"Holzhammermethode" - managed to rescue data off a failing NVMe in my server with a massive hammer (it helped to cool it to allow getting the data off the drive) 😀


@emanuel is the 970 Evo the one that failed, or was it another NVMe?
How did it fail and which NVMe should I not buy in order to avoid having to use a hammer?

· · Web · 2 · 0 · 0

@wolf480pl it was indeed a 970evo 2TB (the one under the Hammer :-) . The one pictured is the replacement, a 970evo plus 2TB. That said, I worked that drive hard for 2 years, it powers a 32GB RAM Server with 7 VMs running 24/7 with pretty much everything I host. Mail Server, Nextcloud, Mastodon, many more things. The drive was 70% full and still worked when cooled AND copies were artificially slowed down to <200MB/s.

@emanuel damn, I was hoping a couple of 970 Evos in RAID 1 would last me 5 years without a replacement, but looks like that may not be the case...

Were you still on warranty / terabytes written below the rated TBW endurance?

@wolf480pl once I have time, I ll try to get SMART stat on that drive and will post here.

@wolf480pl oh I forgot. It returned critical sector read errors hmm 1 for every 20GB, usually only on very large files.

Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!