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.
I just installed a new NAS and was about to re-purpose the 516 to lighter duty. The drives were pulled and a new (new and old) set was put in.
When I try to boot up the NAS now, it fires right up, the nav buttons on the front light up but the display never coes alive. The fan sits at full speed and never falls back. I have tried leaving it for over an hour any nothing seems to happen. I cannot see it on the network and the LAN port LEDs seem to confirm this with no orange activity light at all.
I have tried leaving only the new disks in, and then only the old ones. I tried putting it into boot selection mode, but it is unclear if it ever gets there as the display is dark.
I tried connecting a monitor to the HDMI and again there is no signal to speak of.
Ok, so today I tried to insert the newly built array into the old NAS. The new drives went back into the new NAS and it came up just fine. So there is that.
The old drives went into the old NAS and the unfortunately (although somewhat predictably) the NAS exibited the same behavior.
I then harvested an ATX power supply from a computer I have running in the same cabinet, but could not get it to work at all! No response from the NAS at all. So I reconnected it's own PSU to ensure I had done nothing bad (I had to pull the panel with the motherboard out to get access to the two connectors). Good news is that it was still trying to power on with it's own PSU.
After reading a little on ECC vs. non-ECC memory, I was conforatble pulling a non-ECC module from the same computer and swapping that into the 516. Amazingly it went straight to the boot process and complained that there were no disks!
I replaced the motherboard so that the ATA port daughterboard was mounted and fired it up again, and it came up normally.
So long story short, memory module was bad. Ordering a new ECC one now.
I just pulled all the disks and tried again. Unfortunately still the same behavior. The fan stays on at full speed, the front panel is lit up, with the exception of the LED disply which is dark. I connected a monitor to the HDMI and get no input from the NAS at all.
Note that when I was swapping out the NAS for the new one, the LED display was working well. However in the past I have seen it go dark, but it had no immendiate impact on the NAS performance so I never gave it much thought.
I just pulled all the disks and tried again. Unfortunately still the same behavior. The fan stays on at full speed, the front panel is lit up, with the exception of the LED disply which is dark. I connected a monitor to the HDMI and get no input from the NAS at all.
Yes, it makes no (apparent) difference when the disks are out or in.
I am stumped. I did create a recovery key and tried that, but without feedback from the LED or the HDMI port it is impossible to know if anything is happening. Although I suspect it is just not getting that far.
I would assume it was dead if I was not using it less than 48 hours ago.
Yes, I am. Unfortunately when I looked it up, the date of purchase was sept 2013, so we're already 12 months out of waranty.
Could it be that the memory came unseated? I have never opened it, but wondering what might cause it to fail to try booting. I did move it when it was retired, I am confident it did not drop, but was moved around a bit so may have been knocked?
Thanks for all the continued ideas and consulting.
I tore the NAS down and pulled the memory module and reseated it. I also unplugged and re-plugged all the connectors while taking it apart. I then reassembled it and fired it up again diskless. Unfortunately the exact same behavior as before. I then tried with one of the old NAS disks in it (that should have the partitons and OS) but again the same behavior.
Is there any easy thing I could try to get it going again? Should I replace the memory? I am not sure what would cause it to look the way it does. There is only one connector to the bundle of wires (not a ribbon cable interestingly) that goes to the front panel. So I can only assume that the power going to the front directional buttons is the same as the power going to the LED display?
Is there an easy way to test the LED display? I would love to know if it is supposed to be showing something and it just pooched.
There should be a ribbon cable, with a "snakeskin" cover for protection. If it's just a display issue, though, the fan should quiet down and you should see activity on the NIC LEDs. You could try RAIDar to see if it detects the NAS, but I think it won't.
I purchased a 516 in similar condition. I only needed the door & display, but gave fixing it a shot. It had a dead short on the 5V bus. Likely a blown capacitor, but none looked overstressed. Even with a schematic, finding the problem would have been tough. I never fixed it. Do the tops of any of the electrolytic capacitor "cans" look domed instead of flat?
Thanks for the example. I guess it could have died during one of the poer cycles when I was swapping in the new set of disks.
I'll have another look at everything (still have not tried to pull the CMOS battery) to see if i can get it to kick back into life. Otherwise i guess I'm looking for a new one.
Re. Capacitors, I really have no idea. I have not spent a lot of time at the compnent level for many years. Tracing circuitry is slow and painful especially without a scematic.
Ok, interesting development. I pulled the CMOS battery (there was some sign of dirt at the bottom, but nothing bad I suspect. However when I replaced it about 5 mins later, the LED now lights up!
I get the Welcome to Netgear message, when I have no drives in the NAS.
I turned it off and put a drive in, but now it fires up (LED working with the Welcome message) but then it immediately powers down (fans get to zero) and then automatically tries again, with the same result. I let it go for a few cycles and then powered it off with the rear switch when the fan hit zero.
Now i am not sure what the issue could be? Has anyone seen this behavior? Could it be a power issue?
It does sound a lot like a power issue, but not necessarilly the power supply itself. I'm unsure why removing the CMOS battery had any effect, though another user reported some time ago a similar situation with a Pro6 where battery removal revived it (all the way, in his case).
The 516 supply looked like a standard Flex ATX supply to me (not with swapped pins like older NAS), but I didn't do any actual voltage checks.
Thanks for the reply, and I agree, I am not clear on how and why this made a dfference.
Here is some more detail.
1. Clearing the CMOS get the LED working, unclear as to why, but it has been consistent.
2. The First power on stays on (stuck on the Welcome Screen).
3. All subsequent power on result in the cycle where it tries to start then shutsdown and tries to restarts automatically etc.
4. If I clear the CMOS again the first power on stays on (stuck on the Welcome Screen).
5. Again subsequent power on result in the power cycling as in (3)
I have tried this a few times and it is repeatable. I have also tried to do first power on with disks loaded in and it does stay on (stuck on the Welcome Screen). Not sure how or why this is the case.
I am wondering if there is value in replacing the memory module to try this. Unfortunately there cost of one of these ECC modules is higher than I would like for a throw aay experiment. I can likely dig up an old ATX PSU to try that.
Also would the strength of the CMOS battery factor in here? Should I put a new one in?
Thanks, that's a good idea. Didn't know I could put a keyboard on it? USB I assume?
I am testing a new approach:
I am building another arry on the old disks in the new NAS (shutdown and pulled the drives out of it) to see if that brings it an easier path to boot. If the local flash image is currupted would the NAS ignore it if the disks have an image? This is a working theory that I am going to try. Once I have the array up on three disks (tomorrow) I am going to shut the new NAS down and switch them into the old NAS. Then restore the new one by putting it's original disks back in.
If the local flash image is currupted would the NAS ignore it if the disks have an image?
The boot loader should check the version in the flash against what's on the disks. The flash will update the disks if the flash is newer, similarly the disk image will update the flash if that is newer. Either way, the system eventually boots from the disks.
But of course the boot loader itself might be damaged.
Thanks for the information on the bootloader. My current theory is:
1. The NAs tries to boot from the disks first by default. If they can't find the image there, they try to bot from the flash image and at the same time reset the boot order in the BIOS. Then it fails to boot from a corrupt/bad image in the flash.
2. When I try to boot a second time the boot order is set to try directly from the flash image and this results in the power cycling.
3. It is not a solid theory as I am not clear why it does not power cycle once it fails to boot from the disks the first time. That said, I am hoping that I can encourage it to boot directly from the new 3 disk RAID array I am assembling in the new NAS.
I plan to pul the CMOS battery out before this attempt hoping that the default boot order is the disks first (support this partial theory above).
If the disks are unformatted, then the boot loader will initiate the factory installation from the flash image.
If the disks are formatted, but can't be booted, then the system should try to go into a recovery mode (which does run from flash, but I think is different from the install image in the flash).
FWIW, I am wondering if the CMOS battery is somehow powering the LCD panel (perhaps the same voltage pin is used to keep the real-time clock running - which is the normal function of that battery). If so, that suggests that your problem is power related (hopefully the PSU, since the system board can't be repaired).
If you have a spare disk, you might try seeing if you can boot the NAS with a single disk (which obviously takes less power).
I did try to boot from a single clean (unformatted) disk before I found the CMOS reset would kick the LED back to life. In that case the bahavior was as described before. The fans spun up to max and never came down, the LED was dark and the reset process did not have any effect.
I agree that the PSU is still suspect given the power cycling (seems like it fails on spin up) however it seems to have adequate power when it is the first attempt (disks and fans spin up) and stay running. I'll try the PSU, but really want to try this pre-built array first. I am debating upgrading the other NAS to 6.10.2 given your earlier comment of the comparison, as this would guarantee that the disks have the newest version (the 516 was at 6.10.1 when it died).
I'll let you know tomorrow once the array is built and I can give it a try.
Ok, so today I tried to insert the newly built array into the old NAS. The new drives went back into the new NAS and it came up just fine. So there is that.
The old drives went into the old NAS and the unfortunately (although somewhat predictably) the NAS exibited the same behavior.
I then harvested an ATX power supply from a computer I have running in the same cabinet, but could not get it to work at all! No response from the NAS at all. So I reconnected it's own PSU to ensure I had done nothing bad (I had to pull the panel with the motherboard out to get access to the two connectors). Good news is that it was still trying to power on with it's own PSU.
After reading a little on ECC vs. non-ECC memory, I was conforatble pulling a non-ECC module from the same computer and swapping that into the 516. Amazingly it went straight to the boot process and complained that there were no disks!
I replaced the motherboard so that the ATA port daughterboard was mounted and fired it up again, and it came up normally.
So long story short, memory module was bad. Ordering a new ECC one now.
