First, I'm about to replace a disk (smaller disk with a bigger one). Is it best to shut down the box (to take out a 4TB drive and replace it with an 8TB) or can you simply do a hot swap? Does it matter?
Second, the network has a "dumb" switch so I've always used two ethernet cables to connect the NAS to the network. Is there any advantage in throughput in this set-up or would one cable provide the same throughput as two?
Hot swap is best. You can wait till you're sure the NAS has detected the removal before insertion.
Two Ethenet connections wihtout link agregation is not only not helping, it can cause issues. And if something were to happen to your switch that "blew out" the NAS port, it would do it on both.
Hot swap is best. You can wait till you're sure the NAS has detected the removal before insertion.
I always do hot-swap myself (and recommend it here). The NAS detects the removal and insertion, and that triggers the resync. When you replace a disk with the NAS shut down, then it needs to figure out what you've done. Both methods should work though.
Two Ethenet connections wihtout link agregation is not only not helping, it can cause issues. And if something were to happen to your switch that "blew out" the NAS port, it would do it on both.
Yes. You can use ALB or TLB aggregation with a "dumb" switch, but you shouldn't just connect the two NIC ports. And even ALB/TLB can cause some issues in the clients.
If you are mostly accessing the NAS from one client at a time over gigabit, then there is no speed benefit from connecting both NICs even with link aggregation (since the connection from the switch to the client is still capped at 1 gigabit). There can be some speed up if you have multiple clients that are also connected to the same swtich as the NAS. But it sounds like you aren't doing that.
Thank you for your help. I did a hot swap and overnight the sync was completed. But this morning the NAS is displaying "data DEAD".
It is showing the 4TB drive next to the newly inserted 8TB drive as Dead. OK, I thought, I'll replace that drive with the other 8TB drive I have, but when I inserted that one it is also showing a black circle in the listing. These two 8TB drives have less than 500 hours on them and were working fine in an external backup device.
Is it possible that the 4TB drive that just died and the newly inserted 8TB drive are both bad drives? Should I simply purchase another 8TB drive and expect that to work?
Is it possible that the 4TB drive that just died and the newly inserted 8TB drive are both bad drives? Should I simply purchase another 8TB drive and expect that to work?
Anything is possible, and sometimes new drives do fail. But another (perhaps more likely) possibility is that the 4 TB drive failed before the newly inserted 8 TB drive was able to complete the sync. So the 8 TB drive might be ok.
You should power down the NAS and test the drives in a PC with vendor tools (Seatools for Seagate, Lifeguard for Western Digital). Don't mix up the drives - label them by slot. You can connect the drives to the PC using either SATA or a USB adapter/dock.
Do you have a backup of the data?
How many drives were in the RN516 volume, and what was the RAID mode?
Unfortunately, I'm in a completely Mac environment with no other non-SSD machines. I have an old ReadyNAS (6310) that is used as a nightly backup device just for the ReadyNAS 516 plus an external USB device that used to have the two 8TB drives but now has one 12TB drive in it with another backup. The nightly backup to the old NAS failed last night, but nothing important was not backed up.
To try to test the newly failed 8TB drive I put it in the external USB device (configured now as JBOD) and it formatted fine on a Mac Mini. As I recall none of the drive test tools are available for the Mac.
So, now that the 8TB drive failed in the NAS, but formatted OK on the Mac, I really don't have many choices.
As a precaution I just ordered a new 8TB drive that'll be delivered today. I suppose I'll leave the according-to-the-NAS "failed" 8TB drive in the USB device and install the new 8TB drive when it arrives later today.
FYI: the 516 had five disks (Slot 1: empty, Slot 2: Seagate 8TB, Slot 3: Western Digital 4TB, Slot 4: Seagate 8TB (new), Slot 5: Western Digital 4TB (that died last night), Slot 6: Seagate 8TB), all from the approved Netgear list.
FYI: the 516 had five disks (Slot 1: empty, Slot 2: Seagate 8TB, Slot 3: Western Digital 4TB, Slot 4: Seagate 8TB (new), Slot 5: Western Digital 4TB (that died last night), Slot 6: Seagate 8TB), all from the approved Netgear list.
Ok, so five disks total, likely in XRAID.
When you replace a disk, the contents on the replacement is rebuilt from the remaining disks in the array. If one of those disks fails during the rebuilding, then you lose the volume. Loosing the new 8 TB drive normally won't cause data loss in that situation. So it's more likely that the 4 TB drive failed, and the new 8 TB is actually ok.
One thing you could potentially try is power down the NAS and restore the 4 TB drive you removed when you put in the 8 TB drive. Also remove the 4 TB drive that failed last night. Then power up, and see if that gives you access to your data. If that works, then update your backup immediately.
After that you could try doing a factory default with the two new drives in place. If that builds successfully, you can run a disk test from the volume menu settings. Then check the smart stats on all the drives. If the smart stats look good, then load the data back onto the volume.
Hmm. I shut down the NAS and replaced the bad 4TB drive with the one I had taken out. On reboot it tells me that there is no volume. The four good drives are now colored red with the replaced 4TB drive in black.
Hopefully not. You could still potentially try what I suggested - putting the 4 TB drive you removed back in it's original slot. That is, put everything back the way it was before you hot-swapped in the 8 TB drive.
But If you do try that now, use the boot menu to boot up the volume read-only.
But I have a message that says "Remove inactive volumes to use the disk. #2, #3, #4, #6."
Why are those disks inactive?
The system can't figure out how to assemble your data volume. I think it's starting with the disk you just re-added, and can't connect it with any of the others.
I suspect that will likely happen again (even if you put the disks back in there original positions). But I think it's still worth a try. Mounting read-only should prevent the system from doing more damage (writing to those the disks) if I am wrong.
I want to thank you for all the time and energy you've put into handholding me on this.
What's confusing is that the four disks that are there are in the exact places they were when the sync ended. The only thing that has changed is the 4TB drive I put in Slot 5 and it showed black. Now that drive is out and the four drives are all showing red. Thye should have the volume data on them, but the system tells me I need to create a new volume because none exists.
What's confusing is that the four disks that are there are in the exact places they were when the sync ended.
You started with 5 disks, with protection against a single drive failure. If two or more disks fail, then all the data is lost.
You replaced one disk, and when you did that you lost that protection until the resync completed. But the resync didn't complete, because a disk failed.
Going back to the original disks sometimes works, but not always. For example, it will fail if anything was written to the NAS (or deleted) during the resync. That would include automatic deletions and writes (for example snapshots being pruned, or files being defragged). Those changes result in the 3 original disks becoming out-of-sync with the disk you removed.
Data could still potentially be recovered, but that would require recovery software (or a service).
Another option is that you could attempt to clone the disk that failed to another disk. That might work if it isn't too damaged. But the more you try to do on your own, the more difficult (and problematic) data recovery becomes.
Thanks for the user manual. I did a Boot in read-only, but no change. The system now has two messages about removing inactive volumes, one for disks 2 and 6 and the other for disks 2, 3, 4, and 6. Not having the resources to dig deeper into the failure and/or clone individual drives, I think I simply must resign myself to doing a reset and creating a new volume. When the new (replacement) 8TB drive arrives I think I'll remove the 4TB drive and configure the system with four 8TB drives using X-RAID. If the system install cleanly and the four drives pass muster, I hope my backup NAS can restore the data as of two nights ago to the newly created 516.
I've never tried to restore from the old NAS. Is there a method that's recommended?
I've never tried to restore from the old NAS. Is there a method that's recommended?
I use rsync (creating one backup job per share). I run the backup jobs on the destination NAS (which is slightly easier to set up than the other way around).
Two drives in one group and 4 in the other isn't a situation you re going to fully recover from. 5 and 1 is potentially recoverable. If you have a backup, just cut and run. Factory default and restore data.
I've gone back and forth on how to proceed with the situation. It's easy to be a little confused and/or my posts may be a problem too.
Bottom Line: When I get the drive I intend to do a default setup with the four 8TB drives and start from scratch. I'm keeping my fingers crossed that A. all the drives will pass the formatting process and B. the restore from my backup works.
The drive that was supposed to arrive the same day, didn't. So, while they were trying to figure out where it was and when it would be delivered I decided to rebuild the server with the three good 8TB drives using X-RAID. (I'm running 6.10.1.) That gives it 15+TB and there'll probably be about half that used. Growth should not be terribly fast.
So, the new 8TB drive arrived and I don't know how to proceed. After the problem with two bad drives I'm gun-shy to just add the new one.
Questions:
1. Is there a way to test the new drive without changing the current RAID setup? In case the new drive is not going to format I'd like to know it now.
2. If I can determine that the drive is good, should I keep it in the box and save it? Or would it make more sense to use it as a spare? Although I'm a little fuzzy on what a spare does. Does it change the RAID so that two drives could fail and the system will be preserved? Does it make sense to just keep it spinning, adding to the heat in the unit?
1. Is there a way to test the new drive without changing the current RAID setup?
Can you connect it to a Windows PC (either with a USB adapter/dock or with Sata)? If you can, test it with Lifeguard. Run both the long non-destructive test and the full write-zeros test.
If not, you could try switching to flexraid, and add it to the NAS as a jbod volume. Then run the disk test on the volume from the volume settings wheel. After that you can destroy the volume, format the drive, and then remove it. Finally switch back to XRAID.
Unfortunately there are no Windows machines here. I looked at the documentation for switching RAID modes and it says "You can change the RAID mode that your ReadyNAS storage system uses. Because this process erases all data, if data is stored on your system, you must first back up the data to another storage device before changing the RAID format." I've already spent the last couple of days restoring the NAS with several TB of data (the best back-ups are ones made from another Mac on the network with a USB connection to a backup device so it's been a slow process.).
Because there are three drive slots open, I was hoping there was a way to put in the new drive as a separate disk and simply run the disk test on it. It's too bad that I can't.
Oh, well. It's been a learning experience. Thank you so much for your help and guidance.
I looked at the documentation for switching RAID modes and it says "You can change the RAID mode that your ReadyNAS storage system uses. Because this process erases all data, if data is stored on your system, you must first back up the data to another storage device before changing the RAID format."
You are looking at the documentation for an older ReadyNAS. OS-6 doesn't erase data when you switch to flexraid, and the procedure I posted will work. You can find correct documentation for your NAS here: https://www.netgear.com/support/product/RN516.aspx
