"SMART error (FailedHealthCheck) detected on host"

    "I'm getting this emailed alert:

    In the webgui, under Storage > S.M.A.R.T. > Devices, dev/sdc has a red dot in its Status column.


    The Information > Attributes tab for that device shows "Raw_Read_Error_Rate," and "Seek_Error_Rate," and "Spin_Retry_Count" all with green Status dots, "Reallocated_Sector_Ct" with a red status dot, and all of the other attributes are grayed out.


    So it appears that at least some SMART attributes are being monitored (should I be worried about the Reallocated_Sector_Ct status?).


    The Self-Test Logs show only "Completed without error" statuses. I can't read all the data in the log because there doesn't appear to be a way to make some of the columns labeled "Rem..." and "Life..." wide enough to read the full strings.


    So . . . what's going on? Is SMART working or not? The Extended Information says "SMART support is: Available - device has SMART capability.
    SMART support is: Enabled."


    So I don't know why I am getting an email saying the drive doesn't support SMART, and I am wondering whether the drive is failing.


    Any insight would be appreciated! Thank you.

    I read 'serverpi' (Pi as in Raspberry Pi?), then read SAT (SCSI ATA translation which means we're talking about USB attached storage), then read sdc (which implies there's also sda and sdb). The above combination would be scary.


    In case you're able to login to the OMV installation with SSH you should try to provide the output from 'armbianmonitor -u' and 'for i in /dev/sd? ; do smartctl -d sat -a $i ; done'

    Indeed, my apologies. Yes, a Raspberry Pi; yes, we are talking about USB-attached drives; and yes there exists sda and sdb. Scary that I didn't think to provide that information, or a scary configuration?


    Anyway, thank you for persevering. In response to your suggestions:


    Code
    $ sudo armbianmonitor -u                                                                                                                    
[sudo] password for jack:                                                                                                                   
/var/log/armhwinfo.log has been uploaded to http://sprunge.us/cCKe                                                                          
Please post the URL in the Armbian forum where you've been asked for.


    For your suggested smartctl loop, the forum isn't letting me post the output, "the message is too long, must be under 10,000 characters". This from posting the output into a Code box. Can I p-mail the output to you?



    PS. Apology for the double-post of the original post. I received a "fatal error" message in the browser from the forum when I first posted the topic. And it didn't appear, so I re-posted. Received the fatal error msg again, shrugged and crossed fingers.

    Zitat von JackElliott

    scary configuration?

    This :)


    And you're in trouble, please check the generated log yourself: http://sprunge.us/cCKe (messages wrt sda and sdc). I'm a big fan of energy efficient computing and also NAS operation and based on my personal experience with a lot of ARM devices I consider the Raspberries being the worst choice possible since prone to underpowering troubles and both network and storage behind one single USB2 connection (there's an internal USB hub on the board also containing the USB network adapter).


    And also based on a lot of testing I personally wouldn't use any USB disk enclosure any more that's not based on either JMS567, JMS578 or ASM1153E/ASM1053E. Those older USB-to-SATA bridges in your disk enclosures based on SunPlus or old ASMedia chipsets are the culprit why SMART throws errors (the chipsets have to support SAT -- SCSI / ATA translation -- correctly and only a few and more modern do).

    Addendum: VIA VL715/VL716 are also ok (we got a report about those chipsets being broken wrt 'USB Attached SCSI' but this was as so often simply related to underpowering problems -- maybe problem N°1 in the SBC world)


    And just as some references to study:

    Zitat von JackElliott

    all three of the USB HDDs are self-powered

    I know since the log output also provides your boot environment and max_usb_current=1 is not set (which is basic requirement to be able to power an external HDD somewhat but still more or less inreliably from a Raspberry).


    I'll only repeat what I've written above: the dmesg output collected in the logs indicates that you're in trouble. And the combination of RPi + those outdated USB-to-SATA bridges (not fully supporting SMART for example) is nothing I would ever recommend.

Jetzt mitmachen!

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

Benutzerkonto erstellen Anmelden