Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

Parity Check Performance Problem (600 KB/s) - 120 days needed to finish...

Featured Replies

I have been running a parity for about 10 days now (not sure how to check the full length because when it pauses and restarts it resets the page timer). The first 30% or so ran normal as far as I could tell (50-100MB/s, took about 1.5 days which makes sense as my normal parity runs take 4-7 days). It has been at 500KB/s-1MB/s since then.

 

I am assuming I have a bad disk issue, however my DiskSpeed results are all where I expect them to be, maybe slightly slower... but nothing lower that 35MB/s)... nothing 600KB/s close. Any help would be much appreciated. This is the first parity I have run under unRAID 7 (don't think thats the issue but wanted to note it just in case).

 

This is a 2 parity (14TB each), 18 drive array, all ZFS. 185TB of 210TB used. My last successful parity run was 4-5 days on 12/25/24 and averaged 35MB/s. Logs attached.

Finally, I know those read/write numbers are pretty abysmal in general and would love to know if that just is what it is or if I can look into making that better at all? I guess I never really cared since it has 0 impact on Plex which is all its used for.

hades-diagnostics-20250206-1745.zip

  • Community Expert

It looks like you have writing going on to many of your disks in parallel to the parity check.  While this is happening both the check and the writing will be badly affected.

  • Author
5 minutes ago, itimpi said:

It looks like you have writing going on to many of your disks in parallel to the parity check.  While this is happening both the check and the writing will be badly affected.

writing from what? My parity runs mainly from 1am until 5pm so I should get excellent speeds from 1am-10am when Plex isnt in use no? Nothing else should be burning up the writing.

  • Author

fyi, these are my 4 drives with SMART Errors... Only Disk 3 actually ever reported errors on the main page but I have been monitoring all these drives in case its beyond just a bad connection on the cable/power failure etc etc.

 

DISK 1 (INTERNAL BAY)

    ATA Error Count: 4
        CR = Command Register [HEX]
        FR = Features Register [HEX]
        SC = Sector Count Register [HEX]
        SN = Sector Number Register [HEX]
        CL = Cylinder Low Register [HEX]
        CH = Cylinder High Register [HEX]
        DH = Device/Head Register [HEX]
        DC = Device Command Register [HEX]
        ER = Error register [HEX]
        ST = Status register [HEX]
    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 4 occurred at disk power-on lifetime: 5817 hours (242 days + 9 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 a8 f8 df 40 00   2d+16:13:46.938  READ FPDMA QUEUED
      60 00 00 a8 f4 df 40 00   2d+16:13:43.839  READ FPDMA QUEUED
      2f 00 01 10 00 00 00 00   2d+16:13:43.838  READ LOG EXT
      2f 00 01 10 00 00 00 00   2d+16:13:43.838  READ LOG EXT
      60 00 00 a8 f0 df 40 00   2d+16:13:40.917  READ FPDMA QUEUED

    Error 3 occurred at disk power-on lifetime: 5817 hours (242 days + 9 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 a8 f0 df 40 00   2d+16:13:43.838  READ FPDMA QUEUED
      60 00 00 a8 ec df 40 00   2d+16:13:40.916  READ FPDMA QUEUED
      60 f8 00 b0 e9 df 40 00   2d+16:13:40.906  READ FPDMA QUEUED
      60 08 00 a8 e8 df 40 00   2d+16:13:40.906  READ FPDMA QUEUED
      60 00 00 a8 e4 df 40 00   2d+16:13:40.902  READ FPDMA QUEUED

    Error 2 occurred at disk power-on lifetime: 4166 hours (173 days + 14 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 90 f7 df 40 00   3d+23:08:22.399  READ FPDMA QUEUED
      60 00 00 90 f3 df 40 00   3d+23:08:19.332  READ FPDMA QUEUED
      2f 00 01 10 00 00 00 00   3d+23:08:19.332  READ LOG EXT
      2f 00 01 10 00 00 00 00   3d+23:08:19.332  READ LOG EXT
      60 00 00 90 ef df 40 00   3d+23:08:16.585  READ FPDMA QUEUED

    Error 1 occurred at disk power-on lifetime: 4166 hours (173 days + 14 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 90 ef df 40 00   3d+23:08:19.332  READ FPDMA QUEUED
      60 00 00 90 eb df 40 00   3d+23:08:16.574  READ FPDMA QUEUED
      60 00 00 90 e7 df 40 00   3d+23:08:16.570  READ FPDMA QUEUED
      60 00 00 90 e3 df 40 00   3d+23:08:16.566  READ FPDMA QUEUED
      60 00 00 90 df df 40 00   3d+23:08:16.561  READ FPDMA QUEUED

 

DISK 3 (FRONT BAY)

        ATA Error Count: 3
        CR = Command Register [HEX]
        FR = Features Register [HEX]
        SC = Sector Count Register [HEX]
        SN = Sector Number Register [HEX]
        CL = Cylinder Low Register [HEX]
        CH = Cylinder High Register [HEX]
        DH = Device/Head Register [HEX]
        DC = Device Command Register [HEX]
        ER = Error register [HEX]
        ST = Status register [HEX]
    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 3 occurred at disk power-on lifetime: 23994 hours (999 days + 18 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 41 00 00 00 00 00  Error: ICRC, ABRT at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      61 80 08 80 10 02 40 00  30d+11:40:19.568  WRITE FPDMA QUEUED
      61 80 10 00 11 02 40 00  30d+11:40:19.567  WRITE FPDMA QUEUED
      61 80 00 00 10 02 40 00  30d+11:40:19.566  WRITE FPDMA QUEUED
      61 80 00 80 0f 02 40 00  30d+11:40:19.565  WRITE FPDMA QUEUED
      61 80 08 00 78 f1 40 00  30d+11:40:19.561  WRITE FPDMA QUEUED

    Error 2 occurred at disk power-on lifetime: 13182 hours (549 days + 6 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 41 00 00 00 00 00  Error: ICRC, ABRT at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      61 80 08 00 13 7f 40 00  27d+09:28:58.029  WRITE FPDMA QUEUED
      61 80 38 00 fa 7e 40 00  27d+09:28:58.028  WRITE FPDMA QUEUED
      61 80 30 80 fd 7e 40 00  27d+09:28:58.027  WRITE FPDMA QUEUED
      61 80 28 80 13 7f 40 00  27d+09:28:58.027  WRITE FPDMA QUEUED
      61 80 20 80 12 7f 40 00  27d+09:28:58.027  WRITE FPDMA QUEUED

    Error 1 occurred at disk power-on lifetime: 4352 hours (181 days + 8 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 41 00 00 00 00 00  Error: ICRC, ABRT at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      61 00 18 d0 aa cf 40 00  42d+03:40:30.866  WRITE FPDMA QUEUED
      61 00 10 d0 a8 cf 40 00  42d+03:40:30.851  WRITE FPDMA QUEUED
      61 00 08 d0 a6 cf 40 00  42d+03:40:30.851  WRITE FPDMA QUEUED
      61 00 00 d0 a4 cf 40 00  42d+03:40:30.851  WRITE FPDMA QUEUED
      61 08 38 58 b7 df 40 00  42d+03:40:30.850  WRITE FPDMA QUEUED

 

DISK 12 (FAN BAY)

    ATA Error Count: 4
        CR = Command Register [HEX]
        FR = Features Register [HEX]
        SC = Sector Count Register [HEX]
        SN = Sector Number Register [HEX]
        CL = Cylinder Low Register [HEX]
        CH = Cylinder High Register [HEX]
        DH = Device/Head Register [HEX]
        DC = Device Command Register [HEX]
        ER = Error register [HEX]
        ST = Status register [HEX]
    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 4 occurred at disk power-on lifetime: 9633 hours (401 days + 9 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 f8 af 20 40 00      17:13:06.322  READ FPDMA QUEUED
      60 08 00 f0 af 20 40 00      17:13:03.484  READ FPDMA QUEUED
      60 00 00 f0 ab 20 40 00      17:13:03.480  READ FPDMA QUEUED
      60 00 00 f0 a7 20 40 00      17:13:03.476  READ FPDMA QUEUED
      60 f8 00 f8 a3 20 40 00      17:13:03.472  READ FPDMA QUEUED

    Error 3 occurred at disk power-on lifetime: 9633 hours (401 days + 9 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 f0 6b 20 40 00      17:13:03.147  READ FPDMA QUEUED
      60 00 00 f0 67 20 40 00      17:13:00.240  READ FPDMA QUEUED
      60 00 00 f0 63 20 40 00      17:12:58.043  READ FPDMA QUEUED
      60 00 00 f0 5f 20 40 00      17:12:58.039  READ FPDMA QUEUED
      60 00 00 f0 5b 20 40 00      17:12:58.035  READ FPDMA QUEUED

    Error 2 occurred at disk power-on lifetime: 9633 hours (401 days + 9 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 f8 37 20 40 00      17:12:57.939  READ FPDMA QUEUED
      60 00 00 f8 33 20 40 00      17:12:55.096  READ FPDMA QUEUED
      60 00 00 f8 2f 20 40 00      17:12:55.092  READ FPDMA QUEUED
      60 00 00 f8 2b 20 40 00      17:12:55.088  READ FPDMA QUEUED
      60 00 00 f8 27 20 40 00      17:12:55.079  READ FPDMA QUEUED

    Error 1 occurred at disk power-on lifetime: 9633 hours (401 days + 9 hours)
      When the command that caused the error occurred, the device was active or idle.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      40 41 00 00 00 00 00  Error: UNC at LBA = 0x00000000 = 0

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      60 00 00 f0 0b 20 40 00      17:12:54.997  READ FPDMA QUEUED
      60 00 00 f0 07 20 40 00      17:12:52.144  READ FPDMA QUEUED
      60 00 00 f0 03 20 40 00      17:12:52.141  READ FPDMA QUEUED
      60 00 00 f0 ff 1f 40 00      17:12:52.137  READ FPDMA QUEUED
      60 00 00 f0 fb 1f 40 00      17:12:52.131  READ FPDMA QUEUED

 

DISK 15 (EXTERNAL BAY)

    ATA Error Count: 73 (device log contains only the most recent five errors)
        CR = Command Register [HEX]
        FR = Features Register [HEX]
        SC = Sector Count Register [HEX]
        SN = Sector Number Register [HEX]
        CL = Cylinder Low Register [HEX]
        CH = Cylinder High Register [HEX]
        DH = Device/Head Register [HEX]
        DC = Device Command Register [HEX]
        ER = Error register [HEX]
        ST = Status register [HEX]
    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 73 occurred at disk power-on lifetime: 4049 hours (168 days + 17 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 53 70 47 57 14 40  Error: ICRC, ABRT 112 sectors at LBA = 0x00145747 = 1333063

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      35 00 40 78 56 14 e0 08  19d+14:10:47.052  WRITE DMA EXT
      35 00 40 38 51 14 e0 08  19d+14:10:47.048  WRITE DMA EXT
      25 00 08 c0 e6 68 e0 08  19d+14:10:47.048  READ DMA EXT
      25 00 08 c8 97 1a e0 08  19d+14:10:47.032  READ DMA EXT
      35 00 40 f8 4b 14 e0 08  19d+14:10:47.031  WRITE DMA EXT

    Error 72 occurred at disk power-on lifetime: 4049 hours (168 days + 17 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 53 c1 66 d0 9f 40  Error: ICRC, ABRT 193 sectors at LBA = 0x009fd066 = 10473574

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      35 00 40 e8 cc 9f e0 08  19d+14:10:36.760  WRITE DMA EXT
      35 00 40 a8 c7 9f e0 08  19d+14:10:36.737  WRITE DMA EXT
      35 00 40 68 c2 9f e0 08  19d+14:10:36.736  WRITE DMA EXT
      35 00 40 28 bd 9f e0 08  19d+14:10:36.734  WRITE DMA EXT
      35 00 40 e8 b7 9f e0 08  19d+14:10:36.733  WRITE DMA EXT

    Error 71 occurred at disk power-on lifetime: 4049 hours (168 days + 17 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 53 b1 fe c0 93 40  Error: ICRC, ABRT 177 sectors at LBA = 0x0093c0fe = 9683198

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      35 00 40 70 bf 93 e0 08  19d+14:10:25.846  WRITE DMA EXT
      35 00 40 30 ba 93 e0 08  19d+14:10:25.834  WRITE DMA EXT
      25 00 08 70 71 40 e0 08  19d+14:10:25.834  READ DMA EXT
      35 00 40 f0 b4 93 e0 08  19d+14:10:25.833  WRITE DMA EXT
      35 00 40 b0 af 93 e0 08  19d+14:10:25.805  WRITE DMA EXT

    Error 70 occurred at disk power-on lifetime: 4049 hours (168 days + 17 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 53 f1 66 64 97 40  Error: ICRC, ABRT 241 sectors at LBA = 0x00976466 = 9921638

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      35 00 40 18 63 97 e0 08  19d+14:09:29.972  WRITE DMA EXT
      35 00 40 d8 5d 97 e0 08  19d+14:09:29.957  WRITE DMA EXT
      35 00 40 98 58 97 e0 08  19d+14:09:29.951  WRITE DMA EXT
      35 00 40 58 53 97 e0 08  19d+14:09:29.943  WRITE DMA EXT
      35 00 40 18 4e 97 e0 08  19d+14:09:29.937  WRITE DMA EXT

    Error 69 occurred at disk power-on lifetime: 4006 hours (166 days + 22 hours)
      When the command that caused the error occurred, the device was doing SMART Offline or Self-test.

      After command completion occurred, registers were:
      ER ST SC SN CL CH DH
      -- -- -- -- -- -- --
      84 53 30 d7 73 84 40  Error: ICRC, ABRT 48 sectors at LBA = 0x008473d7 = 8680407

      Commands leading to the command that caused the error were:
      CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
      -- -- -- -- -- -- -- --  ----------------  --------------------
      35 00 40 c8 72 84 e0 08  17d+19:44:14.664  WRITE DMA EXT
      25 00 08 78 bf b0 e0 08  17d+19:44:14.661  READ DMA EXT
      25 00 08 70 c0 b0 e0 08  17d+19:44:14.660  READ DMA EXT
      25 00 08 70 bf b0 e0 08  17d+19:44:14.660  READ DMA EXT
      35 00 40 88 6d 84 e0 08  17d+19:44:14.656  WRITE DMA EXT

 

  • Author

DISREGARD, I mean, definitely see if I am messing something up but I realized manual parity checks were set to no and this was a manual one since I cancelled the scheduled one when I noticed the issue initially and needed to reboot the system.

 

also, one more thing... am I using Parity Check Tuning wrong? I just noticed its still running despite being after my time was up... Shouldnt it have paused at 4pm, its 6:15pm ET now)

 

image.thumb.png.62b22e8a992017f46c26e9b1626ed329.png

Edited by wickedathletes
I am a dolt and missed a setting

  • Author
25 minutes ago, itimpi said:

It looks like you have writing going on to many of your disks in parallel to the parity check.  While this is happening both the check and the writing will be badly affected.

not sure if there is a better way to tell but I don't see anything

image.thumb.png.f910e2fbffe568ea916ee415529f2a70.png

  • Community Expert

According to that screenshot there is something reading from multiple disks

  • Community Expert
10 hours ago, wickedathletes said:

also, one more thing... am I using Parity Check Tuning wrong? I just noticed its still running despite being after my time was up... Shouldnt it have paused at 4pm, its 6:15pm ET now)

Your screenshot for the parity Tuning settings shows you are not using increments for Manual checks.

  • Author
5 hours ago, JorgeB said:

According to that screenshot there is something reading from multiple disks

Sorry, I was referring to the writes.

  • Author
3 hours ago, itimpi said:

Your screenshot for the parity Tuning settings shows you are not using increments for Manual checks.

Yes, I noticed that and adjusted. I shut down all dockers this AM and the best I’m getting is 3MB/s.

 

any other guesses on what could be writing/reading from my disks? Also, why would it run at 30-40MB/s for the first 30-40% then immediately fall apart for the rest of the time? If it was a read/write activity it would be consistent based on activity no? My server activity is roughly the same daily.

  • Community Expert
1 hour ago, wickedathletes said:

Sorry, I was referring to the writes.

Any array reads will also slow down the parity check.

  • Author
34 minutes ago, JorgeB said:

Any array reads will also slow down the parity check.

any best guess on what is the best thing I can look into that is causing this? The first 30-40% run at 30-40MB/s. Then it crawls to 600KB/s. I shutdown all dockers and the best I can get it to is 5MB/s. This is reproducible and have little fluctuation depending on time of day (which to me means this isn't a read/write issue its something else.

Nothing has changed since my last parity run besides all drives now being ZFS (I believe I had a few that still weren't in december) and unraid 7.0 full version (was on beta in dec.). My usage patterns haven't changed in years beyond maybe some more friends in plex (but currently Plex is off anyways).

  • Community Expert
51 minutes ago, wickedathletes said:

The first 30-40% run at 30-40MB/s.

And this was the normal speed?

 

Start a parity check in maintenance mode and let it run for a few minutes, to confirm nothing is really assessing the array, if it's still slow, run the diskspeed docker tests.

  • Author
4 hours ago, JorgeB said:

And this was the normal speed?

 

Start a parity check in maintenance mode and let it run for a few minutes, to confirm nothing is really assessing the array, if it's still slow, run the diskspeed docker tests.

diskspeed docker gives me 35-85 MB/s depending on the drive, I ran that a few days ago, with and without the parity being run, so oddly enough parity process had no impact to the diskspeed performance test. After shutting down all my dockers I am up to about 5-7MB/s and re-enabling them a few hours ago didn't impact that change, but something is still drawing 20+ MB/s from my typical times. And, as mentioned the first 40% or so runs at normal speed and then it crashes. This was seen 3 times in a row. At this point I don't want to restart it again though as at least now i am in a 20 days until finish window and its been running for over 10 days now. I would assume I would see the same thing again (for a 4th time) if I checked it. This parity to me is important because I have 2 drives I have been trying to add to the system to replace older ones but I want to make sure all is well before I do it, and I have major concerns that if a parity check is taking this long I can't have a data rebuild take 40 days...

 

Also, one of the reasons I don't think its read/write is because I just kicked off a data transfer to my PC (200GB) and I am getting 40MB/s and parity check didnt change at all, still 5-7MB/s. Something else is happening I just don't know what. If parity is running and draining my server that bad then shouldn't everything be getting crushed to those speeds too?

Edited by wickedathletes

  • Community Expert
11 hours ago, wickedathletes said:

And, as mentioned the first 40% or so runs at normal speed and then it crashes

That could mean a disk or more with slow sectors, diskspeed test sometimes shows this, but not always, since only a small percentage of the disk surface is tested, would still like to see the graphs, since the speeds you mentioned seem very low for modern disks.

  • Author
11 hours ago, JorgeB said:

That could mean a disk or more with slow sectors, diskspeed test sometimes shows this, but not always, since only a small percentage of the disk surface is tested, would still like to see the graphs, since the speeds you mentioned seem very low for modern disks.

Attached is from when I ran it a day or so ago. Also, I am back down to  700KB/s again. killing dockers 2 nights ago got it up to 7MB/s and restoring the dockers it stayed at 7MB/s but today its been 700KB/s all day.

 

image.thumb.png.68d6b82ca243ab03d261d19fec6dc407.png

  • Community Expert

Well, there's definitely some issue, only two disks are performing normally, all the other ones are super slow, was the test done with the array running or stopped?

 

 

  • Author
8 hours ago, JorgeB said:

Well, there's definitely some issue, only two disks are performing normally, all the other ones are super slow, was the test done with the array running or stopped?

 

 

these results are with the array running (and they produce the same with parity paused or running). I know something is wrong, just dont know what or how to even diagnose what is happening. Everything is functioning as it normally does minus parity running like trash. My recent logs attached.

hades-diagnostics-20250209-1433.zip

Edited by wickedathletes

  • Community Expert

Boot in in safe mode, disable docker and VM services, and see if you still see disk reads.

  • Author
7 hours ago, JorgeB said:

Boot in in safe mode, disable docker and VM services, and see if you still see disk reads.

So running a parity under safe mode has me at about 190 MB/s (its early, sub 1%). I will assume this will be the case until about 40% and then I will hit the same snag area. If not, maybe im good? But when I did it over the last few tries this month 0-40% was roughly 35MB/s (my normal, again with docker and everything running, not in safe mode) and then it haults for no real reason that I can see.

 

Will monitor.

 

image.png.6e22c9dd35d011dd591008cfa63a15c1.png

Edited by wickedathletes

  • Author

Is it common to see the speeds go down over time? 7hrs in (27%) and down to 96MB/s. until about 15% it maintained the 180-190MB/s.

  • Community Expert
1 minute ago, wickedathletes said:

Is it common to see the speeds go down over time? 7hrs in (27%) and down to 96MB/s. until about 15% it maintained the 180-190MB/s.

Yes.  Access speeds of HDD drives decrease as you go from access the first sector to the very last.  For my 16TB drives, that is nearly a 50% decrease.  Parity speeds track with the speed of the drive with the slowest access speed.

  • Author
On 2/10/2025 at 2:58 AM, JorgeB said:

Boot in in safe mode, disable docker and VM services, and see if you still see disk reads.

so running in safe mode had it finish with an average speed of 131MB/s. I noticed my cpu performance was pegged and I disabled folder cache plugin. This solved my processor issues, wondering if that was causing everything? The odd thing is this only started happening since 7.0 release. It worked as it normally did back on the RC's (I run parity every few months). No folder structures of cache folder settings changed, so not sure. I will see how it does with the array running soon (I have a scheduled on in 2 weeks).

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.