Recurring storage failues

  • Please bear with me, I know my setup is far from ideal but I'm saving up for a proper hardware solution.


    I'm running a RPi with 4 external USB HDD's through a USB hub connected to one of the USB3 ports on the RPi.


    This seems to work ok in general, but every day at some point the NFS shares just stop working. I can unmount and remount them fine, but it keeps happening.


    This is the journal logs from when the failures start:


    Code
    Aug 25 21:49:02 nas monit[1333]: 'mountpoint_srv_dev-disk-by-uuid-ecae3a44-f697-44e4-873b-9734bd297ce6' status failed (1) -- /srv/dev-disk-by-uuid-ecae3a44-f697-44e4-873b-9734bd297ce6 is not a mountpoint
    Aug 25 21:49:02 nas monit[1333]: 'mountpoint_srv_dev-disk-by-uuid-ecae3a44-f697-44e4-873b-9734bd297ce6' status failed (1) -- /srv/dev-disk-by-uuid-ecae3a44-f697-44e4-873b-9734bd297ce6 is not a mountpoint
    Aug 25 21:49:02 nas monit[1333]: Filesystem '/srv/dev-disk-by-uuid-63AD-7E9E' not mounted
    Aug 25 21:49:02 nas monit[1333]: 'filesystem_srv_dev-disk-by-uuid-63AD-7E9E' unable to read filesystem '/srv/dev-disk-by-uuid-63AD-7E9E' state
    Aug 25 21:49:02 nas monit[1333]: 'filesystem_srv_dev-disk-by-uuid-63AD-7E9E' trying to restart
    Aug 25 21:49:02 nas monit[1333]: 'mountpoint_srv_dev-disk-by-uuid-63AD-7E9E' status failed (1) -- /srv/dev-disk-by-uuid-63AD-7E9E is not a mountpoint
    Aug 25 21:49:02 nas monit[1333]: 'mountpoint_srv_dev-disk-by-uuid-63AD-7E9E' status failed (1) -- /srv/dev-disk-by-uuid-63AD-7E9E is not a mountpoint
    Aug 25 21:49:21 nas kernel: EXT4-fs error (device sda1): __ext4_find_entry:1614: inode #53739528: comm nfsd: reading directory lblock 0


    Is there anything obvious that can be done about this to make it more reliable? I'm not exactly sure why this is happening.

  • external USB HDD's

    Some portable HDDs go into sleep mode after some time of inactivety.

    This is something that Linux can't deal since most of those drives are built for Windows.

    One possible hack is to keep it alive by making a touch /path-to-drive/awake.txt scheduled job every x minutes so it is always spinning.

  • Some portable HDDs go into sleep mode after some time of inactivety.

    This is something that Linux can't deal since most of those drives are built for Windows.

    One possible hack is to keep it alive by making a touch /path-to-drive/awake.txt scheduled job every x minutes so it is always spinning.

    Thanks, will give this a try :thumbup:

Jetzt mitmachen!

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