Real use case of 3 x Odroid HC2 that have been in production for 5+ years

Post Reply
User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Real use case of 3 x Odroid HC2 that have been in production for 5+ years

Post by domih »

The 3 systems run with a 10TB WD Red (WD100EFAX) and are named hc2a, hc2b and hc2c.

So far the NAS storage is used at 34%.

Code: Select all

Filesystem      Size  Used Avail Use% Mounted on
/dev/sda1       9.1T  2.9T  5.8T  34% /mnt/storage
The 3 systems maintain an identical NAS image using rsync. No distributed storage.

hc2a is the "active" NAS.
hc2b syncs from hc2a every "active" hour, from early morning to late evening.
hc2c is manually synced on demand every few days or once a week.

As of this writing the 3 x HC2 have been working with zero issue. Kudos to Hard Kernel engineers!

I manually apt update them every few months. They are not exposed to the Internet. I usually upgrade hc2c and if it reboots normally, I then upgrade hc2b and if it reboots normally, I finally upgrade hc2a. In doing so, I always has at least one NAS running if a problem were to occur.

Because the Power-Off_Retract_Count does not rise at the same rate on each drive, I do not expect the disks to die at the same time.

The systems have now been running for about 3.2 years :-)

Statistics for each drive:

hc2a

Code: Select all

domih@hc2a:~$ sudo smartctl -d sat -a /dev/sda
...
Device Model:     WDC WD100EFAX-68LHPN0
...
Rotation Rate:    5400 rpm
Form Factor:      3.5 inches
...
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
...
  9 Power_On_Hours          0x0012   096   096   000    Old_age   Always       -       28067
...
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       38
...
192 Power-Off_Retract_Count 0x0032   086   086   000    Old_age   Always       -       17795
193 Load_Cycle_Count        0x0012   086   086   000    Old_age   Always       -       17795
194 Temperature_Celsius     0x0002   180   180   000    Old_age   Always       -       36 (Min/Max 20/49)
...
SMART Error Log Version: 1
No Errors Logged
hc2b

Code: Select all

domih@hc2b:~$ sudo smartctl -d sat -a /dev/sda
...
Device Model:     WDC WD100EFAX-68LHPN0
...
Rotation Rate:    5400 rpm
Form Factor:      3.5 inches
...
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
...
  9 Power_On_Hours          0x0012   097   097   000    Old_age   Always       -       27969
...
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       41
...
192 Power-Off_Retract_Count 0x0032   090   090   000    Old_age   Always       -       12994
193 Load_Cycle_Count        0x0012   090   090   000    Old_age   Always       -       12994
194 Temperature_Celsius     0x0002   185   185   000    Old_age   Always       -       35 (Min/Max 21/47)


SMART Error Log Version: 1
ATA Error Count: 2
...
Error 2 occurred at disk power-on lifetime: 15776 hours (657 days + 8 hours)
  When the command that caused the error occurred, the device was active or idle.
...
Error 1 occurred at disk power-on lifetime: 15776 hours (657 days + 8 hours)
  When the command that caused the error occurred, the device was active or idle.
...
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%     27969         -
# 2  Short offline       Completed without error       00%      2902         -
# 3  Short offline       Completed without error       00%      1875         -
hc2c

Code: Select all

domih@hc2c:~$ sudo smartctl -d sat -a /dev/sda
...
Device Model:     WDC WD100EFAX-68LHPN0
...
Rotation Rate:    5400 rpm
Form Factor:      3.5 inches
...
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
...
  9 Power_On_Hours          0x0012   097   097   000    Old_age   Always       -       27898
...
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       41
...
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       712
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       712
194 Temperature_Celsius     0x0002   180   180   000    Old_age   Always       -       36 (Min/Max 21/48)
...
SMART Error Log Version: 1
No Errors Logged
Last edited by domih on Wed Mar 13, 2024 3:26 am, edited 5 times in total.
These users thanked the author domih for the post (total 2):
odroid (Sat May 07, 2022 10:43 am) • DHGE (Mon May 23, 2022 4:39 pm)

psomero
Posts: 34
Joined: Wed Feb 19, 2014 11:46 am
languages_spoken: english
ODROIDs: XU+E, XU4, C1+, HC1, H2
Location: Silicon Valley
Has thanked: 0
Been thanked: 2 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ years

Post by psomero »

I've had a single HC1 that has been up and running since a month or two after they were available. The only issue I've had is I kicked it off of the top of my desktop computer and it crashed the WD red drive that was in it. I replaced the spinny one with a samsung 960 pro and it's had 100% uptime since then.

A+
These users thanked the author psomero for the post:
domih (Sat Aug 06, 2022 10:26 am)

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ years

Post by domih »

My 3-node NAS (see description in OP) has now reached 35542 hours / 24 / 365 = ~4.06 years of production exploitation :D

These ODROID HC2 are going, going and going!

Code: Select all


domih@hc2a:~$ df -h -t ext4
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p2   59G  3.7G   55G   7% /
/dev/sda1       9.1T  3.3T  5.4T  38% /mnt/storage

domih@hc2a:~$ sudo smartctl -a -d sat /dev/sda
smartctl 6.6 2016-05-31 r4324 [armv7l-linux-4.14.180-178] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Device Model:     WDC WD100EFAX-68LHPN0
.../...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
.../...
Rotation Rate:    5400 rpm
.../...
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
.../...
  9 Power_On_Hours          0x0012   095   095   000    Old_age   Always       -       35542
.../...
192 Power-Off_Retract_Count 0x0032   080   080   000    Old_age   Always       -       24186
193 Load_Cycle_Count        0x0012   080   080   000    Old_age   Always       -       24186
.../...

domih@hc2a:~$ sudo dumpe2fs -h /dev/mmcblk1p2 | grep Lifetime
dumpe2fs 1.44.1 (24-Mar-2018)
Lifetime writes:          1502 GB (Estimated max: 56TB)

===============================================================================

domih@hc2b:~$ df -h -t ext4
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p2   59G  2.7G   56G   5% /
/dev/sda1       9.1T  3.3T  5.4T  38% /mnt/storage

domih@hc2b:~$ sudo smartctl -a -d sat /dev/sda
smartctl 6.6 2016-05-31 r4324 [armv7l-linux-4.14.180-178] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Device Model:     WDC WD100EFAX-68LHPN0
.../...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
.../...
Rotation Rate:    5400 rpm
.../...
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
.../...
  9 Power_On_Hours          0x0012   095   095   000    Old_age   Always       -       35437
.../...
192 Power-Off_Retract_Count 0x0032   086   086   000    Old_age   Always       -       17598
193 Load_Cycle_Count        0x0012   086   086   000    Old_age   Always       -       17598
.../...

Note: same 2 errors as last time at 15776 hours (657 days + 8 hours). No more errors since.

domih@hc2b:~$ sudo dumpe2fs -h /dev/mmcblk1p2 | grep Lifetime
dumpe2fs 1.44.1 (24-Mar-2018)
Lifetime writes:          1709 GB (Estimated max: 56TB)

===============================================================================

domih@hc2c:~$ df -h -t ext4
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p2   59G  2.6G   57G   5% /
/dev/sda1       9.1T  3.3T  5.4T  38% /media/storage


domih@hc2c:~$ sudo smartctl -a -d sat /dev/sda
smartctl 6.6 2016-05-31 r4324 [armv7l-linux-4.14.180-178] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Device Model:     WDC WD100EFAX-68LHPN0
.../...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
.../...
Rotation Rate:    5400 rpm
.../...
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
.../...
  9 Power_On_Hours          0x0012   095   095   000    Old_age   Always       -       35102
.../...
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       928
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       928
.../...

domih@hc2c:~$ sudo dumpe2fs -h /dev/mmcblk1p2 | grep Lifetime
dumpe2fs 1.44.1 (24-Mar-2018)
Lifetime writes:          1877 GB  (Estimated max: 56TB)
Note: They bravely survived a massive power outage of 20+ hours (storms in California this winter) with the UPS miserably failing right away (dead UPS battery which I since replaced) so I did not have time to shut them down properly.
These users thanked the author domih for the post:
odroid (Tue Mar 14, 2023 10:08 am)

User avatar
rooted
Posts: 10622
Joined: Fri Dec 19, 2014 9:12 am
languages_spoken: english
Location: Gulf of Mexico, US
Has thanked: 822 times
Been thanked: 736 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ ==> 4+ years

Post by rooted »

You must have your important files stored on these? Or is it just long term testing for posterity?

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ ==> 4+ years

Post by domih »

rooted wrote:
Wed Mar 15, 2023 6:29 pm
You must have your important files stored on these? Or is it just long term testing for posterity?
My life and my work.

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ ==> 4+ years

Post by domih »

Another storm in the SF Bay Area. Another power outage. I was at the office so the UPSes at home slowly got fully drained. I hope the Duracell batteries are OK with a full cycle. Anyway, no power for quite some time. Apparently the storm destroyed a piece of equipment (probably a transformer) that takes time to replace.

rajprakash
Posts: 13
Joined: Mon Dec 22, 2014 9:42 pm
languages_spoken: english
ODROIDs: ODROID-C1,-XU4
Has thanked: 0
Been thanked: 1 time
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ ==> 4+ years

Post by rajprakash »

Glad to see another bay area user of ODROIDs! Greetings neighbor, i'm over in the east bay.
These users thanked the author rajprakash for the post:
domih (Sun May 07, 2023 7:23 am)

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 3+ ==> 4+ years

Post by domih »

rajprakash wrote:
Sun May 07, 2023 2:56 am
Glad to see another bay area user of ODROIDs! Greetings neighbor, i'm over in the east bay.
Hello there! Hopefully there are more than two ODROID users in the SF Bay Area :mrgreen:

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by domih »

Still running with no issue :D

Code: Select all

domih@hc2a:~$ sudo smartctl -d sat -a /dev/sda
...
Device Model:     WDC WD100EFAX-68LHPN0
...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
...
Rotation Rate:    5400 rpm
...
SMART support is: Enabled
...
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  0x0004   130   130   054    Old_age   Offline      -       108
  3 Spin_Up_Time            0x0007   152   152   024    Pre-fail  Always       -       439 (Average 425)
  4 Start_Stop_Count        0x0012   093   093   000    Old_age   Always       -       28344
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   095   095   000    Old_age   Always       -       40830
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       58
 22 Unknown_Attribute       0x0023   100   100   025    Pre-fail  Always       -       100
192 Power-Off_Retract_Count 0x0032   077   077   000    Old_age   Always       -       28348
193 Load_Cycle_Count        0x0012   077   077   000    Old_age   Always       -       28348
194 Temperature_Celsius     0x0002   191   191   000    Old_age   Always       -       34 (Min/Max 18/49)
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
...
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%     40830         -
# 2  Short offline       Completed without error       00%     24358         -
# 3  Short offline       Completed without error       00%     18573         -
# 4  Short offline       Completed without error       00%      3000         -
# 5  Short offline       Completed without error       00%      1850         -
...


domih@hc2b:~$ sudo smartctl -d sat -a /dev/sda
...
Device Model:     WDC WD100EFAX-68LHPN0
...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
...
Rotation Rate:    5400 rpm
...
SMART support is: Enabled
...
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  0x0004   131   131   054    Old_age   Offline      -       104
  3 Spin_Up_Time            0x0007   146   146   024    Pre-fail  Always       -       449 (Average 449)
  4 Start_Stop_Count        0x0012   095   095   000    Old_age   Always       -       20755
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   095   095   000    Old_age   Always       -       40720
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       64
 22 Unknown_Attribute       0x0023   100   100   025    Pre-fail  Always       -       100
192 Power-Off_Retract_Count 0x0032   083   083   000    Old_age   Always       -       20760
193 Load_Cycle_Count        0x0012   083   083   000    Old_age   Always       -       20760
194 Temperature_Celsius     0x0002   185   185   000    Old_age   Always       -       35 (Min/Max 19/50)
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       -       2
...
Error 2 occurred at disk power-on lifetime: 15776 hours (657 days + 8 hours)
Error 1 occurred at disk power-on lifetime: 15776 hours (657 days + 8 hours)
...
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%     40720         -
# 2  Short offline       Completed without error       00%     27969         -
# 3  Short offline       Completed without error       00%      2902         -
# 4  Short offline       Completed without error       00%      1875         -
...


domih@hc2c:~$ sudo smartctl -d sat -a /dev/sda
...
Device Model:     WDC WD100EFAX-68LHPN0
...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
...
Rotation Rate:    5400 rpm
...
SMART support is: Enabled
...
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  0x0004   129   129   054    Old_age   Offline      -       112
  3 Spin_Up_Time            0x0007   147   147   024    Pre-fail  Always       -       446 (Average 448)
  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       1047
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   095   095   000    Old_age   Always       -       40330
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       64
 22 Unknown_Attribute       0x0023   100   100   025    Pre-fail  Always       -       100
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       1051
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       1051
194 Temperature_Celsius     0x0002   185   185   000    Old_age   Always       -       35 (Min/Max 18/48)
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 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%     40330         -
# 2  Short offline       Completed without error       00%      3169         -
# 3  Short offline       Completed without error       00%      1997         -
...
Usage is slowly climbing (was 34% back in May 04, 2022):

Code: Select all

domih@hc2a:~$ df -h -t ext4
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p2   59G  3.7G   55G   7% /
/dev/sda1       9.1T  3.6T  5.1T  42% /mnt/storage
These users thanked the author domih for the post:
L67GS (Mon Oct 30, 2023 7:09 am)

User avatar
rooted
Posts: 10622
Joined: Fri Dec 19, 2014 9:12 am
languages_spoken: english
Location: Gulf of Mexico, US
Has thanked: 822 times
Been thanked: 736 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by rooted »

Nice there's nothing more to report but increased disk usage
These users thanked the author rooted for the post:
domih (Mon Oct 23, 2023 7:25 am)

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by domih »

rooted wrote:
Mon Oct 23, 2023 7:18 am
Nice there's nothing more to report but increased disk usage Image
Yes the point, IMHO, is to report that our "gadgets" as the main stream PC users see them, can be used for real production work at home.

I'm not even talking about the main stream PC users ignorance about the massive industrial usage of SBCs (signage, medical, factories, etc).

In my humble case, the equation is simple, I don't mind running 3 HC2 24x365 because the power consumption is low enough to not explode the energy bills. Especially also because the WD Red 10TB go to "sleep" pretty quickly.

I would NOT run 3 full blown X86 PCs for having rsync-based replicated NAS servers.

fvolk
Posts: 908
Joined: Sun Jun 05, 2016 11:04 pm
languages_spoken: english
ODROIDs: C4, H3, M1S
Has thanked: 0
Been thanked: 144 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by fvolk »

domih wrote:
Mon Oct 23, 2023 7:33 am
I don't mind running 3 HC2 24x365 because the power consumption is low enough to not explode the energy bills.
....
I would NOT run 3 full blown X86 PCs for having rsync-based replicated NAS servers.
x86-based solutions can also be quite impressive as NAS servers... viewtopic.php?p=375518#p375518

Initially I had two HC1 running as home server(s).
Unfortunately, I had to limit CPU speed, otherwise once in a while they locked up. (not heat, but load changes across BIG-little cores was buggy?)
And they were only 32 bit and they missed CPU-accelerated encyption, which limited I/O performance too much.

(H2/H3 as replacement still have software incompatibilites (missing AVX), but stability, power use and encryption just work....)
These users thanked the author fvolk for the post:
domih (Mon Oct 30, 2023 5:03 am)

fvolk
Posts: 908
Joined: Sun Jun 05, 2016 11:04 pm
languages_spoken: english
ODROIDs: C4, H3, M1S
Has thanked: 0
Been thanked: 144 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by fvolk »

domih wrote:
Thu May 05, 2022 6:15 am
hc2a is the "active" NAS.
hc2b syncs from hc2a every "active" hour, from early morning to late evening.
hc2c is manually synced on demand every few days or once a week.

As of this writing the 3 x HC2 have been working with zero issue.
I'm curious, do you detect random bit-flips or spurious bad sectors?

From my anecdotal experience: I keep important data on 3 separate copies/disks,
and once in a while I do a bit-for-bit compare of the files - and once in a while this shows either a bad byte or a chunk of 000000 in the middle of a file.
Then I overwrite the broken copy with one of the others....
These users thanked the author fvolk for the post:
domih (Mon Oct 30, 2023 5:04 am)

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by domih »

fvolk wrote:
Sun Oct 29, 2023 6:27 pm
domih wrote:
Thu May 05, 2022 6:15 am
hc2a is the "active" NAS.
hc2b syncs from hc2a every "active" hour, from early morning to late evening.
hc2c is manually synced on demand every few days or once a week.

As of this writing the 3 x HC2 have been working with zero issue.
I'm curious, do you detect random bit-flips or spurious bad sectors?
No, I don't. I should :mrgreen:

User avatar
rooted
Posts: 10622
Joined: Fri Dec 19, 2014 9:12 am
languages_spoken: english
Location: Gulf of Mexico, US
Has thanked: 822 times
Been thanked: 736 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by rooted »

I expect if any cosmic radiation hits my drives they deserve it

fvolk
Posts: 908
Joined: Sun Jun 05, 2016 11:04 pm
languages_spoken: english
ODROIDs: C4, H3, M1S
Has thanked: 0
Been thanked: 144 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by fvolk »

domih wrote:
Mon Oct 30, 2023 5:05 am
fvolk wrote:
Sun Oct 29, 2023 6:27 pm
I'm curious, do you detect random bit-flips or spurious bad sectors?
No, I don't. I should
So... not important data then? ....

I also run a weekly Borgbackup:
An external USB disk is attached to a C4, and every machine in LAN runs once a week a full disk Borgbackup job (incremental diff+encrypted backup) to this.
Then the disk is unmounted, detached, and safely stored away.
Once a year I prune the weekly runs to monthly backups, so 12 snapshots per year are preserved, and start a new archive for the new year.

So far I've been able to restore almost all my mistakes... with very little storage use and everything scripted.

...C4+H3 Odroids are the reliable workers in my LAN! :-)

(HC1 and C2 were not reliable enough)

User avatar
rooted
Posts: 10622
Joined: Fri Dec 19, 2014 9:12 am
languages_spoken: english
Location: Gulf of Mexico, US
Has thanked: 822 times
Been thanked: 736 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 4 years and 6 months

Post by rooted »

Just a bunch of movies and shows mostly on my NAS, all my important data is online currently until I get new NAS drives. One is already on the way out (over 62K hours) and the other has more than 48K hours.

One of these days I will build a proper NAS with good redundancy, until then I trust my important data won't be lost in the cloud.

Note the errors on the drive going bad:

Image

User avatar
domih
Posts: 766
Joined: Mon Feb 11, 2019 4:48 pm
languages_spoken: English, French
ODROIDs: UX4, HC2, N2, N2+, H2, H2+, C4, HC4, M1, M1S (with UPS, 4 x Relay) H3, H3+ - 1GbE, 2.5GbE, 10GbE, 45+ GbE
Location: San Francisco Bay Area
Has thanked: 303 times
Been thanked: 229 times
Contact:

Re: Real use case of 3 x Odroid HC2 that have been in production for 5+ years

Post by domih »

My 3 NAS systems just passed 5 years of exploitation :D

Here is the SMART data from the master one:

Code: Select all

domih@hc2a:~$ domih@hc2a:~$ sudo smartctl -d sat -t short /dev/sda

...

domih@hc2a:~$ domih@hc2a:~$ sudo smartctl -d sat -a /dev/sda

smartctl 6.6 2016-05-31 r4324 [armv7l-linux-4.14.180-178] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Device Model:     WDC WD100EFAX-68LHPN0
...
User Capacity:    10,000,831,348,736 bytes [10.0 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Rotation Rate:    5400 rpm
Form Factor:      3.5 inches
...
ATA Version is:   ACS-2, ATA8-ACS T13/1699-D revision 4
SATA Version is:  SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Tue Mar 12 18:16:20 2024 UTC
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


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  0x0004   130   130   054    Old_age   Offline      -       108
  3 Spin_Up_Time            0x0007   150   150   024    Pre-fail  Always       -       440 (Average 439)
  4 Start_Stop_Count        0x0012   093   093   000    Old_age   Always       -       31135
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   094   094   000    Old_age   Always       -       44223
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       59
 22 Unknown_Attribute       0x0023   100   100   025    Pre-fail  Always       -       100
192 Power-Off_Retract_Count 0x0032   075   075   000    Old_age   Always       -       31139
193 Load_Cycle_Count        0x0012   075   075   000    Old_age   Always       -       31139
194 Temperature_Celsius     0x0002   171   171   000    Old_age   Always       -       38 (Min/Max 18/49)
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%     44223         -
# 2  Short offline       Completed without error       00%     40830         -
# 3  Short offline       Completed without error       00%     24358         -
# 4  Short offline       Completed without error       00%     18573         -
# 5  Short offline       Completed without error       00%      3000         -
# 6  Short offline       Completed without error       00%      1850         -
Power_On_Hours is 44223 / 24 = 1842.625 days = 5.05 years :) :) :)

An rsync of 14+ GB from the master to an AMD 4750G system over 1 GbE took 6m12.914s.
The same rsync from the AMD 4750G system to an AMD V1605 system over IPoIB (~20GbE) took 2m15.749s.

Code: Select all

sent 20.84K bytes  received 14.02G bytes  37.53M bytes/sec
total size is 4.32T  speedup is 307.93

real	6m12.914s

sent 17.07K bytes  received 14.02G bytes  102.70M bytes/sec
total size is 4.32T  speedup is 307.93

real	2m15.749s
Another rsync of 1+ GB over the same systems and networking respectively took 0m37.206s and 0m10.139s.

Code: Select all

sent 18.88K bytes  received 1.07G bytes  28.49M bytes/sec
total size is 4.32T  speedup is 4,041.54

real	0m37.206s

sent 15.06K bytes  received 1.07G bytes  101.75M bytes/sec
total size is 4.32T  speedup is 4,041.54

real	0m10.139s
These users thanked the author domih for the post:
odroid (Wed Mar 13, 2024 2:18 pm)

Post Reply

Return to “General Topics”

Who is online

Users browsing this forum: No registered users and 1 guest