My understanding is that TrueNAS/ZFS checks data to make sure it is correct and if it finds issues can self heal. Below in this video (it is linked at the correct time) from Steve Gibson ( grc.com ) he indicates that over time SSD’s need their data to be rewritten every so often to prevent bitrot and…

This has now been explained in the podcast [image] The Care and Feeding of SSDs - TRIM and Charge Refresh | TrueNAS Tech Talk (T3) E049 Announcements On today’s TrueNAS Tech Talk, Chris and Kris take on some SSD truisms and myths - does old data on SSDs “go st…

Consider a larger video, over time the speed it reads at falls but no errors develop. Rewriting this file will restore that performance, i.e. cell charge is restored, whether it is the same cell that is rewritten or the data is remapped. I don’t believe just reading a non faulty cell/sector restores…

[image] Chris Peredun: Detect the marginal charge Is this true for HDDs as well, albeit with a different underlying technology? It may not be user-facing, but don’t HDDs have their own (limited) spare LBA’s reserved for this? Maybe I’m remembering it incorrectly.

My two cents - if this is really bothering you as a serious issue, I’d imagine a script that does the following run manually as you see fit would ‘solve’ this ‘issue’: disable & delete old snapshots copy file, delete original, rename copy to same name as original repeat above for each file, 1 by 1 …

Wow, this is a fairly long thread about SSD’s and if the data needs to be rewritten to stability and ensure optimized I/O speed. I personally do not see this as a problem. How much does a SSD slow down? a few nanoseconds? Is it real world noticeable? I do believe that if TrueNAS (the company) se…

[image] Fleshmauler: I’m not trying to dismiss your concerns, but I’ve never seen or experienced this to be a problem in consumer or enterprise world. In theory, this is a real problem. But, in my opinion, nothing I would want to do anything about. Other than purchase SSDs from well known man…

it makes no difference if the SSD remains powered on or just left in a box, degregation still happens. The solution would be similar to a scrub but each zfs cluster unit that is over X number of days is re-written.

Just to add a small, real-world data point on the issue. Old Samsung 840 (both normal and EVO) and 850 EVO SSD still seem to have some issue with this. Before firmware update, they didn’t do rewrites even after encountering marginal reads. After firmware updates they do rewrites, but the definition…

To me it seems as if @shoulders hasn’t properly read or understood @HoneyBadger ’s explanation. If SSD firmware does detect marginal charges on read and then rewrites the cells themselves (which I believe to be true fopr most reputable SSDs) then a full read during a scrub will rewrite all “rotten” …

Not bitrot, cell charge, but I will leave it at that.

[image] shoulders: to prevent bitrot If I understood everything here correctly if the cell charge decreased enough over time the cell will become unreadable which is then called bitrot. The potential performance decrease should also be eliminated if the marginal cells get rewritten from firmw…

SSD / NVMe - Bits need to be rewritten to maintain performance and integrity

TrueNAS General

shoulders January 17, 2026, 3:57pm 40

This has now been explained in the podcast

Topic		Replies	Views
The Care and Feeding of SSDs - TRIM and Charge Refresh \| TrueNAS Tech Talk (T3) E049 Announcements Podcast , T3 , SSD , TRIM	61	1777	March 12, 2026
Some hdd always become unavailable after a reboot, help needed TrueNAS General CORE , Hardware	20	1104	May 19, 2024
We’re bringing some SMART options back Announcements	75	6363	July 12, 2026
Unexpectedly high SSD-wear TrueNAS General SCALE	45	3555	June 3, 2024
ZFSisms that are not true, or no longer true Resources ZFS	70	1887	June 17, 2025

SSD / NVMe - Bits need to be rewritten to maintain performance and integrity

Related topics