September 20, 20169 yr unRAID OS Version: I've had this issues for a while, finally getting around to rebooting, testing safe mode, etc. I am now on 6.2 final. Description: If a spare USB flash drive is in the system when the array starts, the following error is printed to the console: SG_IO: bad/missing sense data, sb[]: f0 00 05 00 00 00 00 14 00 00 00 00 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Visiting the main tab, particularly the unassigned devices tab, causes the error to be spammed to the console. The error is not in my syslog, just the console. It happens whether the UD plugin is installed or not, and even in safe mode. How to reproduce: Insert a USB flash drive, start the array, visit the unassigned devices tab Expected results: No errors spamming the console Actual results: Lots of SG_IO errors spamming the console Other information: Here is the output of the smartctl and hdparm commands for /dev/sdg: root@Tower:~# smartctl -a /dev/sdg smartctl 6.5 2016-05-07 r4318 [x86_64-linux-4.4.19-unRAID] (local build) Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org /dev/sdg: Unknown USB bridge [0x0781:0x5571 (0x127)] Please specify device type with the -d option. Use smartctl -h to get a usage summary root@Tower:~# hdparm -I /dev/sdg /dev/sdg: SG_IO: bad/missing sense data, sb[]: f0 00 05 00 00 00 00 14 00 00 00 00 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ATA device, with non-removable media Serial Number: ???T@?^?? Firmware Revision: ?? Standards: Likely used: 1 Configuration: Logical max current cylinders 0 0 heads 0 0 sectors/track 510 0 -- Logical/Physical Sector size: 512 bytes device size with M = 1024*1024: 0 MBytes device size with M = 1000*1000: 0 MBytes cache/buffer size = unknown Capabilities: IORDY not likely Buffer type: 04c0: dual port, multi-sector with read caching ability Buffer size: 2029.5kB bytes avail on r/w long: 59904 Cannot perform double-word IO R/W multiple sector transfer: not supported DMA: not supported PIO: pio0 root@Tower:~# I'm guessing the main tab is running hdparm on all the drives. Can it skip flash drives? Or find some other way to suppress this error message on the console? I'll also mention that when viewing the details for this drive: http://tower/Main/New?name=sdg there is an option to "Spin Up" the flash drive, which doesn't do much other than spam the console some more. And the attributes, capabilities and identity tab are essentially empty. tower-diagnostics-20160919-2009.zip Edited January 3, 20188 yr by ljm42
September 20, 20169 yr This is a known issue (some USB drives need an additional parameter when reading the SMART information). The easy workaround is to connect your USB drive after the array is started.
September 20, 20169 yr Author Thanks bonienl, is there a reasonable way to fix this behavior? It seems somewhat related to this issue: https://lime-technology.com/forum/index.php?topic=51814.msg497382#msg497382 where the /config/smart-one.cfg needed an extra param for a specific disk. Would something like that help? Unfortunately, the gui doesn't show SMART settings for flash drives the way it does for hard drives, so I wasn't sure how to do it. Or maybe unraid could simply look at the results of the first hdparm call, and if it is unsuccessful, just stop calling it for that disk? My issue is that I leave the second flash drive plugged in all the time so I can run a separate install of unRAID in a VM for testing purposes: https://lime-technology.com/forum/index.php?topic=44681.msg482553#msg482553 Ironically, this error spamming wasn't an issue when I was stubbing out my usb controller, but as of 6.2.0-rc1 and the unraidlabel parameter, the need to stub the usb controller was removed. But apparently the trade off is that the console is spammed with spurious error messages
September 29, 20169 yr Author Interesting... I re-installed Unassigned Devices and most of the "SG_IO: bad/missing sense data" errors went away! Absolutely none are generated by the Main -> UD tab anymore. The only only time I still see the error is when I view the Dashboard. Does this give any hints on preventing the error?
December 2, 20169 yr Author Is there a reasonable way to fix this on the Dashboard? Nearly every time I check the console it is filled with these errors, completely masking any real errors that might be there.
December 2, 20169 yr Yes this is a known issue. Spamming the console, though not ideal, is harmless to the running system, but yeah needs to be fixed.
September 13, 20178 yr Author It looks like there have been some changes! Following these directions on an unRAID 6.3.5 host, I do not see any more of the SG_IO errors. woot! However, once I start the VM, in their place I see this error on the host's console instead: /dev/sdg: No such file or directory It is repeated on a regular basis whenever I have the host's dashboard open. /dev/sdg is the USB flash drive that is passed to the VM.
January 3, 20188 yr Author With the host at 6.4.0_rc18f I no longer see any of the messages reported here. Thanks!
Archived
This topic is now archived and is closed to further replies.