Hi everyone,
I've got a 1002T NAS running 2 WD 3TB Reds in Raid1. Disk 2 was reporting bad block errors so I pulled the disk out without doing anything to the volume. I hooked it up in my desktop, ran the WD utility which detected and fixed the errors and a subsequent run reported no errors. I didn't Erase or Format the disk.
How should I go about adding this drive back to the NAS? The management button is disabled for the volume. My concern is that if I add the drive back in, it will mirror the bad data from the repaired disk2 to the good disk 1. Does ADM figure this out? Should I use the WD Utility and Erase disk2 to completely rebuild the raid?
Thanks
Replacing problematic drive in Raid1
Moderator: Lillian.W@AST
-
- Posts: 3
- youtube meble na wymiar Warszawa
- Joined: Tue May 22, 2018 10:16 pm
- orion
- Posts: 3482
- Joined: Wed May 29, 2013 11:09 am
Re: Replacing problematic drive in Raid1
Simply put disk-2 back to NAS WHEN NAS IS WORKING. It should re-build the RAID volume (sync data to disk-2).
-
- Posts: 82
- Joined: Tue Jun 27, 2017 1:05 pm
Re: Replacing problematic drive in Raid1
Man, how can you trust this HDD again in a 24/7 situation? I would get another to add to the array and then use the failed HDD in an enclosure for external backups or other low uptime tasks and such.
BTW, if both HDDs were bought together, that is same production batch, I would also look carefully at the other, supposedly healthy one.
How old were these HDDs?
BTW, if both HDDs were bought together, that is same production batch, I would also look carefully at the other, supposedly healthy one.
How old were these HDDs?
-
- Posts: 3
- Joined: Tue May 22, 2018 10:16 pm
Re: Replacing problematic drive in Raid1
They were bought from different sources about 1.5 years ago. I ran tests on disk1 as well but nothing showed up. The errors were only coming from disk2. I was going to RMA it but the WD utility didn't show any issues after fixing and then running the deep scan again. So I'll give it another shot before RMA.Antonios wrote:Man, how can you trust this HDD again in a 24/7 situation? I would get another to add to the array and then use the failed HDD in an enclosure for external backups or other low uptime tasks and such.
BTW, if both HDDs were bought together, that is same production batch, I would also look carefully at the other, supposedly healthy one.
How old were these HDDs?
-
- Posts: 82
- Joined: Tue Jun 27, 2017 1:05 pm
Re: Replacing problematic drive in Raid1
Well, it's magnetic surface defects that's causing the errors, what the utility does is remap those sectors to spare ones, SMART data should indicate that, if compared between both identical HDDs.
Point is, HDD is developing surface defects, you may remap these first bad sectors to spare sectors, this does not make for a healthy magnetic surface at all, imho this HDD's surface is done and can't be trusted to hold data reliably for any length of time.
Just my 2c.
Point is, HDD is developing surface defects, you may remap these first bad sectors to spare sectors, this does not make for a healthy magnetic surface at all, imho this HDD's surface is done and can't be trusted to hold data reliably for any length of time.
Just my 2c.
-
- Posts: 3
- Joined: Tue May 22, 2018 10:16 pm
Re: Replacing problematic drive in Raid1
Thanks, appreciate the response. I'll keep monitoring and continue the regular scheduled tests to see what comes up again. Still have lots of time before the warranty expires.Antonios wrote:Well, it's magnetic surface defects that's causing the errors, what the utility does is remap those sectors to spare ones, SMART data should indicate that, if compared between both identical HDDs.
Point is, HDD is developing surface defects, you may remap these first bad sectors to spare sectors, this does not make for a healthy magnetic surface at all, imho this HDD's surface is done and can't be trusted to hold data reliably for any length of time.
Just my 2c.
I hot-swapped the drive back in, it picked it up and completed the rebuild. Thanks.orion wrote:Simply put disk-2 back to NAS WHEN NAS IS WORKING. It should re-build the RAID volume (sync data to disk-2).