Hi Martin, > > Afterwards, on 8 out of 12 SSDs, root filesystem was corrupted and > > remounted read-only as we are using the "errors=remount-ro" option. > When you say 8 out of 12 you mean that these drives were used in > different systems (i.e. no RAID)? The 12 disks are indeed in 12 different servers, each attached directly to the SATA controller without any RAID. > > Any idea why this would happen? Could this be related to > > https://patchwork.kernel.org/patch/10310641 from March 27? > It sounds likely. That's pretty awful news. I don't have an 860 myself > but I did some due (Google) diligence before acking the patch. > Does it reproduce if you add libata.force=noncqtrim on the kernel > command line? We will soon receive another Samsung 860 Pro SSD in order to make the test you suggested. We will keep you informed. Thanks & regards, Bertrand Blackburn -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
