Hi all, FreeNAS 9.2.1.7 here. I have a 4-disk raidz1 that recently experienced a drive failure. Because this box has only 4 drive bays, in-place replacement is my only option. It's AHCI so I confirmed that the drive was showing as "FAULTED" and then pulled it and replaced it with a brand new disk. Per the docs, this is the right thing to do. However, in the GUI, when I select the disk and click the "Replace" button, the dropdown for "Member disk" is blank and empty. The disk shows up just fine under "View disks" with the expected device name (ada0) and serial number (not the old disk's serial). "camcontrol devlist" on the CLI shows the new disk and not the old disk.
The GUI volume status shows ada1p2 through ada3p2, all with status ONLINE, and gptid/bc23... (was ada0p2) as FAULTED.
I did my best to follow the docs, but the manual seems to assume that you're not doing an in-place replacement. So I definitely had to guess in certain sections. In particular, before I pulled the bad disk, if I clicked on the "Replace" button it would give me one option in the "Member disk" list: "In-place [ada0]". I always canceled out because I was pretty sure you were supposed to pull the bad disk before replacing it. But the fact that I had the exact option I wanted before, and do not have that option any more, suggests to me that maybe I did it wrong. But like I said, this specific process seems to be entirely undocumented. Even Google can't tell me how to do it with the GUI. (Plenty of hits telling me how to do it with the CLI, which is how I replaced the last failed disk. But I want to do it "correctly.")
Code:
[root@freenas] ~# zpool status -v pool: data0 state: DEGRADED status: One or more devices are faulted in response to persistent errors. Sufficient replicas exist for the pool to continue functioning in a degraded state. action: Replace the faulted device, or use 'zpool clear' to mark the device repaired. scan: scrub repaired 44K in 4h36m with 0 errors on Sun May 10 04:36:51 2015 config: NAME STATE READ WRITE CKSUM data0 DEGRADED 0 0 0 raidz1-0 DEGRADED 0 0 0 gptid/bc23ea6b-388f-11e4-bfe3-000c29f8a1e5 FAULTED 1 155 0 too many errors gptid/bc8539c2-388f-11e4-bfe3-000c29f8a1e5 ONLINE 0 0 0 gptid/bce11e3c-388f-11e4-bfe3-000c29f8a1e5 ONLINE 0 0 0 gptid/cefc9e61-d932-11e4-b1a8-000c29f8a1e5 ONLINE 0 0 0 errors: No known data errors
The GUI volume status shows ada1p2 through ada3p2, all with status ONLINE, and gptid/bc23... (was ada0p2) as FAULTED.
I did my best to follow the docs, but the manual seems to assume that you're not doing an in-place replacement. So I definitely had to guess in certain sections. In particular, before I pulled the bad disk, if I clicked on the "Replace" button it would give me one option in the "Member disk" list: "In-place [ada0]". I always canceled out because I was pretty sure you were supposed to pull the bad disk before replacing it. But the fact that I had the exact option I wanted before, and do not have that option any more, suggests to me that maybe I did it wrong. But like I said, this specific process seems to be entirely undocumented. Even Google can't tell me how to do it with the GUI. (Plenty of hits telling me how to do it with the CLI, which is how I replaced the last failed disk. But I want to do it "correctly.")