Re: Ceph Performance of Micron 5210 SATA?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> 
> An observation: We are using proxmox (5.4), and it displays the WEAR level as "N/A", which is unfortunate... :-( Tried upgrading to 6: the same, still no wearout in the GUI.
> 
> Here is smartctl output:
> 
>> smartctl 6.6 2016-05-31 r4324 [x86_64-linux-4.15.18-26-pve] (local build)

Old release, probably came with your OS.

>> Device Model:     Micron_5210_MTFDDAK3T8QDE
>> Firmware Version: D2MU404

Make sure you run the very latest public firmware from Micron.  Ask them if this is current.

>> Device is:        Not in smartctl database [for details use: -P showall]

There you go.

>> 202 Unknown_SSD_Attribute   0x0030   100   100   001    Old_age   Offline      -       0

SMART is not consistently implemented.  Some manufacturers / models use different attribute numbers.  Some eg. will present a counter as lifetime remaining, others as lifetime consumed.  And the calculations used to present lifetime percentages varies among models and may not be the best predictor of failure, eg. right before the drive is going to fail the number may change non-linearly. 

But it’s better than nothing, and I digress.

Some tools match the label string returned by smartctl instead of or in addition to the number.  smartctl will default strings for certain attributes (which sometimes gives wrong answers) if there isn’t an exact match to the actual drive model.

So I bet if you were to add

    "Micron_5210_(EE|MT)FDDAK(1T9|3T8|7T6)QDE|"                        // tested with Micron_5210_MTFDDAK1T9QDC

to your drivedb.h stanza for the 5100/5200, proxmox might pick it up.

Your 6.6 drivedb.h may not even have a 5100/5200 entry.  The latest upstream is here

https://www.smartmontools.org/browser/trunk/smartmontools/drivedb.h

But beware that IIRC the bits at the beginning of the file changed with 7.0, so you won’t be able to inhale the latest file in toto, and it doesn’t actually list the 5210.  Someone from Micron submitted an omnibus update that included it, but it conflicted with the existing 5100/5200 entry, and I haven’t gotten around to submitting a reconciled version.

With the caveat that I’m composing this in my MUA and have not tested it, you might try adding the below to your current drivedb.h.  Now that I again have a 5100 Pro to test against I need to get off my butt and submit a diff upstream so it gets into 7.2.


Be careful with the syntax, it’s easy to get the |{,} characters wrong.


// Reference: https://www.micron.com/resource-details/feec878a-265e-49a7-8086-15137c5f9011
  // TN-FD-34: 5100 SSD SMART Implementation
  {
    "Micron 5100 Pro / 5200 SSDs",
    "(Micron_5100_)?(EE|MT)FDDA[KV](240|480|960|1T9|3T8|7T6)T(BY|CB|CC)|" // Matches both stock and Dell OEM
    "Micron_5210_(EE|MT)FDDAK(1T9|3T8|7T6)QDE|"                        // tested with Micron_5210_MTFDDAK1T9QDC
    "(Micron_5200_)?MTFDDAK(480|960|1T9|3T8|7T6)TD(C|D|N)", // tested with Micron_5200_MTFDDAK3T8TDD/D1MU505
    "", "",
  //"-v 1,raw48,Raw_Read_Error_Rate "
  //"-v 5,raw48,Reallocated_Block_Count "
  //"-v 9,raw24(raw8),Power_On_Hours "  // raw24(raw8)??
  //"-v 12,raw48,Power_Cycle_Count "
    "-v 170,raw48,Reserved_Block_Pct " // Percentage of remaining reserved blocks available
    "-v 171,raw48,Program_Fail_Count "
    "-v 172,raw48,Erase_Fail_Count "
    "-v 173,raw48,Avg_Block-Erase_Count "
    "-v 174,raw48,Unexpect_Power_Loss_Ct "
  //"-v 180,raw48,Reserved_Block_Count " // absolute count of remaining reserved blocks available
    "-v 183,raw48,SATA_Int_Downshift_Ct " // SATA speed downshift count
  //"-v 184,raw48,Error_Correction_Count "
  //"-v 187,raw48,Reported_Uncorrect " // Number of UECC correction failures
  //"-v 188,raw48,Command_Timeouts "
  //"-v 194,tempminmax,Temperature_Celsius " // 100 - degrees C, wraps: 101 reported as 255
  //"-v 195,raw48,Cumulativ_Corrected_ECC "
  //"-v 196,raw48,Reallocation_Event_Ct "
  //"-v 197,raw48,Current_Pending_Sector " // Use the raw value
  //"-v 198,raw48,Offline_Uncorrectable "  // Use the raw value
  //"-v 199,raw48,UDMA_CRC_Error_Count "   // Use the raw value
    "-v 202,raw48,Percent_Lifetime_Remain " // Remaining endurance, trips at 10%
    "-v 206,raw48,Write_Error_Rate "
    "-v 210,raw48,RAIN_Success_Recovered "  // Total number of NAND pages recovered by RAIN
    "-v 211,raw48,Integ_Scan_Complete_Cnt "  // Number of periodic data integrity scans completed
    "-v 212,raw48,Integ_Scan_Folding_Cnt "   // Number of blocks reallocated by integrity scans
    "-v 213,raw48,Integ_Scan_Progress "      // Current is percentage, raw is absolute number of superblocks scanned by the current in\
tegrity scan
    "-v 247,raw48,Host_Program_Page_Count "
    "-v 248,raw48,Bckgnd_Program_Page_Cnt"
  },
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx




[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux