HelloFreeNAS
Dabbler
- Joined
- Sep 15, 2013
- Messages
- 18
Just under a week ago I was trying to access my FreeNAS web gui but was receiving some sort of nginx error. I hadn't set up SSH and couldn't figure out a way to safely shut down my server. I tried orderly shutdown in asrock IPMI and pressing the physical power button on the case (which usually initiates a safe shutdown), neither of which worked.
I ended up having to do a force shutdown through IPMI.
When I turned the server back on I could access the web GUI just fine but I had a couple alarming alerts:
"CRITICAL: 4408 unreadable sectors", "CRITICAL: one or more devices has experienced an unrecoverable error"
Throughout the past 5ish days after turning the server back on in the morning these alerts have showed to up to 6056 unreadable sectors and sometimes "CRITICAL: one or more devices has experienced an unrecoverable error" is there and sometimes it isn't.
The mirror had also been changed to a degraded state at one point.
I just updated to 9.10 stable and it is currently back to only 4408 unreadable sectors and "CRITICAL: one or more devices has experienced an unrecoverable error" has disappeared again and the mirror is in an online state.
What I'm unsure about is whether these notifications are related to some sort of drive errors which resulted in the nginx errors or if the nginx error was just a bug and cutting the power started all this.
My drives are a WD Barracuda 3TB (the one with errors) and WD Red 3TB in a mirror
2x SanDisk 8GB Cruzer Fit for my boot sticks, mirrored
I also have 8GB of ECC RAM.
1. What can I do to test the drive and see whether it's a physical issue and the drive needs replacing?
2. If it is just related to the unsafe shutdown, what can I do to fix these errors?
Thanks
edit: gonna have a look at this:
https://forums.freenas.org/index.php?threads/how-to-hard-drive-burn-in-testing.21451/
ill update if I figure anything out
I ended up having to do a force shutdown through IPMI.
When I turned the server back on I could access the web GUI just fine but I had a couple alarming alerts:
"CRITICAL: 4408 unreadable sectors", "CRITICAL: one or more devices has experienced an unrecoverable error"
Throughout the past 5ish days after turning the server back on in the morning these alerts have showed to up to 6056 unreadable sectors and sometimes "CRITICAL: one or more devices has experienced an unrecoverable error" is there and sometimes it isn't.
The mirror had also been changed to a degraded state at one point.
I just updated to 9.10 stable and it is currently back to only 4408 unreadable sectors and "CRITICAL: one or more devices has experienced an unrecoverable error" has disappeared again and the mirror is in an online state.
What I'm unsure about is whether these notifications are related to some sort of drive errors which resulted in the nginx errors or if the nginx error was just a bug and cutting the power started all this.
My drives are a WD Barracuda 3TB (the one with errors) and WD Red 3TB in a mirror
2x SanDisk 8GB Cruzer Fit for my boot sticks, mirrored
I also have 8GB of ECC RAM.
1. What can I do to test the drive and see whether it's a physical issue and the drive needs replacing?
2. If it is just related to the unsafe shutdown, what can I do to fix these errors?
Thanks
edit: gonna have a look at this:
https://forums.freenas.org/index.php?threads/how-to-hard-drive-burn-in-testing.21451/
ill update if I figure anything out
Last edited: