[ale] SATA drive not recognized by BIOS
Steve Tynor
stevejunk at iintiip.com
Sun Oct 15 20:50:32 EDT 2023
Thanks Jim. It's a software RAID via mdadm. The array was
originally built with Ubuntu 18 - have kept it going through various
system upgrades - now on Ubuntu 22. And you are right - I misspoke when
calling it LVM RAID - if my notes are right from back then, I created it
via the "full disk" partitions:
mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2
/dev/sdb1 /dev/sdc1
When these errors start happening, it brings the machine to its
knees. Anything I can do to "repair" the array to avoid those
segments? Even when the errors are happening, mdstat looks healthy:
ricotta:~> cat /proc/mdstat
Personalities : [raid1] [linear] [multipath] [raid0] [raid6] [raid5]
[raid4] [raid10]
md0 : active raid1 sdc1[0] sdb1[1]
3906884608 blocks super 1.2 [2/2] [UU]
bitmap: 10/30 pages [40KB], 65536KB chunk
unused devices: <none>
For now I've just removed the errant disk from the array again pending
inspiration...
Steve
On 2023-10-15 5:47 PM, Jim Kinney wrote:
> The Errors look like the raid recognized a block/segment failure on
> sdb and handled it by redirects to a new location and copied in the
> data from the mirror.
>
> Is the the drive bad? Not yet. Sectors do fail in time.
>
> But raid system specified sdb, not sdb1, so the raid subsystem is
> either using the entire drive for sdb or there's a raid controlling
> hardware or software that is doing the low level hardware management
> and that layer needs to be worked on to recognize the new drive. It
> really sounds like there's a raid controller in there somewhere.
>
> On Sun, Oct 15, 2023, 2:40 PM Steve Tynor via Ale <ale at ale.org> wrote:
>
> And I guess one more question... I was not able to learn much
> about these errors from the interwebs, but perhaps they don't
> necessarily suggest an actual drive failure but some sort of soft
> problem with the array - perhaps repartition the bad drive and add
> it back to the array and hope the errors don't come back? The
> errors look scary to me, but most people on this list have a lot
> more sysadm chops than I do...
>
> Oct 13 21:11:48 ricotta kernel: [714090.323895] md/raid1:md0: read
> error correct
> ed (8 sectors at 7339070832 on sdb1)
> Oct 13 21:12:01 ricotta kernel: [714104.108320] md/raid1:md0:
> redirecting sector
> 7338806424 to other mirror: sdb1
> Oct 13 21:12:09 ricotta kernel: [714111.347389] ata2.00: exception
> Emask 0x0 SAc
> t 0x140000 SErr 0x0 action 0x0
> Oct 13 21:12:09 ricotta kernel: [714111.347501] ata2.00: irq_stat
> 0x40000008
> Oct 13 21:12:09 ricotta kernel: [714111.347558] ata2.00: failed
> command: READ FP
> DMA QUEUED
> Oct 13 21:12:09 ricotta kernel: [714111.347621] ata2.00: cmd
> 60/78:90:98:5c:71/0
> 1:00:b5:01:00/40 tag 18 ncq dma 192512 in
> Oct 13 21:12:09 ricotta kernel: [714111.347621] res
> 41/40:00:88:5d:71/0
> 0:00:b5:01:00/00 Emask 0x409 (media error) <F>
> Oct 13 21:12:09 ricotta kernel: [714111.347809] ata2.00: status: {
> DRDY ERR }
> Oct 13 21:12:09 ricotta kernel: [714111.347860] ata2.00: error: {
> UNC }
> Oct 13 21:12:09 ricotta kernel: [714111.350759] ata2.00:
> configured for UDMA/133
> Oct 13 21:12:09 ricotta kernel: [714111.350801] sd 1:0:0:0: [sdb]
> tag#18 FAILED
> Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=7s
> Oct 13 21:12:09 ricotta kernel: [714111.350811] sd 1:0:0:0: [sdb]
> tag#18 Sense K
> ey : Medium Error [current]
> Oct 13 21:12:09 ricotta kernel: [714111.350817] sd 1:0:0:0: [sdb]
> tag#18 Add. Se
> nse: Unrecovered read error - auto reallocate failed
> Oct 13 21:12:09 ricotta kernel: [714111.350824] sd 1:0:0:0: [sdb]
> tag#18 CDB: Re
> ad(16) 88 00 00 00 00 01 b5 71 5c 98 00 00 01 78 00 00
> Oct 13 21:12:09 ricotta kernel: [714111.350827]
> blk_update_request: I/O error, d
> ev sdb, sector 7339072664 op 0x0:(READ) flags 0x0 phys_seg 47 prio
> class 0
> Oct 13 21:12:09 ricotta kernel: [714111.351018] md/raid1:md0:
> sdb1: rescheduling
> sector 7338806424
> Oct 13 21:12:09 ricotta kernel: [714111.351111] ata2: EH complete
> Oct 13 21:12:20 ricotta kernel: [714122.174015] md/raid1:md0:
> redirecting sector
> 7338806424 to other mirror: sdb1
> Oct 13 21:12:24 ricotta kernel: [714126.594981] ata2.00: exception
> Emask 0x0 SAc
> t 0x1000 SErr 0x0 action 0x0
> Oct 13 21:12:24 ricotta kernel: [714126.595089] ata2.00: irq_stat
> 0x40000008
> Oct 13 21:12:24 ricotta kernel: [714126.595146] ata2.00: failed
> command: READ FP
> DMA QUEUED
> Oct 13 21:12:24 ricotta kernel: [714126.595209] ata2.00: cmd
> 60/78:60:98:5c:71/0
> 1:00:b5:01:00/40 tag 12 ncq dma 192512 in
> Oct 13 21:12:24 ricotta kernel: [714126.595209] res
> 41/40:00:78:5d:71/0
> 0:00:b5:01:00/00 Emask 0x409 (media error) <F>
> Oct 13 21:12:24 ricotta kernel: [714126.603898] ata2.00: status: {
> DRDY ERR }
> Oct 13 21:12:24 ricotta kernel: [714126.608263] ata2.00: error: {
> UNC }
> Oct 13 21:12:24 ricotta kernel: [714126.615103] ata2.00:
> configured for UDMA/133
> Oct 13 21:12:24 ricotta kernel: [714126.615136] sd 1:0:0:0: [sdb]
> tag#12 FAILED
> Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=4s
> Oct 13 21:12:24 ricotta kernel: [714126.615145] sd 1:0:0:0: [sdb]
> tag#12 Sense K
> ey : Medium Error [current]
> Oct 13 21:12:24 ricotta kernel: [714126.615151] sd 1:0:0:0: [sdb]
> tag#12 Add. Se
> nse: Unrecovered read error - auto reallocate failed
> Oct 13 21:12:24 ricotta kernel: [714126.615158] sd 1:0:0:0: [sdb]
> tag#12 CDB: Re
> ad(16) 88 00 00 00 00 01 b5 71 5c 98 00 00 01 78 00 00
> Oct 13 21:12:24 ricotta kernel: [714126.615162]
> blk_update_request: I/O error, d
> ev sdb, sector 7339072664 op 0x0:(READ) flags 0x0 phys_seg 47 prio
> class 0
> Oct 13 21:12:24 ricotta kernel: [714126.619470] md/raid1:md0:
> sdb1: rescheduling
> sector 7338806424
> Oct 13 21:12:24 ricotta kernel: [714126.623671] ata2: EH complete
>
> On 2023-10-15 3:01 PM, Steve Tynor via Ale wrote:
>>
>> Forgot to include: the original WD drives are also "AF", so I
>> think that rules out sector size incompatibility.
>>
>> Steve
>>
>>
>> On 2023-10-15 2:16 PM, Steve Tynor via Ale wrote:
>>>
>>> I'm not certain how to be sure. A quick duckduckgo search
>>> suggests that 4k native sector also being marketed as "Advanced
>>> Format" - is that right? Looking at the Toshiba spec page, I see
>>> AF listed as "Yes" for the 8T (the first drive I tried) and "No"
>>> for the 4T (which also didn't work).
>>>
>>> https://storage.toshiba.com/docs/support-docs/toshiba_n300_salessheet_english_07-27-21.pdf?Status=Master
>>>
>>> Steve
>>>
>>> On 2023-10-15 2:08 PM, Robert Tweedy via Ale wrote:
>>>> Hi Steve,
>>>>
>>>> Out of curiosity, are the new drives "4K native" sector drives,
>>>> and is this the same as the old drives you're replacing? If the
>>>> system's old enough its BIOS might not support that & is
>>>> expecting to have "512n" (512-native) or "512e" (512-emulated)
>>>> sector-size drives.
>>>>
>>>> Other than that, I'm not sure what it could be beyond being
>>>> Toshiba brand (which I've personally never had good experience
>>>> with, but that's only anecdotal).
>>>>
>>>> -Robert
>>>>
>>>
>>> _______________________________________________
>>> Ale mailing list
>>> Ale at ale.org
>>> https://mail.ale.org/mailman/listinfo/ale
>>> See JOBS, ANNOUNCE and SCHOOLS lists at
>>> http://mail.ale.org/mailman/listinfo
>>
>> _______________________________________________
>> Ale mailing list
>> Ale at ale.org
>> https://mail.ale.org/mailman/listinfo/ale
>> See JOBS, ANNOUNCE and SCHOOLS lists at
>> http://mail.ale.org/mailman/listinfo
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> https://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.ale.org/pipermail/ale/attachments/20231016/d733da55/attachment.htm>
More information about the Ale
mailing list