I am currently using an RN102 on the current beta. I have two disks set up as two volumes (no RAID)
On the weekend I initiated a backup of files to a portable USB disk in anticipation of a disk upgrade (I have run out of space on one of the disks). While the copy was running, I started receiving error warnings about ATA errors on the other disk (ie: the disk not involved in the copy). The copy went for about 14 hours, and the errors started a few hours after and continued through the copy. There were just ATA errors, and there are 0 reallocated sectors.
At the end of the copy I turned off the NAS (so I don't know if they would have continued after that). I decided this was as good a time as any to clean out the dust bunnies (there was a surprising amount of dust, but the CPU and HDD temps were in a normal range). After the clean I turned the system back on and have had no errors since (24 hours uptime). As soon as I powered the unit on I initiated a backup of the drive throwing the errors to a (different) portable USB drive, without issue (ie: no more ATA errors).
Has anyone else ever had a situation where they received ATA errors which were resolved by a power cycle? The only other thing I can imagine this would be related to is an excessive power draw from the USB drive which has led to strange behaviour.
Thanks in advance.
Rob
Model: RN10200|ReadyNAS 100 Series 2- Bay (Diskless)
ATA error is usually caused by SATA controllers not being able to communicate with the disk. So it is possible that the chassis caused the errors logged during the backup procedure and cleaning as you mentioned made the SATA connection better thus not getting any errors after the reboot. If after what happened and you are still getting ATA errors it is also possible that the issue is within the disks internal connection and not with the chassis.
It is always a good practice making backups of your data to protect it and avoid any lost.
I had several days of things operating normally (ie: no further ATA messages), during which time I ran a long SMART test (smartctl from the command line) which came back all clear. I decided to do another backup to the same USB drive I had previously been using when I saw the errors. Within a few minutes of plugging in the USB drive I started receiving ATA Errors again, and they continued until I removed the drive 1 hour later (to be exact, the last error message was 1 min after the drive had been removed, but that could be related to the polling speed of the error collection/reporting system). So, it appears increasingly likely that the ATA errors are somehow related to the USB drive.
Some interesting things to note:
1. The problem only occurs with this drive. In the last few days I performed a backup to a different model drive without issue.
2. This is the second USB drive of this model (Seagate Expansion portable 3TB) I have owned. The first drive did not have these errors, although it has been some months since I have plugged it into the unit as it is normally located offsite.
3. The only other change in the USB realm is that I recently installed a UPS (Cyberpower) which talks to the ReadyNAS on the other back USB port. I wouldn't imagine this would have a significant power draw, but you never know.
Has anyone else ever had similar issues when using portable USB drives?
Some time ago I think there was a report here that errors related to a USB drive were being mis-reported as being on an internal drive. Since this appears to be specific to a particular USB drive, I think the next step is to test that drive in a Windows PC with Seatools.
If you want to pursue the UPS angle: There won't be any power draw from the UPS. But there often are errors in the NAS on the UPS's USB interface (since the UPS often isn't a "well behaved" USB peripheral). I see a lot of these on my own NAS (which also are connected to Cyberpower). You could try disconnecting the USB cable for a while, and see if there is any change to the ATA reports. The NAS will of course still be protected from short power loss, but you would need to manually shut it down if there is an extended power failure.
