0

I have Ubuntu 10.04 server running on a desktop at home. I rebooted, and the machine was unreachable after 10 minutes or so. At this point I should have connected a console (monitor using VGA cable). Instead, I powered off the machine (held down the power button for three seconds). This time I did connect a console, and fsck was running. I'm guessing the box was unreachable the first time because fsck was running then too. Anyway, on the next reboot, I couldn't mount a partition on one of my two 1.5TB drives (/dev/sdb1).

fsck /dev/sdb1 found some errors (short read? something like that) but fsck -Cy /dev/sdb1 fixed them, and I was able to mount and use the partition.

I've since installed smartmontools and configured it to run as a daemon following this guide. If I run sudo service smartmontools restart, the root user gets four emails, with these errors:

Device: /dev/sda, 366 Currently unreadable (pending) sectors
Device: /dev/sdb, 13 Currently unreadable (pending) sectors
Device: /dev/sdb, 28 Offline uncorrectable sectors
Device: /dev/sda, 90 Offline uncorrectable sectors

Both drives (all partitions) otherwise seem fine: I don't see any application errors, strange behvaior, or errors/warnings in /var/log/{syslog,messages}.

So, finally, some questions:

  1. Is there any way to grab SMART logs off the box and analyze them locally using GSmartControl?
  2. Can killing fsck midstream actually trash hardware?
  3. If the drives really are ok, can I squelch/reset the SMART errors somehow?

Thanks, -Adam

UPDATE: The drives are apparently healthy:

# smartctl -H /dev/sda
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

# smartctl -H /dev/sdb
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

1 Answer 1

2

For your first question: Here is a way how I do it with disks on a 3Ware controller:

server $ smartctl -d 3ware,0 -a /dev/twa0 > smart.0
server $ smartctl -d 3ware,1 -a /dev/twa1 > smart.1
server $ smartctl -d 3ware,2 -a /dev/twa2 > smart.2
wks $ scp server:smart.* .
wks $ gsmartcontrol -q --no-scan --add-virtual smart.0 --add-virtual smart.1 --add-virtual smart.2
1
  • Perfect, thanks! That works well for pulling down SMART info. I ended up using smartctl -d ata -a /dev/sda > /tmp/smart.0 and smartctl -d ata -a /dev/sdb > /tmp/smart.1. Analyzed locally. Looked fine, no errors. Dec 3, 2011 at 5:36

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .