I know somebody that has the power on hours count so high that it's about to loop and come back to zero. At least this is what he told me.

I have one that I pulled out of an old eMachines from the single core LGA775 Celeron era. It had 65000 hours on it. That eMachines was on pretty much from when it left the factory (08) til when I got it (early 2016). The machine still works to this day but the board won't boot if any IDE or floppy drives are connected or I try to load any Windows OS. Pretty good endurance for a bottomline machine. Also shows that Seagate barracudas (a 160GB in this case) are actually pretty good.

The amount of.dust in that machine though was horrifying. How it still booted I have no clue. It made me sick when I opened the side panel, the PCB was covered (and I mean COVERED) in at least a 8th of an inch of dust. The CPU cooler literally shot out a dust cloud that would chokw you when you powered the machine on. The dust had fowled the DVD drives mechanism. You could have potted a small plant with the amount of crap inside this PC and that's not an exaggeration. It's the worst case of dust contamination ive ever seen. I think the intakes had just sucked it in from 8 years of (likely) sitting in a room powered on. Pretty freaking horrific.

It doesn't beat either of the previous posters' drives, but here's my brother's old Maxtor:

He doesn't use this drive anymore, but from what I can remember, he would turn his PC on, and... just leave it on. The drive still works perfectly, too, as you can see from the other health parameters (I also ran surface tests on it, which came up clean). Not sure what's up with the CRC Error Count, though; I assume he used a bad cable at some point.

Pretty easy to beat on server type loads.
The longest nonstop operation I observed was a Quantum Atlas 18GB drive that ran from 2001 until around 2015 when it began incrementing it's grown defects list. It's counter was 16bit and it did overflow twice in that time (this can be seen nicely in the SMART selftest log).
It was used for fileserver tasks first, than it began hosting the OS and webpages/logs on the same machine.

My SSD has 16393 power on hours on it. That is SSD, not HDD on my main rig.

My oldest drive in my main rig has 36531 hours on it, 500GB SATA deal.

But all drives are running perfectly fine. I have hardly any wear on any of them. One of them is a Samsung SSD which will probably outlive my hard drives (Around two years of constant service and only 95% wear)
there is my 500 GB drive I use for backups, Samsung, it's my worst for wear drive out of all of them, but it's healthy and fine.

320 GB as a bootloader drive for my really strange bodge job of a dualboot (Windows 7 and Windows Server 2016)

For data storage I have a 2TB HDD, Formatted almost half and half, both with EXT4 partitions. Yes, you've heard me right, EXT4 partitions on Windows. If anybody is interested I can explain.

My oldest drive (Which is hardly a competitor for oldest drive) is a 2.5GB WDC deal, EIDE from 1997. I don't even know if it has SMART on it, and it's in a 486 so I don't have those readings for you.

However, the drives which have probably been through the most hell are my server's Ultra320/SCA-80 drives. I'd image they wouldn't be brand new per-say.

44071 power-on hours with a power-on cycle count of 209 on my current main disk, a Samsung HD103SJ. Pretty good, but I think my previous drive had more.

I think I won. 😁

(Zapnuto hodin == Power On Hours)

What is more funny, it actually is 120 GB drive made by Fujitsu. It worked quite well, I was able to go through the data, but at some point it started behaving weird. It lost about 30% of capacity, started showing as other drive and stopped reading data what was there. I still have the drive, but it has not been powered up for maybe three years?

kaputnik wrote:

So, was over at a friend's place tonight, and among other things, we discussed his home server upgrade plans. When we came to ge […]
Show full quote

So, was over at a friend's place tonight, and among other things, we discussed his home server upgrade plans. When we came to getting new hdd:s vs keeping the old ones, it turned out his current boot drive was an ancient 80GB WD Raptor. The server had been online more or less constantly since he bought that drive, so just had to dump the SMART data, and check power on hours counter:

rhours.png

Could hardly believe my eyes, that thing must be indestructible or something. Almost 10 years of 24/7 service! I thought my record of ~50k rhs was something, until today 😁

Anyone here beating that? 😁

What's even more impressive is the flawless looking SMART stats. The drive looks brand new, other than POH.

I think I have an IBM deskstar with over 80,000 hrs on it too. I'll look and get back here.

One thing to note, older Fujitsu drives (and maybe newer too?) display their power on time in either minutes or seconds (I can't remember), so at first glance those drives would appear to easily beat yours. 🤣.
https://www.smartmontools.org/wiki/FAQ#Iseeso … .Whatdoesitmean

I have a drive on an old Tivo that I decommissioned a few months ago. It was running nonstop for 14 years (after I upgraded it) and probably has 120k hours on it.

krivulak wrote:

http://i1382.photobucket.com/albums/ah276/krivulak/hehe%202_zpsr2pkhynw.png […]
Show full quote

I think I won. 😁

(Zapnuto hodin == Power On Hours)

What is more funny, it actually is 120 GB drive made by Fujitsu. It worked quite well, I was able to go through the data, but at some point it started behaving weird. It lost about 30% of capacity, started showing as other drive and stopped reading data what was there. I still have the drive, but it has not been powered up for maybe three years?

How is that even possible ?!
This count means that this HDD has been turned on for 10 THOUSAND years !! Did you find that in a computer made of rocks in a cave ? 🤣

I've seen failing drives give nonsense POH. I have two of the same model drive (WD800JD-22LSA0) that both give impossible POHs. One is 297,839, the other is 1,065,173. I wonder if it's a weird quirk with this specific model?

Here is my IBM Deskstar 20GB at 84702 POH:

And a WD 40GB with almost 75000 POH:

I've also got a WD800JD-00JNA0 with 80,055 POH. None of these drives I'd consider reliable. they've got SMART stats that indicate they are on their way out. That's what makes the OPs Raptor all the more impressive.

A FreeNAS server. That's the highest i found atm but i had HDDs with 100k+ h.

Deksor wrote:

krivulak wrote:

http://i1382.photobucket.com/albums/ah276/krivulak/hehe%202_zpsr2pkhynw.png […]
Show full quote

I think I won. 😁

(Zapnuto hodin == Power On Hours)

What is more funny, it actually is 120 GB drive made by Fujitsu. It worked quite well, I was able to go through the data, but at some point it started behaving weird. It lost about 30% of capacity, started showing as other drive and stopped reading data what was there. I still have the drive, but it has not been powered up for maybe three years?

How is that even possible ?!
This count means that this HDD has been turned on for 10 THOUSAND years !! Did you find that in a computer made of rocks in a cave ? 🤣

You know Marty got in the Deleron and dropped it off with the cavemen and a solar panel, its been running ever since 🤣.

I seen older servers with upwards to 100k hours on the hdds in my line of work. I seen one with close to 200k.
I seen one with a bad smart table with over 1mil. 😵