FWIW, I've now been running an Eaton 5P 1500 on an RN516 for 5 days and see no such messages. Now, there could be a problem with the ARM version of the OS and it could be that Eaton has a slightly different protocl for different 6P models, but this tends to point to a problem that is not the NAS firmware.
Still waiting for answers as to whether you have tried another USB cable and another USB port on the NAS.
Unfortunately, it appears you cannot delete a configuration for a directly connected UPS and then re-install it, which might be all it takes. I'm opening a separate thread on that.
FramerV, what is your reason for failing to reply for 4 months to the original issue? You told me that you would investigate and reply? Is the fact there is still no solution the reason for your silence?
Can I please get a time frame for someone looking at the new logs? Thank you.
Please forgive my indignant attitute but is it completely unacceptable to me that this forum is Netgear's excuse for customer support. In saying that, thanks very much to those who have actually replied to me.
Looks like that message (and maybe one of yours saying you already did) got lost in one of the forum resets a couple weeks ago. I reposted a lot, but didn't notice this one was gone.
One thing I did not ask before: Do you know if your UPS firmware is up to date? The most recent update was in December.. You can check the version via the front panel.
My UPS doesn't have a front panel, only a power button and some lights. It's the 5S 700. When I checked the Eaton site for firmware, it said "Sorry, no 5S - Firmware available."
For some reason, I was thinking you had a 5P. I should have gone back to the beginning and checked. I have found some interesting messages on the NUT forum with regard to the 5S line, and they sound a lot like your problem.
Here is how one concludes (after many failed suggestions on how to fix it):
These results are the same on two different hosts, two different 5S 1000LCD UPS's, each with their own OEM cable, which is a bit thicker and more substantial looking than the ones that came with the Eaton 5110 aka Powerware 1000i. So while both cables may be bad, it's starting to look like this model UPS may have a flawed USB design?!
While this is from a year ago, I can find nothing hinting that the problem was solved.
FramerV, what is your reason for failing to reply for 5 months to the original issue? You told me that you would investigate and reply? Is the fact there is still no solution the reason for your silence?
Can I please get a time frame for someone looking at the new logs? Thank you.
Please forgive my indignant attitute but is it completely unacceptable to me that this forum is Netgear's excuse for customer support. In saying that, thanks very much to those who have actually replied to me.
Thanks for your observations. Hopefully, as the support people on the phone directed me here for 'support', the occasional look at fixing something wrong might cover my issue here after 8 months.
Latest firmware still has the issue. From what I have observed the issue happend when the drives spin up (when I try to access them) so seems to be prompted by something on the NAS, not the UPS.
We've had a Eaton 5P 1500 connected to a RN104 with 4 disks for two weeks with disk spindown enabled, but we're still unable to reproduce the issue. However, we do have a hunch of what may be happening.
We'll be posting a 6.8.1 Beta with a potential fix for this issue within the next 24 hours.
NUT (the UPS monitoring software) determines if UPS data is stale by comparing the current time to the timestamp of the last UPS update. These updates should occur every few seconds. Depending on a number of factors (which I really don't know), a running system process can be frozen while a disk spins up. So if the disk takes a long time to spin up, and it happens to start spinning up at just the wrong time, then the time difference between the last UPS update and the current time could grow larger than what NUT determines is normal. Then it will mark itself as having stale data, and send a communication error notification. We should be able to fix this by just extending the timeout long enough to allow the disks time to spin up.
Anybody care to have a second look? I was hoping after almost a year this was resolved...
Skywalker did suggest that 6.9 would have the longer timeout, so maybe you should try that. It's at RC-1, so it should be stable. Also, it likely will be released shortly, so you could of course wait a bit longer for that.
