My 7 year old (or so) Synology NAS 2415+ was powered off unexpectedly this morning. When I attempt to power it back on, the power button blinks blue and the alert LED blinks amber and a few of the drive LEDs light steady green. It sits like this for a while, then the remaining drive LEDs come on steady amber and then shortly after it powers itself back off.
I suspect it attempted to apply an automatic update overnight and then failed during the reboot. Intel Atom C2538 CPUs have a known defect, and some older Synology devices were affected. They’re known to work a bit like a fuse where you power down/reboot and then all of a sudden it just stops working.
I contacted Synology Support, and was basically told I was out of luck. The NAS is out of warranty and there is no option for any repair. The agent did mention I should Google for the “100 ohm resistor fix” and I have found this ( C2538 Clock Fix Confirmed - DS2415+ : synology ). I have ordered an overnight Amazon order of a pack of 100 ohm resistors and a soldering station, so I will update this post once I try that “fix.”
I don’t have a proper full backup of the data, so I am a little annoyed that this happened when it did. As luck would have it, I had ordered a replacement NAS anyway, but it’s only a week or two old, and I wasn’t ready yet to trust it. (It’s running newer software, etc, so I wanted to give it a month of “burn in” before I started to relocate the data from one to the other.) Here’s hoping this resistor is temporary enough fix to get it back up and running long enough for me to do the data migration.
TL;DR Hardware fails. RAID is not a backup… have your data backed up even if it’s stored on a RAID device. Synology’s support in this case is kind of shitty for the fact they knew this hardware issue could occur, and they’d rather you make more landfill than provide any paid support option for an out of warranty device.
Does the new NAS have the same number of drives? It should be possible to put the old drives in the new NAS in the same order and then one by one replace it with the new drives from the new NAS and let it rebuild. Or copy the data off to a backup device and then place the new drives in and copy the data back.
I know that works for some models of Synology and QNAP devices, but don’t know whether it will work for the specific models you have.
If the replacement drives are bigger, some NAS will let you then resize the LUNs to take advantage of the extra space, some won’t, so that would be worth checking, before going the RAID repair route.
Well, I guess it’s good news and bad news. The bad news is my aging eye sight made doing the soldering more of a chore because of the small size of everything. The good news is that, with the drives removed (because I had to fully disassemble the device to get at the area to solder) the NAS booted and beeped after I did the soldering and reassembly. I then powered it off, put the drives back in the correct bays (I had carefully numbered each with a post-it note as I removed them), and tried booting again. It took a while longer than I thought it should, but it didn’t behave (oddly/incorrectly) as before and it did eventually beep, as it should when booting normally.
I got logged into the admin panel, and checked around. It appears my suspicions were valid… it was an automatic update that caused the reboot that led to the Atom failure (picture of the event log entry below.) As this was a clean shutdown for the reboot, it appears that I have no worries of corruption.
TL;DR The Atom problem appears to have been the issue as soldering in the 100 ohm resistor appears to have supplied a fix. (For how long it will last is anyone’s guess, but hopefully for long enough for me to copy the data off the NAS and move it to a newer NAS.)
