External USB does not work on BPI-M2 Berry

I have an issue with an external USB Drive: Once connecting it works for some minutes, but after a short time I see in dmesg the following:

> [ 1389.748243] sd 1:0:0:0: rejecting I/O to offline device
> [ 1389.748357] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1389.748425] sd 1:0:0:0: rejecting I/O to offline device
> [ 1389.748471] EXT4-fs error (device sda1): __ext4_journal_start_sb:62: Detected aborted journal
> [ 1389.748482] EXT4-fs (sda1): Remounting filesystem read-only
> [ 1389.748496] EXT4-fs (sda1): previous I/O error to superblock detected
> [ 1389.748654] sd 1:0:0:0: rejecting I/O to offline device
> [ 1389.753237] sd 1:0:0:0: rejecting I/O to offline device
> [ 1389.753332] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1389.947292] sd 1:0:0:0: rejecting I/O to offline device
> [ 1389.947401] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1389.951547] sd 1:0:0:0: rejecting I/O to offline device
> [ 1389.951632] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1391.037660] sd 1:0:0:0: rejecting I/O to offline device
> [ 1391.037766] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1391.042290] sd 1:0:0:0: rejecting I/O to offline device
> [ 1391.042377] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1391.314495] sd 1:0:0:0: rejecting I/O to offline device
> [ 1391.314608] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1391.318626] sd 1:0:0:0: rejecting I/O to offline device
> [ 1391.318714] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1394.335119] sd 1:0:0:0: rejecting I/O to offline device
> [ 1394.335243] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1394.339274] sd 1:0:0:0: rejecting I/O to offline device
> [ 1394.339364] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1394.826744] sd 1:0:0:0: rejecting I/O to offline device
> [ 1394.826861] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1394.831320] sd 1:0:0:0: rejecting I/O to offline device
> [ 1394.831407] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1395.031224] sd 1:0:0:0: rejecting I/O to offline device
> [ 1395.031456] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1395.035344] sd 1:0:0:0: rejecting I/O to offline device
> [ 1395.035432] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1396.657433] sd 1:0:0:0: rejecting I/O to offline device
> [ 1396.657549] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1396.662368] sd 1:0:0:0: rejecting I/O to offline device
> [ 1396.662463] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1396.918469] sd 1:0:0:0: rejecting I/O to offline device
> [ 1396.918597] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1396.922560] sd 1:0:0:0: rejecting I/O to offline device
> [ 1396.922648] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1398.836810] sd 1:0:0:0: rejecting I/O to offline device
> [ 1398.836926] EXT4-fs (sda1): previous I/O error to superblock detected
> [ 1398.836967] sd 1:0:0:0: rejecting I/O to offline device
> [ 1398.837041] EXT4-fs error (device sda1): ext4_find_entry:1369: inode #27789314: comm bash: reading directory lblock 0
> [ 1403.926314] sd 1:0:0:0: rejecting I/O to offline device
> [ 1403.926402] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1403.930481] sd 1:0:0:0: rejecting I/O to offline device
> [ 1403.930571] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1404.889694] sd 1:0:0:0: rejecting I/O to offline device
> [ 1404.889782] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1404.894043] sd 1:0:0:0: rejecting I/O to offline device
> [ 1404.894131] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1405.072990] sd 1:0:0:0: rejecting I/O to offline device
> [ 1405.073075] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1405.076827] sd 1:0:0:0: rejecting I/O to offline device
> [ 1405.076907] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)
> [ 1422.815983] sd 1:0:0:0: rejecting I/O to offline device
> [ 1422.816151] EXT4-fs warning (device sda1): __ext4_read_dirblock:911: error reading directory block (ino 27789314, block 0)

Also tried an USB-Y-Cable with externel USB power Supply, but will not work either. Anybody who tried an external HDD with the BPI?

everything works well

What kind of power supply are you using? Does your external HDD has its own Power Supply?

(I’m using Micro USB PS)

I’m using a 2A Power Supply with Micro USB. I tried with a different HDD, but no difference. The external HDD does not have a power supply, and it has a USB 3.0 connection. I also tried an Y-Cable to get more power, but that doesn’t help either …

What I noticed is, that the error occours mostly when longer write action is performed. Speaks for power issue … but how to solve?

The worse thing is also - once the issue came, I need to reboot the machine, as it doesn’t can been forced back to work …

I have exactly the same problem on a BPI-M2 Ultra. I tried a couple of different HDD closures with the same result, also I tried with different disk without any improvement.

As you said the problem is when you trying write operations, once the error happens it freezes for 2-5min and the it remount the filesystem as readonly, the only way to recover from that I found is just to reboot the whole system.

Unfortunately I don’t have a fix for it :frowning:

Same for me… I moved the usb drives to an other sbc… Before this, I have tried several hard drives, also ssd and usb stick… of course different power supplies… Always the same writing a big bunch of data to a USB drive will crash the connection to the drive… I think it is not a power issue, I think it is timing issue… Actually have posted this issue months ago, without any response…

I have the same problems … the connection to my USB drive crahes during intensive write access. It’s a WD-Elements harddrive with external power supply! Therefore it can’t be a power problem (the drive works fine on any other computer).

System: Debian Stretch 3.10.107 2017-08-28

Here are the messages from syslog:

Dec  2 06:47:40 tuxcloud kernel: [344841.180138] usb 1-1: reset high-speed USB device number 2 using sunxi-ehci
Dec  2 06:47:51 tuxcloud kernel: [344851.800288] usb 1-1: reset high-speed USB device number 2 using sunxi-ehci

Message from syslogd@tuxcloud at Dec  2 06:47:56 ...
 kernel:[344856.881687] journal commit I/O error
Dec  2 06:47:56 tuxcloud kernel: [344856.880675] usb 1-1: USB disconnect, device number 2
Dec  2 06:47:56 tuxcloud kernel: [344856.880720] sd 1:0:0:0: Device offlined - not ready after error recovery
Dec  2 06:47:56 tuxcloud kernel: [344856.880771] sd 1:0:0:0: [sdb] Unhandled error code

Does anybody have an idea how to get more system information about this issue?

I also tested “Ubuntu mate 3.10.108 2017-11-13” with the same error. Could it be a hardware defect?

Thanks a lot!

Does anyone has any news or a clue on what’s going on?

I guess I found a solution - I had the following entry in /etc/fstab for the external HDD:

LABEL=DATA /data xfs rw,user,exec,auto,nofail 0 2

I just thought about, that it could be relying to a cache overflow of some kind, so I have added wsync to disable the cache:

LABEL=DATA /data xfs rw,user,exec,auto,nofail,wsync 0 2

And so, I didn’t got the issue since a couple of days (keep fingers crossed …)

Okay, the previous thing does it not solve entirely - next thing looks promisiing: hdparm -W 0 /dev/sda1 So afterwards I was able to save huge files without issues.

It seems so: hdparm -W 0 /dev/sda1 did it

hello, Did you use sata => usb to connect M2 Berry? and have this problem?

Did you try on board sata interface?

No, I’m using the “normal” USB-Interface to connect to an external USB-HDD. As the external HDD is capsuled, i didn’t have even the chance to connect to the SATA port.

could you take a picture? and please tell me what is your USB HDD device , I’ll do a test, thanks.