QUOTE(Maxieos @ Jun 17 2015, 05:16 PM)
Sorry I dont understand about this SSD problem test ,
https://blog.algolia.com/when-solid-state-d...not-that-solid/Wanted to know whether Intel 330 or 335 affected ?
thanks for this article. it's a good read, bringing me back to those old code-reading days (supervise is still one of my favorite tools to be used in linux boxes lol!)
back to the topic, tl;dr; it's about ssds from samsung (the whole 8 series), certain crucial/micron, ocz, intel, etc. being blacklisted in linux's ata driver (along with other storage types and models), because of non-compliance to standards that involves disabling queued trim on them. (from that article, un-queued trim was used and caused issues too)
imo, it's really depends on usage. those are scenarios and workloads in heavy duty linux servers running 24/7. (those lines of codes can be translated into this, in terms of ssd: in mission critical systems, don't ever use consumer ssds, mainstream or enthusiast; and don't enable trim in linux if the ssd is listed). in reality, our computing loads will certainly not reach that level. i know the key point is still about trim in linux. how many of us have trim issues before that caused data corruption? i have crucial m500, and most of the samsung 8 series in that list. i have no issues so far. (maybe because they are running windows) of course, if you're running linux in your machine, i think it's better to avoid those blacklisted drives, or just disable trim if you insisted.
as for your question, yes, at least 330 is affected. (the answer is not obvious but it's in a link from that page,
http://www.aerospike.com/docs/operations/p...ification.html)
QUOTE(em0ti0n @ Jun 17 2015, 06:20 PM)
thanks.. i will recheck on ur method.. btw, is using parted magic safe? i saw some video on youtube using it.
i am not sure about this, but if it's using the similar method that handles hdd's secure erase, then it's not safe imo.
edit: buggy link fixed.
This post has been edited by horns: Jun 18 2015, 12:15 PM