2

I have a RAID 1 on my Raspi and got mail I don't understand titled "DegradedArray event on /dev/md0:my-host-name" and "Fail event on /dev/md0:my-host-name". I got 6 messages with the former subject and 2 with the latter.

The first kind of mails look like this:

This is an automatically generated mail message from mdadm running on my-host-name

A DegradedArray event had been detected on md device /dev/md0.

Faithfully yours, etc.

P.S. The /proc/mdstat file currently contains the following:

Personalities : [raid1] md0 : active raid1 sda1[0] 124967936 blocks super 1.2 [2/1] [U_] bitmap: 1/1 pages [4KB], 65536KB chunk

unused devices:

The second kind of mails like this:

This is an automatically generated mail message from mdadm running on my-host-name

A Fail event had been detected on md device /dev/md0.

It could be related to component device /dev/sdb1.

Faithfully yours, etc.

P.S. The /proc/mdstat file currently contains the following:

Personalities : [raid1] md0 : active raid1 sdb11(F) sda1[0] 124967936 blocks super 1.2 [2/1] [U_] bitmap: 0/1 pages [0KB], 65536KB chunk

unused devices:

I restarted my Raspi before learning about these messages. This is the current RAID status:

Personalities : [raid1] md0 : active raid1 sdb11 sda1[0] 124967936 blocks super 1.2 [2/2] [UU] bitmap: 0/1 pages [0KB], 65536KB chunk

unused devices:

I did not touch the Raspi or its drives when the event happened. However, I cannot exclude that someone else did and possibly loosened the connection of one of the data cables.

Did this happen? Did something else happen? Did the RAID restore itself to normal operation or do I have to do something?

Possibly related: Meaning of Security Information Mail

3

The first message means that your RAID array went into an inconsistent state, because apparently the sdb drive was detected as failing (second message). The current status looks like the array was restored, but you may want to check the output of smartctl --all /dev/sdb to get the current health status (which is printed before the drive parameters and the error log) and check if the drive parameters are suspicious (things like Reallocated Sector Count or Current Pending Sector hint to a potential problem) or if the device has (new) entries in the error log. You may also want to check dmesg for messages related to sdb.

You could also, for extra safety, remove sdb1 from the RAID array and execute a test with smartctl on it (eg. smartctl -t short /dev/sdb for a short test or smartctl -t long /dev/sdb for a more thorough test).

Please note that you need to use -d <...> for smartctl with a parameter <...> that fits your device. Refer to this list of supported USB devices for the correct one. To get the USB IDs, you can use lsusb. If your device is not listed, you may look for related devices (eg. by the same vendor or having a similar name).

  • There seems to be a problem with the commands. Unfortunately, I can't find you in chat. Can we talk about it there to avoid many comments? More specifically, I get the error "/dev/sda: Unknown USB bridge [0x05dc:0xa838 (0x1100)]". However, I have a desktop environment installed on my Raspi and can read the smart data via gnome-disks, in case your first command is just to read the smart data. – UTF-8 Jan 9 '17 at 19:12
  • Yes, all smartctl commands are intended to either read the parameters or instruct the drive to execute a test. The error message indicates that you are required to use the -d switch, although your device is not on the supported devices list. Guessing from the ID, the device in question is a flash drive, for which the smartctl commands may not be that useful anyway. – Abrixas2 Jan 9 '17 at 19:31
  • /dev/sda and /dev/sdb sometimes switch after a reboot and because I rebooted my Raspi since the mail was created, I don't know which drive was /dev/sda and which was /dev/sdb at the time. One is a thumb drive, the other one an external HDD. The thumb drive doesn't have SMART data (according to gnome-disks) and this is the SMART data gnome-disks shows for the HDD: screenshot 1, screenshot 2. Both Reallocated Sector Count and Current Pending Sector Count seem to be 0. – UTF-8 Jan 9 '17 at 19:36
  • The values seem to be okay, the only thing I see is the UDMA CRC Error Rate, which hints to the possibility of a loose or otherwise impaired connection. OTOH, the raw value is one, making this rather unlikely IMHO. You could add another device (eg. another external HDD or another thumb drive) to the array to compensate a failing drive (either by putting it directly into md0 or by defining it as a sparse drive IIRC). – Abrixas2 Jan 9 '17 at 19:51
  • The rather high value of Hardware ECC Recovered may theoretically point to a problem with the surface of the drive, but may also be normal for high capacity drives. Since all other related values (read errors, write errors) are zero, I would not assume the value pointing out a problem. – Abrixas2 Jan 9 '17 at 19:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.