Neo2 Board with OMV Erasmus, weird behavior in SATA disks (UAS issue?)

  • Hi all,


    I'm facing a weird behavior on the internal SATA disk attached into a Neo2 with NAS 1.2 kit (Sata Seagate M3 disk) and also in any attached USB disk when reading thousands of small files from a TimeCapsule share I have configured within the OMV (3.0.86 Erasmus with 4.11.2 Kernel version), processor Allwinnersun50iw2Family.


    Code
    # uname -a
    Linux dataNEO2 4.11.2 #22 SMP Sat Jul 27 11:14:47 CST 2017 aarch64 GNU/Linux








    After some time backing up the information within the SATA disk to the cloud, I receive errors reading caused by scsi host0: uas_eh_bus_reset_handler start" that causes the FS to be in wrong state:



    Code
    dataNEO2 kernel: EXT4-fs error (device sda1): ext4_xattr_block_get:324: inode #11272713: comm rclone: bad block 45097517
    dataNEO2 kernel: EXT4-fs error (device sda1): ext4_xattr_block_get:324: inode #11272714: comm rclone: bad block 45097518
    dataNEO2 kernel: EXT4-fs error (device sda1): ext4_xattr_block_get:324: inode #11272715: comm rclone: bad block 45097519
    dataNEO2 afpd: ad_open_hf_ea: unexpected: Structure needs cleaning
    dataNEO2 afpd: afp_openfork(1b5b): ad_open: Structure needs cleaning
    dataNEO2 kernel: sd 1:0:0:0: sda tag#2 uas_eh_abort_handler 0 uas-tag 3 inflight: IN
    dataNEO2 kernel: sd 1:0:0:0: sda tag#2 CDB: opcode=0x28 28 00 df 01 09 00 00 00 08 00
    dataNEO2 kernel: scsi host1: uas_eh_bus_reset_handler start
    dataNEO2 kernel: usb 4-1: reset high-speed USB device number 2 using ehci-platform
    dataNEO2 kernel: scsi host1: uas_eh_bus_reset_handler success   ---> Here it comes the reset from UAS

    I have attached an external USB drive (same model, same size) and again the same issue after some time reading files.


    This is the information from the SATA drive



    And the information from the USB ports itself...



    I already tried to disable the UAS feature as per UbuntuForums, StackExchange, ArchiLinux, OpenMediaVault, etc... where they do recommend disabling the UAS feature within the USB ports (usually for USB3 devices, however the behavior is the same):


    • including a blacklist in /etc/modprobe.d/ignore_uas.conf (lsusb output:) with options usb-storage quirks=Vendor_ID:Product_ID:u

    idVendor 0x152d JMicron Technology Corp. / JMicron USA Technology Corp.
    idProduct 0x0578


    Code
    options usb-storage quirks=152d:0578:u


    • including a rule in /etc/udev/rules.d/60-usb-hdd.rule for disabling SMART
    Code
    KERNEL=="sd*[!0-9]", ATTR{removable}=="0", ENV{ID_BUS}=="usb", ENV{DEVTYPE}=="disk", ENV{ID_ATA_FEATURE_SET_SMART}="0", ENV{ID_ATA_FEATURE_SET_SMART_ENABLED}="0", ENV{UDISKS_DISABLE_POLLING}="1"
    KERNEL=="sd*[!0-9]", ATTR{removable}=="0", ENV{ID_BUS}=="ata", ENV{DEVTYPE}=="disk", ENV{ID_ATA_FEATURE_SET_SMART}="0", ENV{ID_ATA_FEATURE_SET_SMART_ENABLED}="0", ENV{UDISKS_DISABLE_POLLING}="1"
    KERNEL=="sd*[!0-9]", ATTR{removable}=="0", ENV{ID_BUS}=="scsi", ENV{DEVTYPE}=="disk", ENV{ID_ATA_FEATURE_SET_SMART}="0", ENV{ID_ATA_FEATURE_SET_SMART_ENABLED}="0", ENV{UDISKS_DISABLE_POLLING}="1"


    • Both together disabling auto suspend and quirks


    Code
    # echo "152d:0578:u"  >> /sys/module/usb_storage/parameters/quirks
    # echo -1 >/sys/module/usbcore/parameters/autosuspend



    But I'm not able to remove the USB resets when reading huge amount of small files (I have been not able to reproduce it with huge size files) and actually, I don't think the UAS is disabled as per dmesg output once reboot.



    I'm wondering whether this is a Kernel issue or a HW issue. Any idea in how to remove that annoying situation?, I'm loosing all data within the Filesystem due to I have to fsck every time the partition, and I'm loosing almost all data inside.

  • chente

    Hat das Thema geschlossen.

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!