[DS1522+, DSM 7.2.2-72806 Update 1] M.2 Drive 2 overheated (WD SN700)

[tsenkov]

Hi,

Something weird happened today - 1 of my SSD’s (used as cache) overheated and the NAS auto-powered down, twice.

My NAS was powered down for a few days (there was a storm → electricity stopped → NAS powered down before UPS died → I didn’t think of turning it ON until today). Today I decided to do some cable management and I realised it’s still OFF so after I finished hiding all wires and cleaned-up (also raised the NAS about 2 feet up / ~ 50cm from the floor level) I started the NAS back on, and in a few minutes I saw it was off again. So I started it back ON again and continued with some Christmas decoration at home, but an hour later I realised it’s OFF again.

This afternoon I powered it back ON and managed to check the logs - the SSD on the second M.2 slot overheated (reached 70ºC) on both occasions. I checked the history of logs too - there hasn’t been such an instance before. (or at least I can’t see it in the logs)

I checked the fan settings, and I bumped it up to the middle position (it was on Quiet before that).

Room temperature is around 22ºC, at the NAS’s location it’s probably 20ºC as it is close to a french/tall window and it’s already close to freezing cold outside, the NAS is very close to a pretty big air purifier which def moves air around it. I don’t think I would need to even bump that fan speed in those conditions?

It seems there was some intensive task, though, that was causing the heat or maybe the fan made the difference very slowly, but about an hour into operating - temperature has come down under 50ºC (~45º ATM). (I’m pretty sure temperature dropped in a short span of time, e.g. from 65ºC to 45ºC in a few min, but still long after I changed speed of the fan, so if I had to bet what was causing it - I would imagine a burst of work is more likely, but I have no idea if/how would operation result in heat for SSD’s, so I could be totally wrong) Update: there IS an intensive task which quickly rises and drops the temp of this stick - there is a still-ongoing drive-check, I guess due to the shutdown(s).

I also think this stick has always ran a bit warmer than the M.2 Drive 1.

Has anyone experienced something like this? Should I be looking to replace this stick? Happy to get your advice.

Best,
Nikolay Tsenkov

[tsenkov]

I tried powering down the NAS to open up the SSD compartment so I can check if there is some mechanical reason for the heat, but it says:

The system is processing the following task(s). Shutting down now may cause package abnormality or data loss. Are you sure you want to proceed?

Storage Manager: Storage pool operations are running. Wait for the operations to complete and try again.

And this is (I guess) the operation that seems to be ongoing:

I guess this disk check is happening because of the shutdowns? The estimate is jumping up & down by a lot, so I have no idea when this will end.

[tsenkov]

This sounds very similar to my case.

https://www.reddit.com/r/synology/comments/1e07n9j/nas_overheated_and_shut_down_should_i_be_concerned/

I am positive the NAS has operated without overheating during the summer, when temperatures indoors accidentally may’ve spiked to 30ºC (AC was stopped for some reason etc.), and the fan speed was likely at Quiet as I’ve never changed this setting before AFAIK.

Could there be something relatively recently introduced at DSM to have led to this?

[SgtNumbers]

I don’t have an SSD in my Synologys but a while back I had a problem with the HDDs just thrashing with 100% use. It was caused by Active Insight. When I deactivated it, the hard drive settled down. The thrashing was slowing things up to the extent that it took over 40 minutes to sign in to a domain connected PC where the Synology was the DC running Synology Directory Services.

[tsenkov]

Hey there, thank you so much for sharing this! I did notice my disks work a bit more, post installing Active Insight. Will try disabling/removing it and will report back here any results. BTW I haven’t had any overheats since those incidents.