What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

Degraded Intel SSD

JD

Moderator
Staff member
Joined
Jul 16, 2007
Messages
11,931
Location
Toronto, ON
I presume this is bad? I mean it still seems functional and boots fine, but should I assume it's "dead" and replace now or keep riding it out? I guess it did well, seems it was only rated at 45TBW and somehow I'm sitting at 349TBW.

1615174234955.png

And of course it also brought to my attention that one of my IronWolf Pro's isn't doing well either:

1615174290982.png
 

sswilson

Moderator
Staff member
Joined
Dec 9, 2006
Messages
24,652
Location
Moncton NB
Was this a 24/7 drive? The hours convert to 6 years of actual up time which is pretty impressive. :)

I'm not completely sure, but I suspect the 1% is indicating that you don't have any overhead at all when it comes to over provisioned cells.
 

Valkyrie

Well-known member
Joined
Dec 3, 2012
Messages
1,270
Location
Chilliwack...Gone...Gone...Gone!
You could try shrinking the OS partition / increase unallocated space. That may make it happier for a while, but yes it does have some miles on it. Depends on how the "media wearout indicator" is determined. If it just needs more room to play it may be fine for a while.
 

Entz

Well-known member
Joined
Jul 17, 2011
Messages
1,878
Location
Kelowna
Intel data center drives are quite robust so if its saying its dying I would believe it. As above my guess is that is triggered when the PE cycles are critically low with no real free blocks.

I would definitely keen an eye on that Ironwolf. If that reallocated sector count starts increasing frequently it needs to be replaced ASAP. The fact that it is exactly 100 may mean it came like that from the factory or its a problem with the smart tool (Does Seagate have one?). A few isn't abnormal as there are spares.
 

Entz

Well-known member
Joined
Jul 17, 2011
Messages
1,878
Location
Kelowna
Actually looking apparently the 100 is a redherring and you need to look at raw. You have 0x10 (16), which is fine IMO as long as it doesn't go up.
 

JD

Moderator
Staff member
Joined
Jul 16, 2007
Messages
11,931
Location
Toronto, ON
Was this a 24/7 drive? The hours convert to 6 years of actual up time which is pretty impressive. :)
This is one of my "servers", and I bought the drive back in 2015. Not really sure why TBW is so high though since it was just the OS drive, only thing I can think is page file? But I guess 6yr of solid runtime would likely generate a fair bit of writes anyhow :)
Actually looking apparently the 100 is a redherring and you need to look at raw. You have 0x10 (16), which is fine IMO as long as it doesn't go up.
I panicked too when I saw 100, but realized it was only 16. Warranty ends April 2022 and Seatools DST says "pass", but I left the "Fix All Long" running last night and that has failed. Guess I will start a RMA to be on the safe side.
 

Shadowarez

Well-known member
Folding Team
Joined
Oct 4, 2013
Messages
4,220
Location
Arctic Canada
im looking at a U2. DC 4600 of this drive its nvme 1.2TB and only $256 was gonna use for caching a 8TB hdd. in new build. ill keep a eye on it.
 

Entz

Well-known member
Joined
Jul 17, 2011
Messages
1,878
Location
Kelowna
im looking at a U2. DC 4600 of this drive its nvme 1.2TB and only $256 was gonna use for caching a 8TB hdd. in new build. ill keep a eye on it.
The endurance on the newer Data center nvme drives is several orders of magnitude better than these older datacenter sata drives, in the multi-petabyte range, even for warranty purposes so I think you will be fine.

I run the P3700 which is the older version of that drive and they make phenomenal low latency caching and SLOGs for the cost.
 

Latest posts

Top