News for hackers. Stuff that matters.

A comment on I *detest* the crazy industry politics that made ECC memory so “special”. Parent comment

Re:I *detest* the crazy industry politics that made ECC memo... by bpye on Sunday October 16, 2022 @01:04PM (#33225218)
On ZFS, I don't think there's any reason why it needs ECC more than any other filesystem [0].

[0] - https://jrs-s.net/2015/02/03/will-zfs-and-non-ecc-ram-kill-y...

[ Parent | on HN ]
Re:I *detest* the crazy industry politics that made ECC memo... by belter on Sunday October 16, 2022 @01:08PM (#33225262)
It looks like ZFS should be run with ECC

"...We further demonstrate that ZFS is less resilient to memory corruption, which can lead to corrupt data being returned to applications or system crashes..." - https://research.cs.wisc.edu/wind/Publications/zfs-corruptio...

"Please Use ZFS With ECC Memory" - https://louwrentius.com/please-use-zfs-with-ecc-memory.html

[ Parent | on HN ]
Re:I *detest* the crazy industry politics that made ECC memo... by PBondurant on Sunday October 16, 2022 @01:44PM (#33225548)
A fuller instance of the first quote: "Through careful and thorough fault injection, we show that ZFS is robust to a wide range of disk faults. We further demonstrate that ZFS is less resilient to memory corruption, which can lead to corrupt data being returned to applications or system crashes"

...ie, ZFS is 'less resilient' in comparison to its robust disk fault handling, not that it's less resilient to memory corruption in comparison to other filesystems. The parent quotation above implies that ZFS is more sensitive to memory corruption than other fs but that is not claimed in the referenced paper.

[ Parent | on HN ]
Re:I *detest* the crazy industry politics that made ECC memo... by puffoflogic on Sunday October 16, 2022 @01:07PM (#33225249)
Presumably zfs is the most-mainstream fs which will actually notice and complain about in-memory bit flips.
[ Parent | on HN ]