USB 3.0 外置硬盘盒在Debian 7上使用，经常丢失硬盘

consatan

系统 Debian GNU/Linux testing (jessie) AMD 64
主板 Gigabyte Q87M-D2H, BIOS F6
外置硬盘盒 ORICO 8628sus3-c（有外置电源插头）
硬盘 Toshiba 3T DT01ACA300 * 2，GPT，每个硬盘单分区EXT4

启动的时候2个硬盘分别是/dev/sdf和/dev/sdg，分别挂载到/media/d1/ /media/d2/上
拷贝数据的时候，可能拷贝到300G或更多点的时候，突然I/O错误了，parted -l看了一下，2个硬盘分别变成是/dev/sdh和/dev/sdi了...
卸载再重新挂载，然后再接着拷贝，拷贝一点数据后，又开始I/O错误了，parted -l，看不到了，只能把移动硬盘盒给拔掉电源，或者重启计算机才能再使用

以下是部分日志
kernel: [96724.528852] EXT4-fs (sdf1): recovery complete
kernel: [96724.535622] EXT4-fs (sdf1): mounted filesystem with ordered data mode. Opts: (null)
kernel: [96731.835544] EXT4-fs (sdg1): recovery complete
kernel: [96731.835755] EXT4-fs (sdg1): mounted filesystem with ordered data mode. Opts: (null)
kernel: [98317.159580] usb 2-6: Disable of device-initiated U1 failed.
kernel: [98322.161717] usb 2-6: Disable of device-initiated U2 failed.
kernel: [98322.312170] Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK
kernel: [98322.312171] Buffer I/O error on device sdg1, logical block 680086147
kernel: [98322.312171] sd 6:0:0:1: [sdg] CDB:
kernel: [98322.312172] Buffer I/O error on device sdg1, logical block 680086148
kernel: [98322.312198] sd 6:0:0:1: [sdg] Unhandled error code
kernel: [98322.312199] Buffer I/O error on device sdg1, logical block 680086177
kernel: [98322.312200] sd 6:0:0:1: [sdg]
kernel: [98322.312200] Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK
kernel: [98322.312201] Buffer I/O error on device sdg1, logical block 680086178
kernel: [98322.312201] sd 6:0:0:1: [sdg] CDB:
kernel: [98324.022455] usb 2-6: new SuperSpeed USB device number 3 using xhci_hcd
kernel: [98324.039337] usb 2-6: New USB device found, idVendor=152d, idProduct=0551
kernel: [98324.039340] usb 2-6: New USB device strings: Mfr=1, Product=2, SerialNumber=5
kernel: [98324.039341] usb 2-6: Product: USB to ATA/ATAPI Bridge
kernel: [98324.039342] usb 2-6: Manufacturer: JMicron
kernel: [98324.039343] usb 2-6: SerialNumber: 433FFFFFFFFF
kernel: [98324.039983] usb-storage 2-6:1.0: USB Mass Storage device detected
kernel: [98324.040036] scsi7 : usb-storage 2-6:1.0
kernel: [98325.039569] scsi 7:0:0:0: Direct-Access     ORIC0    8628US3-C             PQ: 0 ANSI: 5
kernel: [98325.040045] scsi 7:0:0:1: Direct-Access     ORIC0    8628US3-C             PQ: 0 ANSI: 5
kernel: [98325.040753] sd 7:0:0:0: Attached scsi generic sg5 type 0
kernel: [98325.040890] sd 7:0:0:0: [sdh] Very big device. Trying to use READ CAPACITY(16).
kernel: [98325.041152] sd 7:0:0:0: [sdh] Sector size 0 reported, assuming 512.
kernel: [98325.041171] sd 7:0:0:0: [sdh] 5860533168 512-byte logical blocks: (3.00 TB/2.72 TiB)
kernel: [98325.041179] sd 7:0:0:0: [sdh] 0-byte physical blocks
kernel: [98325.041304] sd 7:0:0:1: Attached scsi generic sg6 type 0
kernel: [98325.041533] sd 7:0:0:0: [sdh] Write Protect is off
kernel: [98325.041542] sd 7:0:0:0: [sdh] Mode Sense: 28 00 00 00
kernel: [98325.043164] sd 7:0:0:0: [sdh] No Caching mode page present
kernel: [98325.043356] sd 7:0:0:1: [sdi] Very big device. Trying to use READ CAPACITY(16).
kernel: [98325.045923] sd 7:0:0:1: [sdi] 5860533168 512-byte logical blocks: (3.00 TB/2.72 TiB)
kernel: [98325.046294] sd 7:0:0:1: [sdi] Write Protect is off
kernel: [98325.046301] sd 7:0:0:1: [sdi] Mode Sense: 28 00 00 00
kernel: [98325.046644] sd 7:0:0:1: [sdi] No Caching mode page present
kernel: [98325.046649] sd 7:0:0:1: [sdi] Assuming drive cache: write through
kernel: [98325.047319] sd 7:0:0:1: [sdi] Very big device. Trying to use READ CAPACITY(16).
kernel: [98325.048225] sd 7:0:0:1: [sdi] No Caching mode page present
kernel: [98325.048229] sd 7:0:0:1: [sdi] Assuming drive cache: write through
kernel: [98325.063245] sd 7:0:0:0: [sdh] Assuming drive cache: write through
kernel: [98325.070723] sd 7:0:0:0: [sdh] Very big device. Trying to use READ CAPACITY(16).
kernel: [98325.071051] sd 7:0:0:0: [sdh] Sector size 0 reported, assuming 512.
kernel: [98325.071881] sd 7:0:0:0: [sdh] No Caching mode page present
kernel: [98325.072770] sd 7:0:0:0: [sdh] Assuming drive cache: write through
kernel: [98325.144634]  sdi: sdi1
kernel: [98325.145312] sd 7:0:0:1: [sdi] Very big device. Trying to use READ CAPACITY(16).
kernel: [98325.146204]  sdh: sdh1
kernel: [98325.146479] sd 7:0:0:1: [sdi] No Caching mode page present
kernel: [98325.146981] sd 7:0:0:0: [sdh] Very big device. Trying to use READ CAPACITY(16).
kernel: [98325.147428] sd 7:0:0:1: [sdi] Assuming drive cache: write through
kernel: [98325.148321] sd 7:0:0:1: [sdi] Attached SCSI disk
kernel: [98325.148612] sd 7:0:0:0: [sdh] Sector size 0 reported, assuming 512.
kernel: [98325.149342] sd 7:0:0:0: [sdh] No Caching mode page present
kernel: [98325.150223] sd 7:0:0:0: [sdh] Assuming drive cache: write through
kernel: [98325.151111] sd 7:0:0:0: [sdh] Attached SCSI disk
kernel: [98926.290725] quiet_error: 18 callbacks suppressed
kernel: [98926.290727] Buffer I/O error on device sdf1, logical block 365985792
kernel: [98926.291623] lost page write due to I/O error on sdf1
kernel: [98926.291625] JBD2: Error -5 detected when updating journal superblock for sdf1-8.
kernel: [98927.856073] EXT4-fs error (device sdg1): ext4_put_super:760: Couldn't clean up the journal
kernel: [98940.265762] EXT4-fs (sdh1): recovery complete
kernel: [98940.269293] EXT4-fs (sdh1): mounted filesystem with ordered data mode. Opts: (null)
kernel: [98945.748295] EXT4-fs (sdi1): recovery complete

q1208c

如果你的主板上有 USB 2.0 的接口, 接上去试试看还会不会掉.

consatan

回复 2# q1208c
对不起，最近在忙别的事，忘了上来回复
机器上有USB2.0接口，不过因为机器是在远程的，所以无法马上测试
下次测试看看

q1208c

回复 3# consatan

我以前遇到过类似的问题, 就是 3.0 总掉线. 后来, 接到 2.0 上就没事了. 虽然慢一点.

原因只要你细看那个 USB3.0的 A 端就明白了. 新增的5个点太小, 离边缘太近. 所以, 只要稍稍有人碰一下, 就可以接触不良, 然后表现为掉线, 但 2.0 的 四个点还在, 所以, 有可能被当做2.0的设备.

我个人觉得USB 3.0 的 A端设计是非常失败的. 它就不该用这种方式来兼容2.0
   

consatan

回复 4# q1208c
但我那个很少会有人动到的，不过也有可能是因为硬盘太多，硬盘震动导致的也不一定

如果只能接USB 2.0的话，那我就得考虑其他方案了，毕竟2.0的速度实在太慢了，想想看2个3T的硬盘用2.0得拷贝多久啊...
2 * 2.7T * 1024 * 1024 / 35M / 3600s  ~= 45小时....还是速度恒定在35MB/s的前提下

q1208c

回复 5# consatan

我不是太清楚你的具体需求.

不过, USB 这种, 一般都是用来做备份(一天一次最多了). 拿来一直用的话, 不是太合适.

建议考虑 eSATA 吧, 再慢也有 150M 呢.