This topic has been marked solved and closed to new posts due to inactivity. We hope you'll join the conversation by posting to an open topic or starting a new one.
For 8 years with my predecessor RND210, used soley for Mac's TimeMachine, I have successfully stored one disk offsite, then periodially returned it to resync, then returned it to offsite. Of course, I fully understand and accept I'm without current redundancy during periods of storage. But I always have at least one current disk at ALL times, don't I? This protects my offsite disc, only since last sync, from a NAS theft and fire. To me, this is a reasonable value compromise.
Now that I've upgraded to a RN212, I'm reevaluating my strategy. I like even better the idea of rotating a 3rd disk, as proposed in this post. But I don't buy the reasons for abandoning either strategy, often pitched by NAS purisits and NETGEAR Techs who want me to acquire more equipment. I don't care how long the resync takes or lack of current redundancy. Do backups pause while resyncing? Does this cause more drive thrashing than it would otherwise? I welcome your facts as well as your opinion.
Here are the logs for one cycle (in reverse chronology):
Volume: Volume data health changed from Degraded to Redundant.
Volume: Volume data is resynced. [took 4 hours after 26 hour disk separation]
Volume: Resyncing started for Volume data.
Disk: ... was added to Channel 2 of the head unit.
Volume: Volume data health changed from Redundant to Degraded.
Disk: ... was removed from Channel 2 of the head unit.
Well, I suspect you won't buy my reasons for not doing this either, as I use other NAS for backup. I agree that requires more equipment, but I like the convenience and the fact that it is fully automated.
If you only use two disks, then if the disk already in the NAS fails (or is detected as failed) during the resync you will lose the data on both disks. So that is definitely a risky strategy.
If you use three disks (rotating two for backup), it is safer. But personally I'd rather rotate USB disks instead (or perhaps use a USB dock, rotating two internal drives). That eliminates the windows of poor performance during resync, and any possibility of damaging the SATA backplane or the internal disk(s). It also allows more frequent backups (especially if you use incremental rsync). One additional benefit is that it gives you a backup you can access from a PC (assuming you format the external disks as NTFS). So if the NAS itself fails, you don't need to buy a new one to regain access to your data.
Do backups pause while resyncing? Does this cause more drive thrashing than it would otherwise?
Backups will not pause during resync, but the NAS will be much slower.
With RAID-1, the resync involves reading every single sector on first disk, and writing it to the second disk. This is done in the background. So any access of the NAS (read or write) will create disk thrashing. It will also further slow the resync (because it is a background activity).
In most cases you won't uncover a disk problem (for instance a bad sector) until the system tries to read or write to it. So it is far more likely that the system will detect a disk failure during resync than at any other time - because the system is reading (or writing) every sector on the drive. As I said above, If it's a read failure (50-50 chance it will be), the resync fails and you lose the data on both disks.
Volume: Volume data is resynced. [took 4 hours after 26 hour disk separation]
The time of separation isn't relevant. The resync copies every sector, it is not an incremental update.
If it only took 4 hours, then your disks must be pretty small (1 TB?). So another disaster recovery option might be to use cloud storage. Google Drive perhaps.
Well, I suspect you won't buy my reasons for not doing this either, as I use other NAS for backup. I agree that requires more equipment, but I like the convenience and the fact that it is fully automated.
If you only use two disks, then if the disk already in the NAS fails (or is detected as failed) during the resync you will lose the data on both disks. So that is definitely a risky strategy.
If you use three disks (rotating two for backup), it is safer. But personally I'd rather rotate USB disks instead (or perhaps use a USB dock, rotating two internal drives). That eliminates the windows of poor performance during resync, and any possibility of damaging the SATA backplane or the internal disk(s). It also allows more frequent backups (especially if you use incremental rsync). One additional benefit is that it gives you a backup you can access from a PC (assuming you format the external disks as NTFS). So if the NAS itself fails, you don't need to buy a new one to regain access to your data.
Do backups pause while resyncing? Does this cause more drive thrashing than it would otherwise?
Backups will not pause during resync, but the NAS will be much slower.
With RAID-1, the resync involves reading every single sector on first disk, and writing it to the second disk. This is done in the background. So any access of the NAS (read or write) will create disk thrashing. It will also further slow the resync (because it is a background activity).
In most cases you won't uncover a disk problem (for instance a bad sector) until the system tries to read or write to it. So it is far more likely that the system will detect a disk failure during resync than at any other time - because the system is reading (or writing) every sector on the drive. As I said above, If it's a read failure (50-50 chance it will be), the resync fails and you lose the data on both disks.
Volume: Volume data is resynced. [took 4 hours after 26 hour disk separation]
The time of separation isn't relevant. The resync copies every sector, it is not an incremental update.
If it only took 4 hours, then your disks must be pretty small (1 TB?). So another disaster recovery option might be to use cloud storage. Google Drive perhaps.
Unless the old NAS has failed, why not utilize it for your backup strategy? Ideally, the backup unit would be offsite so everything is taken care of automatically, but bringing it local for the backup is certainly an option. A two-drive NAS isn't that bulky or heavy. If you use RSync, then the backup can be set up as incremental.
Good use for an obsolete 4.1 Duo, but I should have added that it was certifiably fried in a power failure of an ACR UPS, WITHOUT loss of line, due to a way-dead and regrettably unmonitored battery. Didn't know a UPS could be so insulted. Thanks for the suggestions; see my solution in adjacent response.
Thanks for your prompt level-headed judgements and suggestions. Got it, just never wanted to spend any more time under the hood. See adjacent reply for why this issue opened up. It was a frightening experience to find my prior confident practice actually left me with a "safe" offsite disk in the cold. Bought the new RN212, naïvely believing I could somehow migrate my 4.1 within it, but a service contract would've been required. So bought a twin RND210v1 on eBay which really was a misadvertised v2. So bought another RND210 on eBay, but the "safe" offsite disk finally reported DEAD on RAIDar and wouldn't open in FrontView! Perhaps it caught a bug being passed around so many Heads. Laid down and bought the NETGEAR service contact and patiently waded through 6 days of sweat, but they were ultimately successful! After giving up trying to recover on the old and write to the new NAS, she made me buy a 1TB USB drive and she just dumped out the DEAD 4.1 disk to USB within the new NAS due to "a greater abundance of tools there". Too late to make a long story short, so I took the 1TB drive off the shelf and easily set up a push-button backup task. Your reasoned advantages of having a computer-readable backup were very convincing. It's going on vacation! Everybody's happy now.
