ZFS Dying disk or...?

K

Kuzbad

Woke up this morning to a mostly non-responsive server and a console filled up with messages like this (messages log for just the last 8 hours is 126M):

Code: 
Mar  8 07:36:17 xeon kernel: vm_fault: pager read error, pid 1 (init)
Mar  8 07:36:17 xeon kernel: swap_pager: I/O error - pagein failed; blkno 2097173,size 4096, error 6

Including all kinds of funny garbled variants:

Code: 
Mar  8 07:36:17 xeon kernel: swap_pager: I/O error - pagein [B]friled[/B]; blkno 2097173,size 4096, error

This server has a 4-disk ZFS pool, with each disk having swap on it. One of the disks just disappeared earlier:

Code: 
Mar  8 00:44:10 xeon kernel: (ada2:ahcich2:0:0:0): lost device
Mar  8 00:44:10 xeon kernel: pid 635 (devd), uid 0: exited on signal 11
...
Mar  8 00:57:18 xeon kernel: xptioctl: pass driver is not in the kernel
Mar  8 00:57:18 xeon kernel: xptioctl: put "device pass" in your kernel config file

Not sure what the xptioctl error message is about. There are no other error messages other than random programs dying.

I could access the system through Samba just fine, but I could not SSH in, and when I tried to access the console the entire system froze. Hard off and back on.

The system came up again totally fine. Nothing odd in the dmesg. zpool status and smartctl output below.

I can't figure out what happened--any ideas? Do I need to replace this disk ASAP, or is this possibly a motherboard fault?

Code: 
# zpool status -x
  pool: zroot
 state: ONLINE
status: One or more devices has experienced an unrecoverable error.  An
    attempt was made to correct the error.  Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
    using 'zpool clear' or replace the device with 'zpool replace'.
   see: http://illumos.org/msg/ZFS-8000-9P
  scan: resilvered 60.2M in 5h1m with 0 errors on Wed Mar  8 07:39:45 2017
config:

    NAME           STATE     READ WRITE CKSUM
    zroot          ONLINE       0     0     0
      mirror-0     ONLINE       0     0     0
        gpt/disk0  ONLINE       0     0     0
        gpt/disk1  ONLINE       0     0     0
      mirror-1     ONLINE       0     0     0
        gpt/disk3  ONLINE       0     0     0
        ada2p3     ONLINE       0     0     2
    cache
      gpt/cache    ONLINE       0     0     0

errors: No known data errors



Here's the output of smartctl for ada2:

Code: 
# smartctl -a -q noserial /dev/ada2

smartctl 6.3 2014-07-26 r3976 [FreeBSD 8.4-RELEASE-p27 amd64] (local build)
Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:     Western Digital RE4-GP
Device Model:     WDC WD2002FYPS-02W3B1
Firmware Version: 04.01G02
User Capacity:    2,000,398,934,016 bytes [2.00 TB]
Sector Size:      512 bytes logical/physical
Rotation Rate:    5400 rpm
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, 3.0 Gb/s (current: 3.0 Gb/s)
Local Time is:    Wed Mar  8 08:14:37 2017 EST
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:         (43560) 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:    ( 419) minutes.
Conveyance self-test routine
recommended polling time:    (   5) minutes.
SCT capabilities:          (0x303f)    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     0x002f   200   200   051    Pre-fail  Always       -       0
  3 Spin_Up_Time            0x0027   253   253   021    Pre-fail  Always       -       7250
  4 Start_Stop_Count        0x0032   100   100   000    Old_age   Always       -       9
  5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x002e   200   200   000    Old_age   Always       -       0
  9 Power_On_Hours          0x0032   097   097   000    Old_age   Always       -       2825
 10 Spin_Retry_Count        0x0032   100   253   000    Old_age   Always       -       0
 11 Calibration_Retry_Count 0x0032   100   253   000    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       8
192 Power-Off_Retract_Count 0x0032   200   200   000    Old_age   Always       -       7
193 Load_Cycle_Count        0x0032   134   134   000    Old_age   Always       -       199994
194 Temperature_Celsius     0x0022   110   105   000    Old_age   Always       -       42
196 Reallocated_Event_Count 0x0032   200   200   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0032   200   200   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0030   100   253   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -       0
200 Multi_Zone_Error_Rate   0x0008   100   253   000    Old_age   Offline      -       0

SMART Error Log Version: 1
No Errors Logged

SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]

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.
 
First do zpool scrub to check everything. If no errors then backup everything and do the FreeBSD upgrade. It's a security risk to use outdated FreeBSD versions. FreeBSD 11.0 is stable for a production server.
 
Hi SirDice,

That's the plan. I've got a FreeBSD 11 VM in testing currently. This is an internal server, I'm the only one with console access, and all services are firewall inside and out. I'm not worried about security in the OS itself.

Does the version of FreeBSD actually change anything about this error message, however? I'm uncertain why a disk would just disconnect without any other obvious errors?
 
Remington said:
First do zpool scrub to check everything. If no errors then backup everything and do the FreeBSD upgrade. It's a security risk to use outdated FreeBSD versions. FreeBSD 11.0 is stable for a production server.
Click to expand...

I'm currently doing a full ZFS send to make sure my backup is as up-to-date as possible. As soon as that's done, then I'm going to do the scrub.
 
Kuzbad said:
I'm uncertain why a disk would just disconnect without any other obvious errors?
Click to expand...
I've had this happen with an old Promise SATA controller card. Turns out the controller was near death and it didn't take too long for it to fail completely. Is this an onboard controller or a card? Do any of the other disks disappear? Or is it always the same disk?

Also, use smartctl(8) to run a short, then a long test. Hopefully that catches whatever is wrong with the disk (if it's the disk itself that's causing it).
 
SirDice said:
I've had this happen with an old Promise SATA controller card. Turns out the controller was near death and it didn't take too long for it to fail completely. Is this an onboard controller or a card? Do any of the other disks disappear? Or is it always the same disk?
Click to expand...

This is an onboard controller and I've never had this happen before. I think, at the very least, I will move my upgrade plans forward a bit!! Thanks.
 
Disable the power-management on green power disks. It's causing the drives to spin down hundreds or thousands of times a day and completely wears out the mechanics. Your spinup times are already quite high, so prepare for a failing drive in the near future...
ZFS also isn't particularly keen with drives that seem to not respond because they first have to spin up and get thrown out of the vdev.

The powersaving can be disabled by setting APM to 254 with camcontrol. I also disabled the acoustics management (aam) on the GP drives I had, as it can increase latency drastically.
Even with APM and AAM disabled from the first day I had to RMA 3 out of 5 RE4-GPs within, so as said: be prepared for it to fail any time!
 
sko said:
Your spinup times are already quite high, so prepare for a failing drive in the near future...
Click to expand...
You are wrong. Spin-up time is normal.
What is not normal is
Code: 
193 Load_Cycle_Count        0x0032   134   134   000    Old_age   Always       -       199994
It is close to the limit (300 000) I believe, and in 3 000 hours? Amazing.
Are all disks have this parameter so high?
Re series shouldn't do it.
Code: 
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x002f   200   200   051    Pre-fail  Always       -       0
  3 Spin_Up_Time            0x0027   149   147   021    Pre-fail  Always       -       11516
  4 Start_Stop_Count        0x0032   100   100   000    Old_age   Always       -       75
  5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x002e   200   200   000    Old_age   Always       -       0
  9 Power_On_Hours          0x0032   087   087   000    Old_age   Always       -       10211
 10 Spin_Retry_Count        0x0032   100   253   000    Old_age   Always       -       0
 11 Calibration_Retry_Count 0x0032   100   253   000    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       75
 16 Unknown_Attribute       0x0022   024   176   000    Old_age   Always       -       428506950898
183 Runtime_Bad_Block       0x0032   100   100   000    Old_age   Always       -       0
192 Power-Off_Retract_Count 0x0032   200   200   000    Old_age   Always       -       73
193 Load_Cycle_Count        0x0032   200   200   000    Old_age   Always       -       2
194 Temperature_Celsius     0x0022   116   105   000    Old_age   Always       -       36
196 Reallocated_Event_Count 0x0032   200   200   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0032   200   200   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0030   200   200   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -       0
200 Multi_Zone_Error_Rate   0x0008   200   200   000    Old_age   Offline      -       0
 
sorry, of course i meant the load cycle count with nearly 200k. Spinup time is normal (253 / 21)

Unfortunately the RE-GP also spin down after very short idle periods. IIRC its a bit longer than on desktop drives, but still causes massive mechanical wear.
The drives mostly die when power-cycling and just won't spin up again but making only clicking noises.
 
sko said:
sorry, of course i meant the load cycle count with nearly 200k. Spinup time is normal (253 / 21)

Unfortunately the RE-GP also spin down after very short idle periods. IIRC its a bit longer than on desktop drives, but still causes massive mechanical wear.
The drives mostly die when power-cycling and just won't spin up again but making only clicking noises.
Click to expand...
I see. I've made a mistake. Mine are Western Digital Se and according SMART above they are not parking at all (I like this behaviour more). WD should be assamed for doing this on drives thay are positioning as Enterprise.
 
In the past 10 years the manufactures stop using the gold plated contacts between the hard disk controller and the motor of the hard disk and the actuator pcb inside the disk. Most of this cause the copper contacts to oxides that lead to poor contact and sudden disconnects between the sata controller and the hard disk. You can give a try to remove the hard disk controller with torx screwdriver and check and clean with pencil gum those contacts then use some alcohol to wipe the contacts. This method resolve me a lot of similar problems that i used to have in the years.

You always must have a backup of your information before you do this.
 
VladiBG said:
In the past 10 years the manufactures stop using the gold plated contacts between the hard disk controller and the motor of the hard disk and the actuator pcb inside the disk. Most of this cause the copper contacts to oxides that lead to poor contact and sudden disconnects between the sata controller and the hard disk. You can give a try to remove the hard disk controller with torx screwdriver and check and clean with pencil gum those contacts then use some alcohol to wipe the contacts. This method resolve me a lot of similar problems that i used to have in the years.
Click to expand...

Probably best to do this in a clean room or in a small room with ionized air filter to remove dust and use latex gloves.
 
Remington said:
Probably best to do this in a clean room or in a small room with ionized air filter to remove dust and use latex gloves.
Click to expand...
He's not talking about opening up the drive but removing the controller board at the bottom and checking the 'pads' that connect to the inside of the disk via springs/lashes.
 
You must log in or register to reply here.
Back
Top