. ************
163 2016-11-13 15:31 31 ************
164 2016-11-13 15:32 ? -
165 2016-11-13 15:33 20 *
166 2016-11-13 15:34 21 **
167 2016-11-13 15:35 21 **
168 2016-11-13 15:36 21 **
169 2016-11-13 15:37 22 ***
170 2016-11-13 15:38 22 ***
171 2016-11-13 15:39 22 ***
172 2016-11-13 15:40 23 ****
173 2016-11-13 15:41 24 *****
174 2016-11-13 15:42 24 *****
175 2016-11-13 15:43 24 *****
176 2016-11-13 15:44 25 ******
177 2016-11-13 15:45 25 ******
178 2016-11-13 15:46 25 ******
179 2016-11-13 15:47 26 *******
... ..( 2 skipped). .. *******
182 2016-11-13 15:50 26 *******
183 2016-11-13 15:51 27 ********
... ..( 7 skipped). .. ********
191 2016-11-13 15:59 27 ********
192 2016-11-13 16:00 28 *********
... ..( 4 skipped). .. *********
197 2016-11-13 16:05 28 *********
198 2016-11-13 16:06 29 **********
... ..( 13 skipped). .. **********
212 2016-11-13 16:20 29 **********
213 2016-11-13 16:21 30 ***********
... ..( 5 skipped). .. ***********
219 2016-11-13 16:27 30 ***********
220 2016-11-13 16:28 31 ************
221 2016-11-13 16:29 31 ************
222 2016-11-13 16:30 31 ************
223 2016-11-13 16:31 30 ***********
224 2016-11-13 16:32 30 ***********
225 2016-11-13 16:33 31 ************
... ..( 2 skipped). .. ************
228 2016-11-13 16:36 31 ************
229 2016-11-13 16:37 30 ***********
... ..( 5 skipped). .. ***********
235 2016-11-13 16:43 30 ***********
236 2016-11-13 16:44 31 ************
237 2016-11-13 16:45 30 ***********
... ..( 8 skipped). .. ***********
246 2016-11-13 16:54 30 ***********
247 2016-11-13 16:55 31 ************
248 2016-11-13 16:56 30 ***********
... ..( 9 skipped). .. ***********
258 2016-11-13 17:06 30 ***********
259 2016-11-13 17:07 31 ************
260 2016-11-13 17:08 30 ***********
... ..( 8 skipped). .. ***********
269 2016-11-13 17:17 30 ***********
270 2016-11-13 17:18 31 ************
271 2016-11-13 17:19 31 ************
272 2016-11-13 17:20 31 ************
273 2016-11-13 17:21 30 ***********
274 2016-11-13 17:22 30 ***********
275 2016-11-13 17:23 31 ************
276 2016-11-13 17:24 31 ************
277 2016-11-13 17:25 30 ***********
... ..( 7 skipped). .. ***********
285 2016-11-13 17:33 30 ***********
286 2016-11-13 17:34 31 ************
... ..( 17 skipped). .. ************
304 2016-11-13 17:52 31 ************
305 2016-11-13 17:53 30 ***********
306 2016-11-13 17:54 31 ************
307 2016-11-13 17:55 30 ***********
... ..( 5 skipped). .. ***********
313 2016-11-13 18:01 30 ***********
314 2016-11-13 18:02 31 ************
315 2016-11-13 18:03 31 ************
316 2016-11-13 18:04 30 ***********
... ..( 3 skipped). .. ***********
320 2016-11-13 18:08 30 ***********
321 2016-11-13 18:09 31 ************
... ..( 11 skipped). .. ************
333 2016-11-13 18:21 31 ************
334 2016-11-13 18:22 32 *************
335 2016-11-13 18:23 31 ************
336 2016-11-13 18:24 31 ************
337 2016-11-13 18:25 32 *************
338 2016-11-13 18:26 31 ************
... ..( 11 skipped). .. ************
350 2016-11-13 18:38 31 ************
351 2016-11-13 18:39 32 *************
352 2016-11-13 18:40 31 ************
... ..( 5 skipped). .. ************
358 2016-11-13 18:46 31 ************
359 2016-11-13 18:47 32 *************
360 2016-11-13 18:48 32 *************
361 2016-11-13 18:49 32 *************
362 2016-11-13 18:50 31 ************
... ..( 14 skipped). .. ************
377 2016-11-13 19:05 31 ************
378 2016-11-13 19:06 30 ***********
379 2016-11-13 19:07 31 ************
380 2016-11-13 19:08 30 ***********
... ..( 4 skipped). .. ***********
385 2016-11-13 19:13 30 ***********
386 2016-11-13 19:14 31 ************
387 2016-11-13 19:15 31 ************
388 2016-11-13 19:16 30 ***********
... ..( 4 skipped). .. ***********
393 2016-11-13 19:21 30 ***********
394 2016-11-13 19:22 31 ************
395 2016-11-13 19:23 30 ***********
396 2016-11-13 19:24 31 ************
... ..( 10 skipped). .. ************
407 2016-11-13 19:35 31 ************
408 2016-11-13 19:36 32 *************
409 2016-11-13 19:37 ? -
410 2016-11-13 19:38 32 *************
411 2016-11-13 19:39 31 ************
... ..( 2 skipped). .. ************
414 2016-11-13 19:42 31 ************
415 2016-11-13 19:43 32 *************
416 2016-11-13 19:44 31 ************
417 2016-11-13 19:45 32 *************
418 2016-11-13 19:46 31 ************
419 2016-11-13 19:47 32 *************
420 2016-11-13 19:48 31 ************
421 2016-11-13 19:49 32 *************
422 2016-11-13 19:50 31 ************
423 2016-11-13 19:51 31 ************
424 2016-11-13 19:52 31 ************
425 2016-11-13 19:53 32 *************
426 2016-11-13 19:54 31 ************
... ..( 4 skipped). .. ************
431 2016-11-13 19:59 31 ************
432 2016-11-13 20:00 32 *************
433 2016-11-13 20:01 32 *************
434 2016-11-13 20:02 31 ************
... ..( 4 skipped). .. ************
439 2016-11-13 20:07 31 ************
440 2016-11-13 20:08 32 *************
441 2016-11-13 20:09 31 ************
442 2016-11-13 20:10 32 *************
443 2016-11-13 20:11 31 ************
... ..( 2 skipped). .. ************
446 2016-11-13 20:14 31 ************
447 2016-11-13 20:15 32 *************
448 2016-11-13 20:16 31 ************
... ..( 5 skipped). .. ************
454 2016-11-13 20:22 31 ************
455 2016-11-13 20:23 32 *************
456 2016-11-13 20:24 31 ************
457 2016-11-13 20:25 32 *************
458 2016-11-13 20:26 32 *************
459 2016-11-13 20:27 31 ************
... ..( 2 skipped). .. ************
462 2016-11-13 20:30 31 ************
SCT Error Recovery Control command not supported
Device Statistics (GP Log 0x04) not supported
SATA Phy Event Counters (GP Log 0x11)
ID Size Value Description
0x0001 2 0 Command failed due to ICRC error
0x0002 2 0 R_ERR response for data FIS
0x0003 2 0 R_ERR response for device-to-host data FIS
0x0004 2 0 R_ERR response for host-to-device data FIS
0x0005 2 0 R_ERR response for non-data FIS
0x0006 2 0 R_ERR response for device-to-host non-data FIS
0x0007 2 0 R_ERR response for host-to-device non-data FIS
0x000a 2 3 Device-to-host register FISes sent due to a COMRESET
0x000b 2 0 CRC errors within host-to-device FIS
0x8000 4 3578 Vendor specific
[-- Attachment #3: sdb.txt --]
[-- Type: text/plain, Size: 17448 bytes --]
smartctl 6.2 2013-07-26 r3841 [x86_64-linux-4.4.0-47-generic] (local build)
Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org
=== START OF INFORMATION SECTION ===
Model Family: Western Digital Caviar Green (AF, SATA 6Gb/s)
Device Model: WDC WD20EARX-00PASB0
Serial Number: WD-WCAZA9552721
LU WWN Device Id: 5 0014ee 2b0e3f07a
Firmware Version: 51.0AB51
User Capacity: 2,000,398,934,016 bytes [2.00 TB]
Sector Sizes: 512 bytes logical, 4096 bytes physical
Device is: In smartctl database [for details use: -P show]
ATA Version is: ATA8-ACS (minor revision not indicated)
SATA Version is: SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is: Sun Nov 13 20:30:01 2016 SAST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
AAM feature is: Unavailable
APM feature is: Unavailable
Rd look-ahead is: Enabled
Write cache is: Enabled
ATA Security is: Disabled, NOT FROZEN [SEC1]
Wt Cache Reorder: Enabled
=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED
General SMART Values:
Offline data collection status: (0x84) Offline data collection activity
was suspended by an interrupting command from host.
Auto Offline Data Collection: Enabled.
Self-test execution status: ( 0) The previous self-test routine completed
without error or no self-test has ever
been run.
Total time to complete Offline
data collection: (37260) seconds.
Offline data collection
capabilities: (0x7b) SMART execute Offline immediate.
Auto Offline data collection on/off support.
Suspend Offline collection upon new
command.
Offline surface scan supported.
Self-test supported.
Conveyance Self-test supported.
Selective Self-test supported.
SMART capabilities: (0x0003) Saves SMART data before entering
power-saving mode.
Supports SMART auto save timer.
Error logging capability: (0x01) Error logging supported.
General Purpose Logging supported.
Short self-test routine
recommended polling time: ( 2) minutes.
Extended self-test routine
recommended polling time: ( 360) minutes.
Conveyance self-test routine
recommended polling time: ( 5) minutes.
SCT capabilities: (0x3035) SCT Status supported.
SCT Feature Control supported.
SCT Data Table supported.
SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAGS VALUE WORST THRESH FAIL RAW_VALUE
1 Raw_Read_Error_Rate POSR-K 200 200 051 - 0
3 Spin_Up_Time POS--K 168 162 021 - 6566
4 Start_Stop_Count -O--CK 097 097 000 - 3748
5 Reallocated_Sector_Ct PO--CK 200 200 140 - 0
7 Seek_Error_Rate -OSR-K 200 200 000 - 0
9 Power_On_Hours -O--CK 081 081 000 - 13883
10 Spin_Retry_Count -O--CK 100 100 000 - 0
11 Calibration_Retry_Count -O--CK 100 100 000 - 0
12 Power_Cycle_Count -O--CK 097 097 000 - 3683
192 Power-Off_Retract_Count -O--CK 200 200 000 - 49
193 Load_Cycle_Count -O--CK 001 001 000 - 837570
194 Temperature_Celsius -O---K 119 108 000 - 31
196 Reallocated_Event_Count -O--CK 200 200 000 - 0
197 Current_Pending_Sector -O--CK 200 001 000 - 3
198 Offline_Uncorrectable ----CK 200 200 000 - 4
199 UDMA_CRC_Error_Count -O--CK 200 200 000 - 0
200 Multi_Zone_Error_Rate ---R-- 200 200 000 - 3
||||||_ K auto-keep
|||||__ C event count
||||___ R error rate
|||____ S speed/performance
||_____ O updated online
|______ P prefailure warning
General Purpose Log Directory Version 1
SMART Log Directory Version 1 [multi-sector log support]
Address Access R/W Size Description
0x00 GPL,SL R/O 1 Log Directory
0x01 SL R/O 1 Summary SMART error log
0x02 SL R/O 5 Comprehensive SMART error log
0x03 GPL R/O 6 Ext. Comprehensive SMART error log
0x06 SL R/O 1 SMART self-test log
0x07 GPL R/O 1 Extended self-test log
0x09 SL R/W 1 Selective self-test log
0x10 GPL R/O 1 NCQ Command Error log
0x11 GPL R/O 1 SATA Phy Event Counters
0x80-0x9f GPL,SL R/W 16 Host vendor specific log
0xa0-0xa7 GPL,SL VS 16 Device vendor specific log
0xa8-0xb7 GPL,SL VS 1 Device vendor specific log
0xbd GPL,SL VS 1 Device vendor specific log
0xc0 GPL,SL VS 1 Device vendor specific log
0xc1 GPL VS 93 Device vendor specific log
0xe0 GPL,SL R/W 1 SCT Command/Status
0xe1 GPL,SL R/W 1 SCT Data Transfer
SMART Extended Comprehensive Error Log Version: 1 (6 sectors)
Device Error Count: 2
CR = Command Register
FEATR = Features Register
COUNT = Count (was: Sector Count) Register
LBA_48 = Upper bytes of LBA High/Mid/Low Registers ] ATA-8
LH = LBA High (was: Cylinder High) Register ] LBA
LM = LBA Mid (was: Cylinder Low) Register ] Register
LL = LBA Low (was: Sector Number) Register ]
DV = Device (was: Device/Head) Register
DC = Device Control Register
ER = Error register
ST = Status register
Powered_Up_Time is measured from power on, and printed as
DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes,
SS=sec, and sss=millisec. It "wraps" after 49.710 days.
Error 2 [1] occurred at disk power-on lifetime: 9505 hours (396 days + 1 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER -- ST COUNT LBA_48 LH LM LL DV DC
-- -- -- == -- == == == -- -- -- -- --
40 -- 51 00 00 00 00 25 b3 05 58 40 00 Error: UNC at LBA = 0x25b30558 = 632489304
Commands leading to the command that caused the error were:
CR FEATR COUNT LBA_48 LH LM LL DV DC Powered_Up_Time Command/Feature_Name
-- == -- == -- == == == -- -- -- -- -- --------------- --------------------
60 00 80 00 48 00 00 25 b3 0b 80 40 08 17:33:58.845 READ FPDMA QUEUED
60 00 80 00 50 00 00 25 b3 0b 00 40 08 17:33:58.845 READ FPDMA QUEUED
60 00 80 00 58 00 00 25 b3 0a 80 40 08 17:33:58.844 READ FPDMA QUEUED
60 00 80 00 60 00 00 25 b3 0a 00 40 08 17:33:58.844 READ FPDMA QUEUED
60 00 80 00 80 00 00 25 b3 09 80 40 08 17:33:58.832 READ FPDMA QUEUED
Error 1 [0] occurred at disk power-on lifetime: 9505 hours (396 days + 1 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER -- ST COUNT LBA_48 LH LM LL DV DC
-- -- -- == -- == == == -- -- -- -- --
40 -- 51 00 fe 00 00 25 b2 fa 58 40 00 Error: UNC at LBA = 0x25b2fa58 = 632486488
Commands leading to the command that caused the error were:
CR FEATR COUNT LBA_48 LH LM LL DV DC Powered_Up_Time Command/Feature_Name
-- == -- == -- == == == -- -- -- -- -- --------------- --------------------
60 00 80 00 d8 00 00 25 b2 ff 00 40 08 17:33:56.041 READ FPDMA QUEUED
60 00 80 00 d0 00 00 25 b2 fe 80 40 08 17:33:56.041 READ FPDMA QUEUED
60 00 80 00 c8 00 00 25 b2 fe 00 40 08 17:33:56.041 READ FPDMA QUEUED
60 00 80 00 c0 00 00 25 b2 fd 80 40 08 17:33:56.041 READ FPDMA QUEUED
60 00 80 00 b8 00 00 25 b2 fd 00 40 08 17:33:56.040 READ FPDMA QUEUED
SMART Extended Self-test Log Version: 1 (1 sectors)
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Short offline Completed without error 00% 13880 -
# 2 Short offline Aborted by host 10% 13880 -
# 3 Short offline Interrupted (host reset) 10% 13880 -
SMART Selective self-test log data structure revision number 1
SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS
1 0 0 Not_testing
2 0 0 Not_testing
3 0 0 Not_testing
4 0 0 Not_testing
5 0 0 Not_testing
Selective self-test flags (0x0):
After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.
SCT Status Version: 3
SCT Version (vendor specific): 258 (0x0102)
SCT Support Level: 1
Device State: Active (0)
Current Temperature: 31 Celsius
Power Cycle Min/Max Temperature: 22/31 Celsius
Lifetime Min/Max Temperature: 21/42 Celsius
Under/Over Temperature Limit Count: 0/0
SCT Temperature History Version: 2
Temperature Sampling Period: 1 minute
Temperature Logging Interval: 1 minute
Min/Max recommended Temperature: 0/60 Celsius
Min/Max Temperature Limit: -41/85 Celsius
Temperature History Size (Index): 478 (433)
Index Estimated Time Temperature Celsius
434 2016-11-13 12:33 32 *************
... ..( 15 skipped). .. *************
450 2016-11-13 12:49 32 *************
451 2016-11-13 12:50 31 ************
452 2016-11-13 12:51 32 *************
... ..( 2 skipped). .. *************
455 2016-11-13 12:54 32 *************
456 2016-11-13 12:55 31 ************
... ..( 16 skipped). .. ************
473 2016-11-13 13:12 31 ************
474 2016-11-13 13:13 32 *************
475 2016-11-13 13:14 32 *************
476 2016-11-13 13:15 32 *************
477 2016-11-13 13:16 31 ************
0 2016-11-13 13:17 31 ************
1 2016-11-13 13:18 32 *************
... ..( 14 skipped). .. *************
16 2016-11-13 13:33 32 *************
17 2016-11-13 13:34 ? -
18 2016-11-13 13:35 22 ***
19 2016-11-13 13:36 22 ***
20 2016-11-13 13:37 22 ***
21 2016-11-13 13:38 23 ****
22 2016-11-13 13:39 24 *****
23 2016-11-13 13:40 24 *****
24 2016-11-13 13:41 25 ******
25 2016-11-13 13:42 25 ******
26 2016-11-13 13:43 25 ******
27 2016-11-13 13:44 26 *******
28 2016-11-13 13:45 26 *******
29 2016-11-13 13:46 27 ********
... ..( 5 skipped). .. ********
35 2016-11-13 13:52 27 ********
36 2016-11-13 13:53 28 *********
37 2016-11-13 13:54 28 *********
38 2016-11-13 13:55 29 **********
39 2016-11-13 13:56 28 *********
40 2016-11-13 13:57 29 **********
... ..( 11 skipped). .. **********
52 2016-11-13 14:09 29 **********
53 2016-11-13 14:10 30 ***********
... ..( 11 skipped). .. ***********
65 2016-11-13 14:22 30 ***********
66 2016-11-13 14:23 31 ************
... ..( 10 skipped). .. ************
77 2016-11-13 14:34 31 ************
78 2016-11-13 14:35 32 *************
... ..( 26 skipped). .. *************
105 2016-11-13 15:02 32 *************
106 2016-11-13 15:03 33 **************
107 2016-11-13 15:04 32 *************
... ..( 10 skipped). .. *************
118 2016-11-13 15:15 32 *************
119 2016-11-13 15:16 33 **************
120 2016-11-13 15:17 32 *************
121 2016-11-13 15:18 32 *************
122 2016-11-13 15:19 33 **************
123 2016-11-13 15:20 32 *************
... ..( 2 skipped). .. *************
126 2016-11-13 15:23 32 *************
127 2016-11-13 15:24 33 **************
128 2016-11-13 15:25 32 *************
... ..( 5 skipped). .. *************
134 2016-11-13 15:31 32 *************
135 2016-11-13 15:32 ? -
136 2016-11-13 15:33 21 **
137 2016-11-13 15:34 21 **
138 2016-11-13 15:35 21 **
139 2016-11-13 15:36 22 ***
140 2016-11-13 15:37 22 ***
141 2016-11-13 15:38 23 ****
142 2016-11-13 15:39 23 ****
143 2016-11-13 15:40 23 ****
144 2016-11-13 15:41 24 *****
145 2016-11-13 15:42 25 ******
... ..( 2 skipped). .. ******
148 2016-11-13 15:45 25 ******
149 2016-11-13 15:46 26 *******
150 2016-11-13 15:47 26 *******
151 2016-11-13 15:48 26 *******
152 2016-11-13 15:49 27 ********
... ..( 5 skipped). .. ********
158 2016-11-13 15:55 27 ********
159 2016-11-13 15:56 28 *********
... ..( 5 skipped). .. *********
165 2016-11-13 16:02 28 *********
166 2016-11-13 16:03 29 **********
... ..( 10 skipped). .. **********
177 2016-11-13 16:14 29 **********
178 2016-11-13 16:15 30 ***********
... ..( 6 skipped). .. ***********
185 2016-11-13 16:22 30 ***********
186 2016-11-13 16:23 31 ************
... ..( 9 skipped). .. ************
196 2016-11-13 16:33 31 ************
197 2016-11-13 16:34 32 *************
... ..( 8 skipped). .. *************
206 2016-11-13 16:43 32 *************
207 2016-11-13 16:44 31 ************
208 2016-11-13 16:45 31 ************
209 2016-11-13 16:46 31 ************
210 2016-11-13 16:47 32 *************
211 2016-11-13 16:48 31 ************
... ..( 2 skipped). .. ************
214 2016-11-13 16:51 31 ************
215 2016-11-13 16:52 32 *************
216 2016-11-13 16:53 32 *************
217 2016-11-13 16:54 31 ************
218 2016-11-13 16:55 32 *************
219 2016-11-13 16:56 31 ************
... ..( 3 skipped). .. ************
223 2016-11-13 17:00 31 ************
224 2016-11-13 17:01 32 *************
225 2016-11-13 17:02 32 *************
226 2016-11-13 17:03 31 ************
... ..( 2 skipped). .. ************
229 2016-11-13 17:06 31 ************
230 2016-11-13 17:07 32 *************
231 2016-11-13 17:08 32 *************
232 2016-11-13 17:09 31 ************
... ..( 2 skipped). .. ************
235 2016-11-13 17:12 31 ************
236 2016-11-13 17:13 32 *************
... ..( 39 skipped). .. *************
276 2016-11-13 17:53 32 *************
277 2016-11-13 17:54 31 ************
... ..( 7 skipped). .. ************
285 2016-11-13 18:02 31 ************
286 2016-11-13 18:03 32 *************
287 2016-11-13 18:04 32 *************
288 2016-11-13 18:05 31 ************
289 2016-11-13 18:06 32 *************
... ..( 14 skipped). .. *************
304 2016-11-13 18:21 32 *************
305 2016-11-13 18:22 33 **************
306 2016-11-13 18:23 32 *************
307 2016-11-13 18:24 32 *************
308 2016-11-13 18:25 33 **************
309 2016-11-13 18:26 32 *************
... ..( 19 skipped). .. *************
329 2016-11-13 18:46 32 *************
330 2016-11-13 18:47 33 **************
331 2016-11-13 18:48 32 *************
332 2016-11-13 18:49 32 *************
333 2016-11-13 18:50 33 **************
334 2016-11-13 18:51 32 *************
... ..( 9 skipped). .. *************
344 2016-11-13 19:01 32 *************
345 2016-11-13 19:02 31 ************
... ..( 16 skipped). .. ************
362 2016-11-13 19:19 31 ************
363 2016-11-13 19:20 32 *************
... ..( 15 skipped). .. *************
379 2016-11-13 19:36 32 *************
380 2016-11-13 19:37 ? -
381 2016-11-13 19:38 33 **************
382 2016-11-13 19:39 32 *************
... ..( 29 skipped). .. *************
412 2016-11-13 20:09 32 *************
413 2016-11-13 20:10 33 **************
... ..( 14 skipped). .. **************
428 2016-11-13 20:25 33 **************
429 2016-11-13 20:26 32 *************
430 2016-11-13 20:27 33 **************
431 2016-11-13 20:28 32 *************
432 2016-11-13 20:29 32 *************
433 2016-11-13 20:30 32 *************
SCT Error Recovery Control command not supported
Device Statistics (GP Log 0x04) not supported
SATA Phy Event Counters (GP Log 0x11)
ID Size Value Description
0x0001 2 0 Command failed due to ICRC error
0x0002 2 0 R_ERR response for data FIS
0x0003 2 0 R_ERR response for device-to-host data FIS
0x0004 2 0 R_ERR response for host-to-device data FIS
0x0005 2 0 R_ERR response for non-data FIS
0x0006 2 0 R_ERR response for device-to-host non-data FIS
0x0007 2 0 R_ERR response for host-to-device non-data FIS
0x000a 2 2 Device-to-host register FISes sent due to a COMRESET
0x000b 2 0 CRC errors within host-to-device FIS
0x8000 4 3575 Vendor specific
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-13 18:46 What to do about Offline_Uncorrectable and Pending_Sector in RAID1 Bruce Merry
@ 2016-11-13 20:18 ` Anthony Youngman
2016-11-13 20:51 ` Bruce Merry
0 siblings, 1 reply; 17+ messages in thread
From: Anthony Youngman @ 2016-11-13 20:18 UTC (permalink / raw)
To: Bruce Merry, linux-raid
Quick first response ...
On 13/11/16 18:46, Bruce Merry wrote:
> Hi
>
> I'm running software RAID1 across two drives in my home machine (LVM
> on LUKS on RAID1). I've just installed smartmontools and run short
> tests, and promptly received emails to tell me that one of the drives
> has 4 offline uncorrectable sectors and 3 current pending sectors.
> I've attached smartctl --xall output for sda (good) and sdb (bad).
>
> These drives are pretty old (over 5 years) so I'm going to replace
> them as soon as I have time (and yes, I have backups), but in the
> meantime I'd like advice on:
>
What drives are they? I'm guessing they're hunky-dory, but they don't
fall foul of timeout mismatch, do they?
https://raid.wiki.kernel.org/index.php/Timeout_Mismatch
> 1. What exactly this means. My understanding is that some data has
> been lost (or may have been lost) on the drive, but the drive still
> has spare sectors to remap things once the failed sectors are written
> to. Is that correct?
It may also mean that the four sectors at least, have already been
remapped ... I'll let the experts confirm. The three pending errors
might be where a read has failed but there's not yet been a re-write -
and you won't have noticed because the raid dealt with it.
>
> 2. How can I tell which sectors are problematic? If it's in the swap
> partition I'm far less worried. I can see two LBAs for offline
> uncorrectable errors in the --xall output, but that still leaves
> another two at large.
I don't think you need to be worried at all. It's only a few sectors,
there's no sign of any further trouble? and as it's raided, when the
drive returns an error the raid code will sort it out for you.
>
> 3. Assuming my understanding is correct, and that the sector falls
> within the RAID1 partition on the drive, is there some way I can
> recover the sectors from the other drive in the RAID1? As a last
> resort I imagine I could wipe the suspect drive and then rebuild it
> from the good one, but I'm hoping there's something less risky I can
> do.
Do a scrub? You've got seven errors total, which some people will say
"panic on the first error" and others will say "so what, the odd error
every now and then is nothing to worry about". The point of a scrub is
it will background-scan the entire array, and if it can't read anything,
it will re-calculate and re-write it.
Just make sure you've not got that timeout problem, or a scrub will make
matters a whole lot worse ...
>
> Thanks in advance
> Bruce
>
Cheers,
Wol
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-13 20:18 ` Anthony Youngman
@ 2016-11-13 20:51 ` Bruce Merry
2016-11-13 21:06 ` Wols Lists
0 siblings, 1 reply; 17+ messages in thread
From: Bruce Merry @ 2016-11-13 20:51 UTC (permalink / raw)
To: Anthony Youngman; +Cc: linux-raid
On 13 November 2016 at 22:18, Anthony Youngman <antlists@youngman.org.uk> wrote:
> Quick first response ...
>
> On 13/11/16 18:46, Bruce Merry wrote:
>>
>> Hi
>>
>> I'm running software RAID1 across two drives in my home machine (LVM
>> on LUKS on RAID1). I've just installed smartmontools and run short
>> tests, and promptly received emails to tell me that one of the drives
>> has 4 offline uncorrectable sectors and 3 current pending sectors.
>> I've attached smartctl --xall output for sda (good) and sdb (bad).
>>
>> These drives are pretty old (over 5 years) so I'm going to replace
>> them as soon as I have time (and yes, I have backups), but in the
>> meantime I'd like advice on:
>>
> What drives are they? I'm guessing they're hunky-dory, but they don't fall
> foul of timeout mismatch, do they?
>
> https://raid.wiki.kernel.org/index.php/Timeout_Mismatch
smartctl reports "SCT Error Recovery Control command not supported".
Does that mean I should be worried? Is there any way to tell whether a
given drive I can buy online supports it?
>> 1. What exactly this means. My understanding is that some data has
>> been lost (or may have been lost) on the drive, but the drive still
>> has spare sectors to remap things once the failed sectors are written
>> to. Is that correct?
>
>
> It may also mean that the four sectors at least, have already been remapped
> ... I'll let the experts confirm. The three pending errors might be where a
> read has failed but there's not yet been a re-write - and you won't have
> noticed because the raid dealt with it.
I'm guessing nothing has been remapped yet, because the
Reallocated_Sector_Ct and Reallocator_Event_ct are both zero.
>> 3. Assuming my understanding is correct, and that the sector falls
>> within the RAID1 partition on the drive, is there some way I can
>> recover the sectors from the other drive in the RAID1? As a last
>> resort I imagine I could wipe the suspect drive and then rebuild it
>> from the good one, but I'm hoping there's something less risky I can
>> do.
>
>
> Do a scrub? You've got seven errors total, which some people will say "panic
> on the first error" and others will say "so what, the odd error every now
> and then is nothing to worry about". The point of a scrub is it will
> background-scan the entire array, and if it can't read anything, it will
> re-calculate and re-write it.
Yes, that sounds like what I need. Thanks to Google I found
/usr/share/mdadm/checkarray to trigger this. It still has a few hours
to go, but now the bad drive has pending sectors == 65535 (which is
suspiciously power-of-two and I assume means it's actually higher and
is being clamped), and /sys/block/md0/md/mismatch_cnt is currently at
1408. If scrubbing is supposed to rewrite on failed reads I would have
expected pending sectors to go down rather than up, so I'm not sure
what's happening.
Thanks
Bruce
--
Dr Bruce Merry
bmerry <@> gmail <.> com
http://www.brucemerry.org.za/
http://blog.brucemerry.org.za/
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-13 20:51 ` Bruce Merry
@ 2016-11-13 21:06 ` Wols Lists
2016-11-13 23:03 ` Phil Turmel
2016-11-14 15:52 ` Bruce Merry
0 siblings, 2 replies; 17+ messages in thread
From: Wols Lists @ 2016-11-13 21:06 UTC (permalink / raw)
To: Bruce Merry; +Cc: linux-raid
On 13/11/16 20:51, Bruce Merry wrote:
> On 13 November 2016 at 22:18, Anthony Youngman <antlists@youngman.org.uk> wrote:
>> Quick first response ...
>>
>> On 13/11/16 18:46, Bruce Merry wrote:
>>>
>>> Hi
>>>
>>> I'm running software RAID1 across two drives in my home machine (LVM
>>> on LUKS on RAID1). I've just installed smartmontools and run short
>>> tests, and promptly received emails to tell me that one of the drives
>>> has 4 offline uncorrectable sectors and 3 current pending sectors.
>>> I've attached smartctl --xall output for sda (good) and sdb (bad).
>>>
>>> These drives are pretty old (over 5 years) so I'm going to replace
>>> them as soon as I have time (and yes, I have backups), but in the
>>> meantime I'd like advice on:
>>>
>> What drives are they? I'm guessing they're hunky-dory, but they don't fall
>> foul of timeout mismatch, do they?
>>
>> https://raid.wiki.kernel.org/index.php/Timeout_Mismatch
>
> smartctl reports "SCT Error Recovery Control command not supported".
> Does that mean I should be worried? Is there any way to tell whether a
> given drive I can buy online supports it?
You need drives that explicitly support raid. WD Reds, Seagate NAS, some
Toshibas - my 2TB laptop drive does ... Try and find a friend with a
drive you like, and check it out, or ask on this list :-)
Did you run that script to increase the kernel timeout?
>
>>> 1. What exactly this means. My understanding is that some data has
>>> been lost (or may have been lost) on the drive, but the drive still
>>> has spare sectors to remap things once the failed sectors are written
>>> to. Is that correct?
>>
>>
>> It may also mean that the four sectors at least, have already been remapped
>> ... I'll let the experts confirm. The three pending errors might be where a
>> read has failed but there's not yet been a re-write - and you won't have
>> noticed because the raid dealt with it.
>
> I'm guessing nothing has been remapped yet, because the
> Reallocated_Sector_Ct and Reallocator_Event_ct are both zero.
>
>>> 3. Assuming my understanding is correct, and that the sector falls
>>> within the RAID1 partition on the drive, is there some way I can
>>> recover the sectors from the other drive in the RAID1? As a last
>>> resort I imagine I could wipe the suspect drive and then rebuild it
>>> from the good one, but I'm hoping there's something less risky I can
>>> do.
>>
>>
>> Do a scrub? You've got seven errors total, which some people will say "panic
>> on the first error" and others will say "so what, the odd error every now
>> and then is nothing to worry about". The point of a scrub is it will
>> background-scan the entire array, and if it can't read anything, it will
>> re-calculate and re-write it.
>
> Yes, that sounds like what I need. Thanks to Google I found
> /usr/share/mdadm/checkarray to trigger this. It still has a few hours
> to go, but now the bad drive has pending sectors == 65535 (which is
> suspiciously power-of-two and I assume means it's actually higher and
> is being clamped), and /sys/block/md0/md/mismatch_cnt is currently at
> 1408. If scrubbing is supposed to rewrite on failed reads I would have
> expected pending sectors to go down rather than up, so I'm not sure
> what's happening.
>
Ummm....
Sounds like that drive could need replacing. I'd get a new drive and do
that as soon as possible - use the --replace option of mdadm - don't
fail the old drive and add the new. Dunno where you're based, but 5mins
on the internet ordering a new drive is probably time well spent.
Note that Seagate Barracudas don't have the best of reputations if
they're the drive you've already got, and the 3TB drives are best
avoided. Sod's law, I've got two of them ...
Advice I always give ... if you're getting new drives, always consider
increasing capacity. I don't know what size your current drives are, but
look at prices of drives a bit larger than what they are, and is it
worth paying the extra?
If you do get bigger drives, there's nothing stopping you making the
paritions on it bigger before you add them in to the array. It'll be
wasted space until you increase the size of all the drives, but once
you've replaced both drives, you can use mdadm to increase the array
size. I don't know about LUKS, but I would expect you can grow that, and
then you can expand your data partitions within that.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-13 21:06 ` Wols Lists
@ 2016-11-13 23:03 ` Phil Turmel
2016-11-14 6:50 ` Bruce Merry
2016-11-14 15:52 ` Bruce Merry
1 sibling, 1 reply; 17+ messages in thread
From: Phil Turmel @ 2016-11-13 23:03 UTC (permalink / raw)
To: Bruce Merry; +Cc: Wols Lists, linux-raid
Hi Bruce,
On 11/13/2016 04:06 PM, Wols Lists wrote:
> On 13/11/16 20:51, Bruce Merry wrote:
>> On 13 November 2016 at 22:18, Anthony Youngman <antlists@youngman.org.uk> wrote:
>>> Quick first response ...
>>> https://raid.wiki.kernel.org/index.php/Timeout_Mismatch
>>
>> smartctl reports "SCT Error Recovery Control command not supported".
>> Does that mean I should be worried? Is there any way to tell whether a
>> given drive I can buy online supports it?
You should be worried. It is vital for proper MD raid operation that
drive timeouts be shorter than the kernel timeout for that device. If
you can't make the drive timeout short, you *must* make the kernel
timeout long.
> You need drives that explicitly support raid. WD Reds, Seagate NAS, some
> Toshibas - my 2TB laptop drive does ... Try and find a friend with a
> drive you like, and check it out, or ask on this list :-)
Manufacturers' data sheets for system builders usually contain enough
information to determine if ERC is supported. Nowadays, the "NAS"
families work out of the box. And enterprise drives, too, of course.
>>>> 1. What exactly this means. My understanding is that some data has
>>>> been lost (or may have been lost) on the drive, but the drive still
>>>> has spare sectors to remap things once the failed sectors are written
>>>> to. Is that correct?
Generally, yes.
>>> Do a scrub? You've got seven errors total, which some people will say "panic
>>> on the first error" and others will say "so what, the odd error every now
>>> and then is nothing to worry about". The point of a scrub is it will
>>> background-scan the entire array, and if it can't read anything, it will
>>> re-calculate and re-write it.
>>
>> Yes, that sounds like what I need. Thanks to Google I found
>> /usr/share/mdadm/checkarray to trigger this. It still has a few hours
>> to go, but now the bad drive has pending sectors == 65535 (which is
>> suspiciously power-of-two and I assume means it's actually higher and
>> is being clamped), and /sys/block/md0/md/mismatch_cnt is currently at
>> 1408. If scrubbing is supposed to rewrite on failed reads I would have
>> expected pending sectors to go down rather than up, so I'm not sure
>> what's happening.
>>
> Ummm....
>
> Sounds like that drive could need replacing. I'd get a new drive and do
> that as soon as possible - use the --replace option of mdadm - don't
> fail the old drive and add the new. Dunno where you're based, but 5mins
> on the internet ordering a new drive is probably time well spent.
You have two other possibilities:
1) Swap volumes in the raid. These are known to drop unneeded writes
when the data isn't needed, even if it made it to one of the mirrors.
That makes harmless mismatches.
2) Trim. Well-behaved drive firmware guarantees zeros for trimmed
sectors, but many drives return random data instead. Zing, mismatches.
It's often unhelpful with encrypted volumes, as even well-behaved
firmware can't deliver zeroed sectors *inside* the encryption.
I wouldn't panic just yet. The check scrub should (with mitigated
timeouts) fix all of your pending sectors. Then look at your actual
relocations to determine if you really do have a problem.
Phil
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-13 23:03 ` Phil Turmel
@ 2016-11-14 6:50 ` Bruce Merry
2016-11-14 16:01 ` Phil Turmel
0 siblings, 1 reply; 17+ messages in thread
From: Bruce Merry @ 2016-11-14 6:50 UTC (permalink / raw)
To: Phil Turmel; +Cc: Wols Lists, linux-raid
On 14 November 2016 at 01:03, Phil Turmel <philip@turmel.org> wrote:
> Hi Bruce,
>
> On 11/13/2016 04:06 PM, Wols Lists wrote:
>> On 13/11/16 20:51, Bruce Merry wrote:
>>> On 13 November 2016 at 22:18, Anthony Youngman <antlists@youngman.org.uk> wrote:
>>>> Quick first response ...
>
>>>> https://raid.wiki.kernel.org/index.php/Timeout_Mismatch
>>>
>>> smartctl reports "SCT Error Recovery Control command not supported".
>>> Does that mean I should be worried? Is there any way to tell whether a
>>> given drive I can buy online supports it?
>
> You should be worried. It is vital for proper MD raid operation that
> drive timeouts be shorter than the kernel timeout for that device. If
> you can't make the drive timeout short, you *must* make the kernel
> timeout long.
Okay, I'll give that script a go to increase my kernel timeout. If I
understand correctly, it's not the end of the world if the drive
doesn't support SCTERC, provided I have a long kernel timeout (and
when things go wrong it might take much longer to recover, but I can
live with that). Is that correct?
>>> Yes, that sounds like what I need. Thanks to Google I found
>>> /usr/share/mdadm/checkarray to trigger this. It still has a few hours
>>> to go, but now the bad drive has pending sectors == 65535 (which is
>>> suspiciously power-of-two and I assume means it's actually higher and
>>> is being clamped), and /sys/block/md0/md/mismatch_cnt is currently at
>>> 1408. If scrubbing is supposed to rewrite on failed reads I would have
>>> expected pending sectors to go down rather than up, so I'm not sure
>>> what's happening.
>>>
>> Ummm....
>>
>> Sounds like that drive could need replacing. I'd get a new drive and do
>> that as soon as possible - use the --replace option of mdadm - don't
>> fail the old drive and add the new. Dunno where you're based, but 5mins
>> on the internet ordering a new drive is probably time well spent.
Oh don't worry, I wasted no time in ordering new drives already.
> You have two other possibilities:
>
> 1) Swap volumes in the raid. These are known to drop unneeded writes
> when the data isn't needed, even if it made it to one of the mirrors.
> That makes harmless mismatches.
It won't be that - I keep have separate non-RAIDed partitions for swap.
> 2) Trim. Well-behaved drive firmware guarantees zeros for trimmed
> sectors, but many drives return random data instead. Zing, mismatches.
> It's often unhelpful with encrypted volumes, as even well-behaved
> firmware can't deliver zeroed sectors *inside* the encryption.
Weee, sounds like fun. I hope it's that. Is there any way to tell
which blocks mismatch, so that I can tell which files are in trouble
(assuming I can figure out how to map through LVM, LUKS and
debuge2fs).
Bruce
--
Dr Bruce Merry
bmerry <@> gmail <.> com
http://www.brucemerry.org.za/
http://blog.brucemerry.org.za/
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-13 21:06 ` Wols Lists
2016-11-13 23:03 ` Phil Turmel
@ 2016-11-14 15:52 ` Bruce Merry
2016-11-14 15:58 ` Wols Lists
1 sibling, 1 reply; 17+ messages in thread
From: Bruce Merry @ 2016-11-14 15:52 UTC (permalink / raw)
To: Wols Lists; +Cc: linux-raid
On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
> Sounds like that drive could need replacing. I'd get a new drive and do
> that as soon as possible - use the --replace option of mdadm - don't
> fail the old drive and add the new.
Would you mind explaining why I should use --replace instead of taking
out the suspect drive? I guess I lose redundancy for any writes that
occur while the rebuild is happening, but I'd plan to do this with the
filesystem unmounted so there wouldn't be any writes.
What I'd quite like to do is treat the "good" drive as the source for
all the data (unless it turns out to have bad sectors too...), even if
the other drive doesn't have a read error; which I'd achieve by
failing the "bad" drive. I want to do this because after doing one
scrub and starting on a second, I'm still seeing non-zero
mismatch_cnt. Does --replace do anything clever about using the old
drive only when it must, or does it just read from the whole array
like normal, which might mean taking the mismatched data from the bad
drive?
Thanks
Bruce
--
Dr Bruce Merry
bmerry <@> gmail <.> com
http://www.brucemerry.org.za/
http://blog.brucemerry.org.za/
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 15:52 ` Bruce Merry
@ 2016-11-14 15:58 ` Wols Lists
2016-11-14 16:03 ` Bruce Merry
2016-11-15 18:14 ` Peter Sangas
0 siblings, 2 replies; 17+ messages in thread
From: Wols Lists @ 2016-11-14 15:58 UTC (permalink / raw)
To: Bruce Merry; +Cc: linux-raid
On 14/11/16 15:52, Bruce Merry wrote:
> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>> > Sounds like that drive could need replacing. I'd get a new drive and do
>> > that as soon as possible - use the --replace option of mdadm - don't
>> > fail the old drive and add the new.
> Would you mind explaining why I should use --replace instead of taking
> out the suspect drive? I guess I lose redundancy for any writes that
> occur while the rebuild is happening, but I'd plan to do this with the
> filesystem unmounted so there wouldn't be any writes.
Because a replace will copy from the old drive to the new, recovering
any failures from the rest of the array. A fail-and-add will have to
rebuild the entire new array from what's left of the old, stressing the
old array much more.
Okay, in your case, it probably won't make an awful lot of difference,
but it does make you vulnerable to problems on the "good" drive. To
alter your wording slightly, you lose redundancy for writes AND READS
that occur while the array is rebuilding. It's just good practice (and I
point it out because --replace is new and not well known at present).
Cheers,
Wol
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 6:50 ` Bruce Merry
@ 2016-11-14 16:01 ` Phil Turmel
2016-11-14 16:09 ` Bruce Merry
0 siblings, 1 reply; 17+ messages in thread
From: Phil Turmel @ 2016-11-14 16:01 UTC (permalink / raw)
To: Bruce Merry; +Cc: Wols Lists, linux-raid
Hi Bruce,
On 11/14/2016 01:50 AM, Bruce Merry wrote:
> Okay, I'll give that script a go to increase my kernel timeout. If I
> understand correctly, it's not the end of the world if the drive
> doesn't support SCTERC, provided I have a long kernel timeout (and
> when things go wrong it might take much longer to recover, but I can
> live with that). Is that correct?
Yes.
>> 2) Trim. Well-behaved drive firmware guarantees zeros for trimmed
>> sectors, but many drives return random data instead. Zing, mismatches.
>> It's often unhelpful with encrypted volumes, as even well-behaved
>> firmware can't deliver zeroed sectors *inside* the encryption.
>
> Weee, sounds like fun. I hope it's that. Is there any way to tell
> which blocks mismatch, so that I can tell which files are in trouble
> (assuming I can figure out how to map through LVM, LUKS and
> debuge2fs).
The check operation doesn't log the sector addresses, unfortunately. At
least I don't see any such operation in the code that increments
mismatch count. Not even a tracepoint. Hmmm.
In the meantime, run a "repair" scrub instead of a "check" scrub to
affirmatively force no mismatches. (Writes first member of mirrors to
the others.)
Phil
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 15:58 ` Wols Lists
@ 2016-11-14 16:03 ` Bruce Merry
2016-11-14 16:09 ` Wols Lists
2016-11-14 16:10 ` Phil Turmel
2016-11-15 18:14 ` Peter Sangas
1 sibling, 2 replies; 17+ messages in thread
From: Bruce Merry @ 2016-11-14 16:03 UTC (permalink / raw)
To: Wols Lists; +Cc: linux-raid
On 14 November 2016 at 17:58, Wols Lists <antlists@youngman.org.uk> wrote:
> On 14/11/16 15:52, Bruce Merry wrote:
>> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>>> > Sounds like that drive could need replacing. I'd get a new drive and do
>>> > that as soon as possible - use the --replace option of mdadm - don't
>>> > fail the old drive and add the new.
>> Would you mind explaining why I should use --replace instead of taking
>> out the suspect drive? I guess I lose redundancy for any writes that
>> occur while the rebuild is happening, but I'd plan to do this with the
>> filesystem unmounted so there wouldn't be any writes.
>
> Because a replace will copy from the old drive to the new, recovering
> any failures from the rest of the array. A fail-and-add will have to
> rebuild the entire new array from what's left of the old, stressing the
> old array much more.
Okay, I can see how for RAID5 that might be a bad thing.
In my case however, it sounds like --replace will copy everything from
the failing drive, whereas I'd rather it copied everything from the
good drive. Same stress on the array, less chance of copying dodgy
data.
Bruce
--
Dr Bruce Merry
bmerry <@> gmail <.> com
http://www.brucemerry.org.za/
http://blog.brucemerry.org.za/
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 16:03 ` Bruce Merry
@ 2016-11-14 16:09 ` Wols Lists
2016-11-14 16:10 ` Phil Turmel
1 sibling, 0 replies; 17+ messages in thread
From: Wols Lists @ 2016-11-14 16:09 UTC (permalink / raw)
To: Bruce Merry; +Cc: linux-raid
On 14/11/16 16:03, Bruce Merry wrote:
> On 14 November 2016 at 17:58, Wols Lists <antlists@youngman.org.uk> wrote:
>> On 14/11/16 15:52, Bruce Merry wrote:
>>> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>>>>> Sounds like that drive could need replacing. I'd get a new drive and do
>>>>> that as soon as possible - use the --replace option of mdadm - don't
>>>>> fail the old drive and add the new.
>>> Would you mind explaining why I should use --replace instead of taking
>>> out the suspect drive? I guess I lose redundancy for any writes that
>>> occur while the rebuild is happening, but I'd plan to do this with the
>>> filesystem unmounted so there wouldn't be any writes.
>>
>> Because a replace will copy from the old drive to the new, recovering
>> any failures from the rest of the array. A fail-and-add will have to
>> rebuild the entire new array from what's left of the old, stressing the
>> old array much more.
>
> Okay, I can see how for RAID5 that might be a bad thing.
>
> In my case however, it sounds like --replace will copy everything from
> the failing drive, whereas I'd rather it copied everything from the
> good drive. Same stress on the array, less chance of copying dodgy
> data.
>
So long as the data on the drive is correct (it should be) and the drive
reports a fault where it can't read it, it'll only copy good data off
the bad drive. It'll copy it from the other drive if it's dud.
Cheers,
Wol
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 16:01 ` Phil Turmel
@ 2016-11-14 16:09 ` Bruce Merry
2016-11-14 16:14 ` Phil Turmel
0 siblings, 1 reply; 17+ messages in thread
From: Bruce Merry @ 2016-11-14 16:09 UTC (permalink / raw)
To: Phil Turmel; +Cc: Wols Lists, linux-raid
On 14 November 2016 at 18:01, Phil Turmel <philip@turmel.org> wrote:
> In the meantime, run a "repair" scrub instead of a "check" scrub to
> affirmatively force no mismatches. (Writes first member of mirrors to
> the others.)
I think that's good news. I wasn't sure which direction "repair" would
copy, but the good drive is the first member ("Device Role : Active
device 0"). I'll give that a go once the current scrub is done.
Thanks
Bruce
--
Dr Bruce Merry
bmerry <@> gmail <.> com
http://www.brucemerry.org.za/
http://blog.brucemerry.org.za/
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 16:03 ` Bruce Merry
2016-11-14 16:09 ` Wols Lists
@ 2016-11-14 16:10 ` Phil Turmel
1 sibling, 0 replies; 17+ messages in thread
From: Phil Turmel @ 2016-11-14 16:10 UTC (permalink / raw)
To: Bruce Merry, Wols Lists; +Cc: linux-raid
On 11/14/2016 11:03 AM, Bruce Merry wrote:
> On 14 November 2016 at 17:58, Wols Lists <antlists@youngman.org.uk> wrote:
>> On 14/11/16 15:52, Bruce Merry wrote:
>>> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>>>>> Sounds like that drive could need replacing. I'd get a new drive and do
>>>>> that as soon as possible - use the --replace option of mdadm - don't
>>>>> fail the old drive and add the new.
>>> Would you mind explaining why I should use --replace instead of taking
>>> out the suspect drive? I guess I lose redundancy for any writes that
>>> occur while the rebuild is happening, but I'd plan to do this with the
>>> filesystem unmounted so there wouldn't be any writes.
>>
>> Because a replace will copy from the old drive to the new, recovering
>> any failures from the rest of the array. A fail-and-add will have to
>> rebuild the entire new array from what's left of the old, stressing the
>> old array much more.
I entirely endorse Anthony's advice on this one. You are at great risk
of not completing a fail/add resync with the new drive.
> Okay, I can see how for RAID5 that might be a bad thing.
>
> In my case however, it sounds like --replace will copy everything from
> the failing drive, whereas I'd rather it copied everything from the
> good drive. Same stress on the array, less chance of copying dodgy
> data.
You simply don't have that choice, sorry. And drives returning dodgy
data is ungodly rare. The sector checksum algorithms are that good.
You have a URE crisis in your array that is far more significant.
Phil
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 16:09 ` Bruce Merry
@ 2016-11-14 16:14 ` Phil Turmel
0 siblings, 0 replies; 17+ messages in thread
From: Phil Turmel @ 2016-11-14 16:14 UTC (permalink / raw)
To: Bruce Merry; +Cc: Wols Lists, linux-raid
On 11/14/2016 11:09 AM, Bruce Merry wrote:
> On 14 November 2016 at 18:01, Phil Turmel <philip@turmel.org> wrote:
>> In the meantime, run a "repair" scrub instead of a "check" scrub to
>> affirmatively force no mismatches. (Writes first member of mirrors to
>> the others.)
>
> I think that's good news. I wasn't sure which direction "repair" would
> copy, but the good drive is the first member ("Device Role : Active
> device 0"). I'll give that a go once the current scrub is done.
Direction is truly immaterial, as any bad sectors on device 0 will be
reconstructed from the other. Neil Brown has a blog entry on this topic
that applies here:
http://neil.brown.name/blog/20100211050355
Phil
^ permalink raw reply [flat|nested] 17+ messages in thread
* RE: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-14 15:58 ` Wols Lists
2016-11-14 16:03 ` Bruce Merry
@ 2016-11-15 18:14 ` Peter Sangas
2016-11-15 18:49 ` Anthony Youngman
1 sibling, 1 reply; 17+ messages in thread
From: Peter Sangas @ 2016-11-15 18:14 UTC (permalink / raw)
To: 'Wols Lists', 'Bruce Merry'; +Cc: linux-raid
Hi Wol,
-----Original Message-----
From: Wols Lists [mailto:antlists@youngman.org.uk]
Sent: Monday, November 14, 2016 7:58 AM
To: Bruce Merry
Cc: linux-raid@vger.kernel.org
Subject: Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
On 14/11/16 15:52, Bruce Merry wrote:
> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>> > Sounds like that drive could need replacing. I'd get a new drive
>> > and do that as soon as possible - use the --replace option of mdadm
>> > - don't fail the old drive and add the new.
> Would you mind explaining why I should use --replace instead of taking
> out the suspect drive? I guess I lose redundancy for any writes that
> occur while the rebuild is happening, but I'd plan to do this with the
> filesystem unmounted so there wouldn't be any writes.
>Because a replace will copy from the old drive to the new, recovering any failures from the rest of the array. A fail-and-add will have to rebuild the entire new array >from what's left of the old, stressing the old array much more.
>Okay, in your case, it probably won't make an awful lot of difference, but it does make you vulnerable to problems on the "good" drive. To alter your wording >slightly, you lose redundancy for writes AND READS that occur while the array is rebuilding. It's just good practice (and I point it out because --replace is new and >not well known at present).
>Cheers,
>Wol
With respect to the --replace switch and "replacing a failed drive" documented on the wiki here:
https://raid.wiki.kernel.org/index.php/Replacing_a_failed_drive Can you clear a few things up for me ?
1. If I just want to replace a working drive in a RAID1 and the array is still redundant I can
issue the following command as in your example:
mdadm /dev/mdN [--fail /dev/sdx1] --remove /dev/sdx1 --add /dev/sdy1
which fails and removes sdx1 and replaces it with sdy1.
Question1. How is this different from first doing a fail/remove on sdx1, physically replacing sdx1 with sdy1 and doing an add on sdy1?
2. If one of the drives as an error in a RAID1 and gets kicked out of the array and the array loses redundancy the wiki has the following example:
mdmad /dev/mdN --re-add /dev/sdX1
mdadm /dev/mdN --add /dev/sdY1 --replace /dev/sdX1 --with /dev/sdY1
Question2. Is this point here to first try and re-add sdX1 with the "--re-add" (first line above) and if that fails do a replace (second line above)?
Thanks,
Peter
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-15 18:14 ` Peter Sangas
@ 2016-11-15 18:49 ` Anthony Youngman
2016-11-15 19:22 ` Peter Sangas
0 siblings, 1 reply; 17+ messages in thread
From: Anthony Youngman @ 2016-11-15 18:49 UTC (permalink / raw)
To: Peter Sangas; +Cc: linux-raid
On 15/11/16 18:14, Peter Sangas wrote:
> Hi Wol,
>
>
> -----Original Message-----
> From: Wols Lists [mailto:antlists@youngman.org.uk]
> Sent: Monday, November 14, 2016 7:58 AM
> To: Bruce Merry
> Cc: linux-raid@vger.kernel.org
> Subject: Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
>
> On 14/11/16 15:52, Bruce Merry wrote:
>> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>>>> Sounds like that drive could need replacing. I'd get a new drive
>>>> and do that as soon as possible - use the --replace option of mdadm
>>>> - don't fail the old drive and add the new.
>> Would you mind explaining why I should use --replace instead of taking
>> out the suspect drive? I guess I lose redundancy for any writes that
>> occur while the rebuild is happening, but I'd plan to do this with the
>> filesystem unmounted so there wouldn't be any writes.
>
>> Because a replace will copy from the old drive to the new, recovering any failures from the rest of the array. A fail-and-add will have to rebuild the entire new array >from what's left of the old, stressing the old array much more.
>
>> Okay, in your case, it probably won't make an awful lot of difference, but it does make you vulnerable to problems on the "good" drive. To alter your wording >slightly, you lose redundancy for writes AND READS that occur while the array is rebuilding. It's just good practice (and I point it out because --replace is new and >not well known at present).
>
>> Cheers,
>> Wol
>
> With respect to the --replace switch and "replacing a failed drive" documented on the wiki here:
> https://raid.wiki.kernel.org/index.php/Replacing_a_failed_drive Can you clear a few things up for me ?
>
> 1. If I just want to replace a working drive in a RAID1 and the array is still redundant I can
> issue the following command as in your example:
>
> mdadm /dev/mdN [--fail /dev/sdx1] --remove /dev/sdx1 --add /dev/sdy1
>
> which fails and removes sdx1 and replaces it with sdy1.
>
> Question1. How is this different from first doing a fail/remove on sdx1, physically replacing sdx1 with sdy1 and doing an add on sdy1?
>
Not really different at all. It's just that you (obviously) can't do the
remove and add in the same command if you physically swap the drive in
the middle.
But I bang on a bit about having access to spare port to stick a drive
on, so I've assumed you can have both the old and the new drive
physically (and logically) in the system at the same time.
>
> 2. If one of the drives as an error in a RAID1 and gets kicked out of the array and the array loses redundancy the wiki has the following example:
>
> mdmad /dev/mdN --re-add /dev/sdX1
> mdadm /dev/mdN --add /dev/sdY1 --replace /dev/sdX1 --with /dev/sdY1
>
> Question2. Is this point here to first try and re-add sdX1 with the "--re-add" (first line above) and if that fails do a replace (second line above)?
>
Correct. You've lost redundancy, and (you NEED a bitmap here) the idea
is to get sdX1 back in to the array to restore redundancy before you
copy its contents to sdY1.
You need the bitmap because, without it, a re-add becomes a normal add,
and it's not only a waste of time, it adds stress to the array and
increases your chances of a total failure.
>
> Thanks,
> Peter
>
Cheers,
Wol
^ permalink raw reply [flat|nested] 17+ messages in thread
* RE: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
2016-11-15 18:49 ` Anthony Youngman
@ 2016-11-15 19:22 ` Peter Sangas
0 siblings, 0 replies; 17+ messages in thread
From: Peter Sangas @ 2016-11-15 19:22 UTC (permalink / raw)
To: 'Anthony Youngman'; +Cc: linux-raid
Got it. Thanks again.
-----Original Message-----
From: Anthony Youngman [mailto:antlists@youngman.org.uk]
Sent: Tuesday, November 15, 2016 10:50 AM
To: Peter Sangas
Cc: linux-raid@vger.kernel.org
Subject: Re: What to do about Offline_Uncorrectable and Pending_Sector in RAID1
On 15/11/16 18:14, Peter Sangas wrote:
> Hi Wol,
>
>
> -----Original Message-----
> From: Wols Lists [mailto:antlists@youngman.org.uk]
> Sent: Monday, November 14, 2016 7:58 AM
> To: Bruce Merry
> Cc: linux-raid@vger.kernel.org
> Subject: Re: What to do about Offline_Uncorrectable and Pending_Sector
> in RAID1
>
> On 14/11/16 15:52, Bruce Merry wrote:
>> On 13 November 2016 at 23:06, Wols Lists <antlists@youngman.org.uk> wrote:
>>>> Sounds like that drive could need replacing. I'd get a new drive
>>>> and do that as soon as possible - use the --replace option of mdadm
>>>> - don't fail the old drive and add the new.
>> Would you mind explaining why I should use --replace instead of
>> taking out the suspect drive? I guess I lose redundancy for any
>> writes that occur while the rebuild is happening, but I'd plan to do
>> this with the filesystem unmounted so there wouldn't be any writes.
>
>> Because a replace will copy from the old drive to the new, recovering any failures from the rest of the array. A fail-and-add will have to rebuild the entire new array >from what's left of the old, stressing the old array much more.
>
>> Okay, in your case, it probably won't make an awful lot of difference, but it does make you vulnerable to problems on the "good" drive. To alter your wording >slightly, you lose redundancy for writes AND READS that occur while the array is rebuilding. It's just good practice (and I point it out because --replace is new and >not well known at present).
>
>> Cheers,
>> Wol
>
> With respect to the --replace switch and "replacing a failed drive" documented on the wiki here:
> https://raid.wiki.kernel.org/index.php/Replacing_a_failed_drive Can you clear a few things up for me ?
>
> 1. If I just want to replace a working drive in a RAID1 and the array
> is still redundant I can issue the following command as in your example:
>
> mdadm /dev/mdN [--fail /dev/sdx1] --remove /dev/sdx1 --add /dev/sdy1
>
> which fails and removes sdx1 and replaces it with sdy1.
>
> Question1. How is this different from first doing a fail/remove on sdx1, physically replacing sdx1 with sdy1 and doing an add on sdy1?
>
Not really different at all. It's just that you (obviously) can't do the remove and add in the same command if you physically swap the drive in the middle.
But I bang on a bit about having access to spare port to stick a drive on, so I've assumed you can have both the old and the new drive physically (and logically) in the system at the same time.
>
> 2. If one of the drives as an error in a RAID1 and gets kicked out of the array and the array loses redundancy the wiki has the following example:
>
> mdmad /dev/mdN --re-add /dev/sdX1
> mdadm /dev/mdN --add /dev/sdY1 --replace /dev/sdX1 --with /dev/sdY1
>
> Question2. Is this point here to first try and re-add sdX1 with the "--re-add" (first line above) and if that fails do a replace (second line above)?
>
Correct. You've lost redundancy, and (you NEED a bitmap here) the idea is to get sdX1 back in to the array to restore redundancy before you copy its contents to sdY1.
You need the bitmap because, without it, a re-add becomes a normal add, and it's not only a waste of time, it adds stress to the array and increases your chances of a total failure.
>
> Thanks,
> Peter
>
Cheers,
Wol
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 17+ messages in thread
end of thread, other threads:[~2016-11-15 19:22 UTC | newest]
Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-11-13 18:46 What to do about Offline_Uncorrectable and Pending_Sector in RAID1 Bruce Merry
2016-11-13 20:18 ` Anthony Youngman
2016-11-13 20:51 ` Bruce Merry
2016-11-13 21:06 ` Wols Lists
2016-11-13 23:03 ` Phil Turmel
2016-11-14 6:50 ` Bruce Merry
2016-11-14 16:01 ` Phil Turmel
2016-11-14 16:09 ` Bruce Merry
2016-11-14 16:14 ` Phil Turmel
2016-11-14 15:52 ` Bruce Merry
2016-11-14 15:58 ` Wols Lists
2016-11-14 16:03 ` Bruce Merry
2016-11-14 16:09 ` Wols Lists
2016-11-14 16:10 ` Phil Turmel
2016-11-15 18:14 ` Peter Sangas
2016-11-15 18:49 ` Anthony Youngman
2016-11-15 19:22 ` Peter Sangas
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.