Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[fordem]

Here's an interesting one for you guys to puzzle over...

Hardware is a Rev A1, running 1.10 firmware, configured to hibernate 5 minutes after the drives are idle, but there is a Securicam configured to put a snapshot on the NAS every 15 seconds between 6:00 am & 6:00 pm, so the unit stays active at least 12 hours a day.

About a week ago I noticed I was getting overtemp shutdowns so I pulled it apart and cleaned out the dust bunnies, which hasn't helped - I now have the fan set to always on high, which does prevent the shut down, but, here's the strange part - idle temps when the drives are hibernating are higher than operating temps.

Operating temps run around 46*C, idle temps are around 54*C - who wants to propose a theory?

[JavaLawyer]

On the DNS-323, if both HDDs are hibernating, does the fan still kick-in?

[fordem]

I have no clue - but since I now have the fan set to always on high - I don't think the temp rise is related to that.  I'm determining that the unit has hibernated from the power switch backlight going dim, around the time I expect it to - it's 8:33pm now, the unit has been hibernated for over two hours, and the indicated temp is 57*C

Prior to this incident, with the fan set to auto off/low/high - the idle temps would drop somewhat lower than active temps, but I never really paid it any attention because I never got over temp shut downs.

[fordem]

7:51 am - unit active, indicated temp 47*C

I'm going to assume I have a failing drive - my IR gun shows both drives around 44*C, I'll check them again this evening.

[dosborne]

In a hot room, mine are in the low 40's

Wed, 4 Jul 2012 00:02:09 NAS41 Size:3662GB Free:3185GB [86%] Temp:42°C Up:5 days 10 hours 28 minutes
Wed, 4 Jul 2012 00:02:09 NAS42 Size:3662GB Free:1107GB [30%] Temp:42°C Up:5 days 10 hours 28 minutes
Wed, 4 Jul 2012 01:00:03 NAS40 Size:1830GB Free:0952GB [52%] Temp:44°C Up:5 days 11 hours 26 minutes
Wed, 4 Jul 2012 01:00:03 NAS41 Size:3662GB Free:3185GB [86%] Temp:43°C Up:5 days 11 hours 26 minutes
Wed, 4 Jul 2012 01:00:03 NAS42 Size:3662GB Free:1107GB [30%] Temp:42°C Up:5 days 11 hours 26 minutes
Wed, 4 Jul 2012 02:00:03 NAS40 Size:1830GB Free:0952GB [52%] Temp:42°C Up:5 days 12 hours 26 minutes
Wed, 4 Jul 2012 02:00:03 NAS41 Size:3662GB Free:3185GB [86%] Temp:41°C Up:5 days 12 hours 26 minutes
Wed, 4 Jul 2012 02:00:03 NAS42 Size:3662GB Free:1107GB [30%] Temp:40°C Up:5 days 12 hours 25 minutes

[fordem]

8:30 pm, IR gun shows 43*C on the drives with them hibernated, displayed temp is 56*C and the fan is OFF even though it is set to manual always on high - I'm changing my opinion, this is not a failing drive, it's D-Link's screwy firmware not doing what it should.

[JavaLawyer]

This thread has me wondering whether the DNS-343 engages the fans if all four HDDs are hibernating.  The DNS-323 and DNS-343 share many firmware commonalities.  I'll have to test this out.

That said, IR guns don't lie, but firmware can.  

If it adds value to this discussion, and although intuitively obvious, but mildly interesting, I find that the temperature of hibernating HDDs directly correlates to the physical proximity to active HDDs in my DNS-343s. Heat is radiated from the active HDDs and contributes to the temperature of the neighboring hibernating HDDs. However, even on the hottest of days, I've never had a hibernating HDD broach 45*.

All HDDs Hibernating
HDD1 (Slot-1) - 37* (Hibernating)
HDD2 (Slot-2) - 38* (Hibernating)
HDD3 (Slot-3) - 37* (Hibernating)
HDD4 (Slot-4) - 37* (Hibernating)

One HDD Active, Three HDD Hibernating
HDD1 (Slot-1) - 42* (Active)
HDD2 (Slot-2) - 40* (Hibernating)
HDD3 (Slot-3) - 38* (Hibernating)
HDD4 (Slot-4) - 37* (Hibernating)

One HDD Active, Three HDD Hibernating
HDD1 (Slot-1) - 41* (Hibernating)
HDD2 (Slot-2) - 43* (Active)
HDD3 (Slot-3) - 39* (Hibernating)
HDD4 (Slot-4) - 38* (Hibernating)

[gbaker3]

That really is strange. I guess it could be a faulty temperature sensor. Have you tried flashing back to an earlier f/w version yet? Usually you're the one handing out advice, so I'm sure you will figure it out. I'm just really curious what the problem turns out to be.

EDIT: My idle temps are 38^oC with an ambient temp of about 25^oC with fan set to Auto (Low/High)

[JavaLawyer]

Somewhat surprisingly, the DNS-343 firmware does not have a fan speed control option.  

[fordem]

Somewhat surprisingly, the DNS-343 firmware does not have a fan speed control option. 

Why should that be surprising - none of the other NAS units I've used have them, very few servers have them (and some of those suckers are LOUD!), I've never understood the obsession that some forum members have with it on this NAS, and in fact, until I started having "overtemp" shutdowns, I never went near it.

Anyway - none of those setting seems to make a difference to my idle temps, the NAS has not done an overtemp shutdown in about a week, possibly because it's no longer in  the rack where it usually sits, possible because it's been diassembled and all dust bunnies evacuated.

I'm loath to do a reset to factory defaults, because it means having to recreate user, passwords & scripts that I have long forgotten (I use the firmware scheduler to do backups from other servers), but I guess that's the next step.

[JavaLawyer]

Why should that be surprising - none of the other NAS units I've used have them, very few servers have them (and some of those suckers are LOUD!), I've never understood the obsession that some forum members have with it on this NAS, and in fact, until I started having "overtemp" shutdowns, I never went near it.

The missing function is surprising to me from a continuity perspective, since the DNS-323 and DNS-343 seem to have evolved from the same (or similar) technical specifications. I wholly agree with you regarding the lack of utility the fan speed function serves. The mere fact that it took me four years to notice that the fan speed function was missing speaks to how much I value it.

I'm loath to do a reset to factory defaults, because it means having to recreate user, passwords & scripts that I have long forgotten (I use the firmware scheduler to do backups from other servers), but I guess that's the next step.

Won't the Configuration Settings file retain a good amount of that information?

[fordem]

What are the chances of "restoring the problem" when restoring the configuration?

[JavaLawyer]

As you alluded to, without a known root cause "restoring the problem" is a real danger since we don't know whether the error originates from corrupt data that will be subsequently copied to a new Configuration Settings file. That said, there are other forms of corruption that could occur independent of the data copied to the Configuration Settings file (in which case you are free and clear). Without a definitive root cause, this leaves trial and error.

Whenever I make a significant change to the ShareCenter settings or perform a firmware update, I save a copy of the Configuration Settings file and append the date and firmware version to the file name. I also maintain a dated archive of Configuration Settings files for restoration purposes, so I can regress to a settings file that I know to be stable or recover settings from accidental deletion. This is a practice I also follow with my router and media player.