Intermittent loss of a disk, re-boot brings it back

[R.G.]

I ran into a funny one. My RAIDZ3 with seven 3TB disks has twice now had a disk go off line. Same disk each time. On reboot, the disk reappears and appears to be undamaged by both data content and SMART data error records each time. Well, actually, it did this once when I set the system up and I reseated all the cables, took the reappearance as verification of a cable problem. It did it again while actually on line. A reboot did not bring it back. I took this as permanent failure.

I ordered a replacement disk, and since I was to be out of town for some days, shut the system down. When I rebooted to make sure the disk was offlined from the pool, it had reappeared with no errors.

Any ideas?

 

[anodos]

Post hardware details and version of freenas.

 

[diedrichg]

And the SMART data of that drive. 90% chance it's a bad drive. Offline, replace, online, resilver.

 

[R.G.]

OK, the mystery unfolds a bit.

I was sitting in the same room as the NAS box, and we had a power glitch. I am out in the country, and glitches happen much more frequently than they do in the big city. That's why I have all my machines on UPS power.

I have most of the work cluster on a 1000VA backup, but the NAS has its own Tripp Lite SMART700USB. The UPS went into "charge only" mode on the glitch and did not pick up into battery backup, so the NAS went down. I messed with the UPS, unplugging, exercising the self tests and such, and came up with an NTF on it, got it back on line.

Then I rebooted the NAS. The subject disk was missing from the pool. I bet when I reboot, the disk is there again.

This is looking like an interaction between a sensitive disk, power line disturbances, and a flaky UPS.

 

[R.G.]

Yep, reboot put the drive back in again.

SMARTCTL data for the drive:

Code:

smartctl -a /dev/ada0
smartctl 6.3 2014-07-26 r3976 [FreeBSD 9.3-RELEASE-p5 amd64] (local build)
Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:  Toshiba 3.5" HDD DT01ACA...
Device Model:  TOSHIBA DT01ACA300
Serial Number:  84G9V94GS
LU WWN Device Id: 5 000039 ff4e0a90e
Firmware Version: MX6OABB0
User Capacity:  3,000,592,982,016 bytes [3.00 TB]
Sector Sizes:  512 bytes logical, 4096 bytes physical
Rotation Rate:  7200 rpm
Form Factor:  3.5 inches
Device is:  In smartctl database [for details use: -P show]
ATA Version is:  ATA8-ACS T13/1699-D revision 4
SATA Version is:  SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:  Mon Feb  9 10:13:54 2015 CST
SMART support is: Available - device has SMART capability.
SMART support is: 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:  (22078) seconds.
Offline data collection
capabilities:  (0x5b) SMART execute Offline immediate.
  Auto Offline data collection on/off support.
  Suspend Offline collection upon new
  command.
  Offline surface scan supported.
  Self-test supported.
  No 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:  (  1) minutes.
Extended self-test routine
recommended polling time:  ( 368) minutes.
SCT capabilities:  (0x003d) SCT Status supported.
  SCT Error Recovery Control 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  FLAG  VALUE WORST THRESH TYPE  UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate  0x000b  100  100  016  Pre-fail  Always  -  0
  2 Throughput_Performance  0x0005  140  140  054  Pre-fail  Offline  -  69
  3 Spin_Up_Time  0x0007  157  157  024  Pre-fail  Always  -  305 (Average 426)
  4 Start_Stop_Count  0x0012  100  100  000  Old_age  Always  -  62
  5 Reallocated_Sector_Ct  0x0033  100  100  005  Pre-fail  Always  -  0
  7 Seek_Error_Rate  0x000b  100  100  067  Pre-fail  Always  -  0
  8 Seek_Time_Performance  0x0005  119  119  020  Pre-fail  Offline  -  35
  9 Power_On_Hours  0x0012  100  100  000  Old_age  Always  -  1209
10 Spin_Retry_Count  0x0013  100  100  060  Pre-fail  Always  -  0
12 Power_Cycle_Count  0x0032  100  100  000  Old_age  Always  -  62
192 Power-Off_Retract_Count 0x0032  100  100  000  Old_age  Always  -  64
193 Load_Cycle_Count  0x0012  100  100  000  Old_age  Always  -  64
194 Temperature_Celsius  0x0002  181  181  000  Old_age  Always  -  33 (Min/Max 18/38)
196 Reallocated_Event_Count 0x0032  100  100  000  Old_age  Always  -  0
197 Current_Pending_Sector  0x0022  100  100  000  Old_age  Always  -  0
198 Offline_Uncorrectable  0x0008  100  100  000  Old_age  Offline  -  0
199 UDMA_CRC_Error_Count  0x000a  200  200  000  Old_age  Always  -  0

SMART Error Log Version: 1
No Errors Logged

SMART Self-test log structure revision number 1
Num  Test_Description  Status  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Short offline  Completed without error  00%  13  -

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.

I can't figure out how to remove formatting on the section here. Sorry.
I don't see errors in the SMART data that relate to this issue.

The hardware is Supermicro MBD-X9SCL-F-O, Xeon E3-1220, 16GB ECC, six other 3TB disks from various manufacturers.

 

[R.G.]

OK, latest oddity. I had a drive fail. It was *not* the reboot-me drive, it was another one in the array entirely. For this one, FreeNAS reported that the drive was faulted due to write errors.

The smartctl reports had not been listing any errors in any of the drives, so I went off and did short and long tests on the drive. No trouble found by smart tests.

I swapped in a replacement drive and resilvered, and that seems to be working just fine, but now I have this 3TB drive that shows no errors and works fine to any other test I can give it in other machines.

My experience leads me directly to either bad power or data cables, bad connector, or ...ugh... a problem on the motherboard or power supply. Given that I suspected power supply issues before, I went back through the whole power supply design and came up with a figure for it needing an ATX power supply of only 234W. What's in there is a Corsair 400W, about two years old. Going to a bigger power supply seems outright wasteful. I'll go put the Kill-A-Watt on it and get the actual AC power used, but last time I checked a similar configuration, it was under 50W at idle.

I wobbled/reseated all the power and data cables, and it's acting like all is well - which is how it acted before.

So do I have a drive that FreeNAS doesn't like that is otherwise OK, or do I have a drive that has subtle, hidden errors that don't show up on any tests I can give it.

Or do I have an intermittent port on my fancy new server-grade Supermicro board that also doesn't leave a trail?

 

[Fraoch]

Going to a bigger power supply isn't wasteful if the power supply is efficient enough. All the power ups and downs (often without proper shutdown) will be hard on all components. It's impossible to say what it is right now though.

One thing you should do if the drives are acting flaky is put in a SMART testing schedule - for example, short tests daily and long tests weekly or so. Your drive has 1209 hours on it but only shows one test when it had been in use for 13 hours. You might also want to try a one-time conveyance test just to be sure.