Install the app
How to install the app on iOS

Follow along with the video below to see how to install our site as a web app on your home screen.

Note: This feature may not be available in some browsers.

Estimated endurance of SSD M.2 has reached its limit message

As an Amazon Associate, we may earn commissions from qualifying purchases. Learn more...

16
3
NAS
DS1621+
Operating system
  1. macOS
Mobile operating system
  1. iOS
I have a DS1621+ (20GB of RAM) with two expansion units (DSM 7.2.2) and I have two Samsung 970 Evo Plus 500GB drives in an SSD Cache (they are just over 4 years old at this point). The Estimated Endurance is 1% (the other SSD currently has 2%) and is in critical condition saying to replace the SSD to prevent data loss (I only use it for cache to help performance):

1737566706698.webp


1737566883700.webp


Multiple questions if I may seeking advice and recommendations:

1) My SSD Cache was never upgraded to the latest version when I upgraded to DSM 7. If I remove the SSD cache and create a new one to get on latest version, would this help my situation at all? Or it doesn't matter because Synology knows that these SSDs are completely worn through and needs to be replaced? Wondering if anything gets 'reset' when recreating the SSD cache.

2) Can I safely ignore this message and keep using the SSDs since they are only used for cache purposes? I'm trying to think if I would ever really lose any data if I continue to use them until they truly die.

3) These have a 5-year warranty from Samsung, but not sure if they would really be eligible anyways for replacement? If it's not worth the effort, I'd rather not waste my time with the RMA process, etc. if it would be painful.

4) Since these SSDs lasted me over 4 years, I'm thinking these are pretty good to use if I buy the same ones again (970 Evo Plus while still sticking with 500GB as I'm not sure if more is really any better when using as a cache). I also see ones at BHPhoto that are from Crucial and Lexar (and others) at pretty good prices. Just curious what people recommend when using for cache purposes only.

Thanks!
 
Or it doesn't matter because Synology knows that these SSDs are completely worn through and needs to be replaced?
This is the case 99%. Back in 2018, I have had 2x 960EVO for caching in my 918 at the time.

I have burned those in 11 months time.

1737619853358.webp

1737619863246.webp


Can I safely ignore this message and keep using the SSDs since they are only used for cache purposes? I'm trying to think if I would ever really lose any data if I continue to use them until they truly die.
I wouldn't advise it. Retire the cache asap so that the date can be saved onto the main volume, and then de-attach them.

These have a 5-year warranty from Samsung, but not sure if they would really be eligible anyways for replacement? If it's not worth the effort, I'd rather not waste my time with the RMA process, etc. if it would be painful.
I have done the exact same thing because mine were under a 3y warranty. The RMA was completed in matter of days and I got brand new 970 models as replacement.

That was my 1st hand experience, and have had no data loss, and a positive experience during the RMA process.
 
Thank you so much for the tips, @Rusty. I'll give it a try with the RMA process and see how it goes.

Retire the cache asap so that the date can be saved onto the main volume, and then de-attach them.
I'm thinking I'll remove the cache from Volume 1 altogether, and hopefully safe to just keep them inside the Synology (without configuring them) until I get replacements (if this sounds wise/okay to do). I guess I'll find out if Synology barks at me (I'd hate to open her up twice if I can avoid it).

I'm curious how you are able to see the SMART info above in your screenshots for the SSDs. When I look at 'Health Info', I only get the below info:

1737652002957.webp

1737652017165.webp


I guess I'm wondering how much info Samsung may need to RMA both SSDs (since only one is Critical at this point, but the Estimated Endurance % is hopefully all they need to claim both since both are so low?).
 
I'm curious how you are able to see the SMART info above in your screenshots for the SSDs. When I look at 'Health Info', I only get the below info
The reason is simple. Those are DSM 6 screens from 2018.

During the RMA I only send them the critical screenshot that shows life and hours as well as the serial numbers, proof of purchase etc. As soon as they got it and as long as those drives are inside the warranty you should have no issues at all with all the evidence you have atm.


and hopefully safe to just keep them inside the Synology (without configuring them) until I get replacements
As long as they are not configured, there is no harm in keeping them inside to avoid opening up the NAS twice. Perfectly valid reason.


Estimated Endurance % is hopefully all they need to claim both since both are so low?)
That Critical screenshot with life and hours should be more then fine for them along all the rest that I mentioned above.
 
I RMA'd the two SSDs with Samsung - it was actually pretty painless (I called 800-726-7864 and asked for Memory). The rep took down all my details, opened a ticket, and I replied back with the info they needed (I included the Critical screenshot with my docs). I just shipped them out today via UPS (they even provided a label). Hopefully I will receive them pretty soon because I really really miss having an SSD cache (it's amazing how much of a difference it makes - it's a bit painful using my media server and accessing files without it).

Thanks again for all the help and tips! Hopefully will get her running again like before real soon ... and it will be great if these replacements gives me another 4+ years of service.
 
So I got both good and bad news from Samsung. SSD#1 that is ‘Critical’ in Synology they won’t replace and are returning to me: After arrival at our facility, it was determined that this unit's Total Bytes Written, or TBW, is 444.3 TB. The 970 EVO Plus 500 GB model has a warranty period of 5 years or 300 TB TBW, whichever comes first. As a result, unfortunately, your drive has been deemed ineligible for warranty service, as the TBW exceeds the warranty limit. I apologize for any inconvenience. This unit will be returned to you without repair shortly.

SSD#2 that had Estimated Endurance of 2% left on Synology they are replacing with an upgraded unit (MZ-V8V1T0B/AM Samsung 980 1TB (not the Pro version)). They didn’t mention anything about the TBW on this returned unit, and I did have to ship both units separately to Samsung with different RMA numbers (so not sure if I got two different people doing different things in looking at them or if they both went through the same testing at Samsung). I just don’t understand how one SSD can be above the TBW limit, and the other one being below it(?) (they were installed together at the same time and always used as a R/W cache).

Hopefully the replacement Evo 980 1TB is indeed going to be okay performance and lifetime-wise over the Evo 970 Plus 500GB.

I think my plan of action then is to buy another Evo 980 1TB to match what they are sending me to re-setup my R/W cache.
 
Does the Synology cache work symmetrically? Meaning reads and writes to both drives? Or asymmetrically where one stick is for reads, and one is for writes. There is a good chance there is a lot more reading than writing going on.
 
Does the Synology cache work symmetrically? Meaning reads and writes to both drives? Or asymmetrically where one stick is for reads, and one is for writes. There is a good chance there is a lot more reading than writing going on.
I’m assuming it works symmetrically - I believe the setup has to be RAID 1 (for protection in case one fails in a R/W cache scenario with how I set it up with pinning the btrfs metadata to it) - so it should always be writing to both SSDs to keep everything in line for the RAID. For a read-only cache you really just need one SSD and it can fail and not cause any issues to the pool (from what I understand).

Going from 500GB to 1TB with both SSDs in my setup should help a lot as the TBW increases from 300GB to 600GB (from what I read with the specs).
 
I’m assuming it works symmetrically - I believe the setup has to be RAID 1 (for protection in case one fails in a R/W cache scenario with how I set it up with pinning the btrfs metadata to it) - so it should always be writing to both SSDs to keep everything in line for the RAID. For a read-only cache you really just need one SSD and it can fail and not cause any issues to the pool (from what I understand).

Going from 500GB to 1TB with both SSDs in my setup should help a lot as the TBW increases from 300GB to 600GB (from what I read with the specs).

So the synology estimator is so full of bs, it always recommends its now ancient and laughable 800GB option. That drive formats down to 700GB usable (in the dual stick config). My Occupied is over 800GB. And reusable space at 2.8TB (this is on a 4tb drive that formats down to 3.6tb usable). If I had just their lame 'approved' of SSD the pressure and refresh on the 'occupied' portion would be very heavy with heavy turnover. Having some cushion beyond that, I think, should release that memory pressure a bit.

That said, by the time the 4tb sticks wear out, replacements will be dirty cheap. Also, I hope by that time Synology might actually have some modern hardware with U.2 drives so a lot of this cache stuff becomes unnecessary.
 

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

Popular tags from this forum

Similar threads

@Telos, my goal was to maximize transfer speed as I may have mentioned in my original post. As far as the...
Replies
18
Views
602

Thread Tags

Welcome to SynoForum.com!

SynoForum.com is an unofficial Synology forum for NAS owners and enthusiasts.

Registration is free, easy and fast!

Trending content in this forum

Back
Top