3

I noticed the following messages on my system:

Personalities : [raid1] [raid6] [raid5] [raid4] 
Md3 : active raid5 sda5[0] sdc5[3] sdb5[1]
      3858519040 blocks super 1.2 level 5, 512k chunk, algorithm 2 [3/3] [UUU]
      [=================>...]  check = 86.0% (1659909504/1929259520) finish=182.1min speed=24645K/sec

What does check status mean and why did this happen?!

Improve this question

1 Answer 1

Reset to default
6

When this happens one of the members in the RAID array has likely gone out of sync with the array, this is the resyncing check that's occurring and it's 86% complete.

When the RAID array is in a healthy state the output will look similar to this:

$ sudo cat /proc/mdstat 
Personalities : [raid1] 
md0 : active raid1 sdc1[0] sdb1[1]
      2930266432 blocks [2/2] [UU]

unused devices: <none>

Why did this happen?

I've had these happen from time to time when one of the members gets out of sync. However if they continue to occur then it's likely a good indication that one of the members is starting to fail, and so you should heed the warning and fully diagnose all the members to make sure they're in good health and also identify which HDD (member) is failing and plan to replace it sooner rather than later.

Some distributions may perform a check periodically so you might be getting your array in this state due to these scheduled checks. I believe Debian/Ubuntu may perform a check like this weekly, look for a entry in your system's crontab files or /etc/cron.d directories.

Lastly you can manually trigger a check such as this by echoing the command "check" to the RAID's sync_action file under /dev.

Example

As root:

# echo check > /sys/block/md0/sync_action

If you need to pause an on going check then you can echo the command "idle" to stop it.

# echo idle > /sys/block/md0/sync_action

References

Improve this answer
3
  • thanks for your explanation slm can a lot of write / read operations make this happen?! like under http flood attack which may make a lot of write / read operations? thanks again
    – user205537
    Feb 8, 2014 at 20:07
  • Sure any additional wear and tear on the RAID could've led to this happening too.
    – slm
    Feb 8, 2014 at 20:10
  • 1
    On Ubuntu (for sure 16.04, probably others too), this is simply a cronjob in /etc/cron.d/mdadm that runs /usr/share/mdadm/checkarray with cron schedule 57 0 * * 0.
    – nh2
    Nov 6, 2016 at 15:16

You must log in to answer this question.

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