Skip to main content

Replace data disk

What happend?

I have a nightly cronjob configured to execute a script which

  • runs a SnapRAID sync and scrub when some requirements are met (e.g. number of deleted files do not exceed a threshold)
  • and sends me an email to notify me about what happend during the script execution

This is how I got notified that the latest SnapRAID sync could not be executed: Some files on one of the data disks could not be read because of an "Unexpected input/output read error".
This did not really surprise me because SMART was reporting the affected disk to have a 100% estimated probability to fail soon for a while now. Additonally, this already happend on this disk before, but I could usually easily fix it by replacing the broken file. Not this time. Now much more files were affected (including files I newly added trying to fix the broken ones as before).

There was no way around it now: I had to replace the disk.

....