Once in a while, there’s a disk in a RAID array that we believe to be “suspicious” and would like to mark as a failed drive. But using Areca’s web interface, there’s no obvious possibility to do so.
You may have already guessed: According to Arca’s FAQ section, there’s a hidden way to do so:
We have added the “FailDisk” hidden on the “Rescue Raid Set” function. The “FailDisk” function can kill off the member disk of RAID set that you enter on the Rescue Raid Set. Please use the “FailDisk Device Location” on the Rescue Raid Set. You can get the correctly “Device Location” from the area “Device Location” of the “Device Information”. Such as, you want to kill off the “Enclosure#2 SLOT 01″ HDD. You can get the right disk channel from the Device-Information and enter ” FailDisk Enclosure#2 SLOT 01″.
But when you look at the web interface of the Areca ARC-1261ML… where’s the “Device Information”, to look up the “Device Location”?
Let me translate that FAQ answer to the ARC-1261ML: You simply make use of the channel identification, as found on the “RaidSet hierarchy” page in the “IDE Channels” column and duplicated in the “Channel” column of the “IDE Channels” listing:
| Raid Set | IDE Channels | Volume Set(Ch/Id/Lun) | Volume State | Capacity |
|---|---|---|---|---|
| Raid6-0000 | Ch01 | RAID6-0000v0000 (0/0/0) | Normal | 45000GB |
| Ch02 | ||||
| … | ||||
IDE Channels
| Channel | Usage | Capacity | Model |
|---|---|---|---|
| Ch01 | Raid6-0000 | 15000GB | MySuperDuperFastDisk |
| Ch02 | Raid6-0000 | 15000GB | MySuperDuperFastDisk |
| Ch03 | Raid6-0000 | 15000GB | MySuperDuperFastDisk |
| … | |||
| Ch16 | N.A. | N.A. | N.A. |
So all you have to do to mark i.e. disk #8 as "failing", is to enter "DiskFail Ch8" as the "keyword" in "RaidSet Functions" – "Rescue Raidset", check the "Confirm Operation" box and submit the form.
Invoking the same function via the Areca CLI failed for me: According to the help message, it should be invoked as “disk fail drv=8”, but unfortunately the only response I receive is “GuiErrMsg: Invalid Parameter.”, even after setting the admin password first. (Of course I tried “drv=Ch08” as well, which gave a “ErrMsg: Drive# Is Missing.” response.)
Be careful which this command, you get what you ask for: The controller will mark the drive as “failed” and start to recover the RAID set. Not only may this take a long time (and degrading your RAID’s performance during rebuild) – it may be dangerous, too. So use this at your own risk, don’t toy with this in a productive environment!