RAID 1 Hijinks

[Lord Nicoll]

HA, what are the chances the 2nd HDD I bought for data redundancy (purchases a month later) would be the one to fail. Although RAID 1 mirrors both drives and they server as each others redundancy, I still consider the drive in bay 2 to be the redundancy and guess what one failed? I'm thinking manufacturing fuck up, I even got low profile rubber isolation mounts for it as server racks can be unforgiving for HDDs. The backing up the files and reinstalling in RAID 1 wasn't that annoying, Linux makes it easy, but it sure feels like a waste now.

[Kobathor]

They're both for redundancy- one can fail and all of your data is still there. It would be the exact same outcome if Drive 1 were to have failed. If you didn't have that redundant storage, Drive 1 might have failed and you would've lost all of your data.
 

Just buy another drive and redo the RAID. If you don't care about redundancy anymore, don't replace the drive. It's your choice. If your drive is still under warranty, you could probably send it in and be sent a fresh one- the drive failing is probably a manufacturer defect if it died during the warranty period.

[Lord Nicoll]

1 minute ago, Kobathor said:

They're both for redundancy- one can fail and all of your data is still there. It would be the exact same outcome if Drive 1 were to have failed. If you didn't have that redundant storage, Drive 1 might have failed and you would've lost all of your data.
 

Just buy another drive and redo the RAID. If you don't care about redundancy anymore, don't replace the drive. It's your choice. If your drive is still under warranty, you could probably send it in and be sent a fresh one- the drive failing is probably a manufacturer defect if it died during the warranty period.

It's a new SAS drive, it's going back to them for a replacement. Yeah, I did point that out, both drives are for redundancy to each other, but the drive that was bought to make the data safe was the one that failed, and in only 26 days, the drive has roughly  468 hours of use, which is well below what I expect. The server is remote so slow right speeds might have been there as an indication of a bad drive and I wouldn't have noticed. 

[Kobathor]

Just now, Lord Nicoll said:

It's a new SAS drive, it's going back to them for a replacement. Yeah, I did point that out, both drives are for redundancy to each other, but the drive that was bought to make the data safe was the one that failed, and in only 26 days, the drive has roughly  468 hours of use, which is well below what I expect. The server is remote so slow right speeds might have been there as an indication of a bad drive and I wouldn't have noticed. 

Only 26 days and it failed? Dang! They'd better send you a new one.

And yeah- good thing you noticed. 

[Lord Nicoll]

1 minute ago, Kobathor said:

Only 26 days and it failed? Dang! They'd better send you a new one.

And yeah- good thing you noticed. 

Well when I booted it again (I only run it 18 hours a day) the LED indicators listed a Red fatal error, after hooking up a monitor, the HP Smart Array failed to initialize due to a physical HDD failure which resulted in a logical drive failure as it knew something was wrong. The HDD smells like burned electronics, so I guess the PCB done fucked the goose. The other drive is a different batch (always a good idea to use different batches of HDD) but it was probably a bad solder joint, but this could initiate a recall, but I doubt it, out of 4 drives (two servers) I've had no problems so far.