I have bought the X670E Pro RS recently and added some ECC memories to it. Now I wanted to enable memory scrubbing, however when I look in /sys/devices/system/edac/mc/mc0/ the sdram_scrub_rate is missing. I can't seem to find an answer as to why online. I changed the ECC from auto to enable just to force it to be enabled. lsmod | fgrep edac reports: amd64_edac 69632 0 edac_mce_amd 40960 1 amd64_edac Which is correct, yes(?). Am I missing a driver, or maybe I missed that edac-utils is required. Why is sdram_scrub_rate missing is the question.

I use btrfs atop of bcache. It caches small reads and that works great. When running a scrub of the btrfs however, I'd like those reads to go through to the backing devices because their contents are the ground truth; what I want to verify. One way to achieve that would be to detach the individual bcache devices from the bcache but doing so necessarily also detaches the bcache state which means it gets lost. It'd have to rebuild the cache every time I run a scrub. Is it possible to somehow achieve that without losing the cache? I want bcache to keep doing all of its regular state tracking; updating and invalidating the cache. When it comes to utilising the cache though, it should simply read directly from the backing drives and ignore any cached data.

My NAS is running zfs and does a scrub now and then. When it runs it read all data (as expected), but it also writes a litte during the run (around 0.5%): ~~~ Device rkB/s wkB/s %util sda 13628.80 69.60 58.04 sdb 13625.60 73.60 59.48 sdc 13625.60 70.80 58.96 sdd 13625.60 68.40 58.24 sde 13625.60 69.60 60.00 ~~~ This could be explained by other systems using it. But there are no systems using the NAS, and no user processes running on the server, that would access the NAS. Also the rate is constant for hours: Full read for 4.5 seconds followed by 0.5 seconds write. It is a fixed rythm: silence for 4.5 seconds, and then the DRRDRRD sound of seeking when it is writing. But why does it write *anything* when there are no clients and nothing to correct? I was under the impression that as long as all data are clean, consistent, and unchanged since last scrub, there would be no need to write any data.

I've recently had problems with a faulty power supply to a Raspberry PI system used as a media center. Before tracking it and replacing with a new power unit it seems it did some damages to an old disk (that could be troubled by itself) and to a not so old disk that is physically ok. It's formatted with btrfs and scrub gives errors in "super": # btrfs scrub status -d /dev/sda1 scrub status for ************************** scrub device /dev/sda1 (id 1) history scrub resumed at Mon Feb 1 22:55:03 2021 and finished after 01:44:48 total bytes scrubbed: 160.77GiB with 2 errors error details: super=2 corrected errors: 0, uncorrectable errors: 0, unverified errors: 0 btrfs check doesn't find any problem: # btrfs check --progress /dev/sda1 Opening filesystem to check... Checking filesystem on /dev/sda1 UUID: ************************** [1/7] checking root items (0:00:00 elapsed, 36660 items checked) [2/7] checking extents (0:00:26 elapsed, 16745 items checked) [3/7] checking free space cache (0:00:06 elapsed, 240 items checked) [4/7] checking fs roots (0:00:32 elapsed, 684 items checked) [5/7] checking csums (without verifying data) (0:00:29 elapsed, 35432 items checked) [6/7] checking root refs (0:00:00 elapsed, 3 items checked) [7/7] checking quota groups skipped (not enabled on this FS) found 252808790016 bytes used, no error found total csum bytes: 246555120 total tree bytes: 274219008 total fs tree bytes: 11255808 total extent tree bytes: 3080192 btree space waste bytes: 11029386 file data blocks allocated: 252534571008 referenced 252534571008 Now, given that I do have a full backup and an automated script to sync to the media center, so the less troubling thing would be to reformat and populate the disk, I think it's a good context to make some experiment. So: is there any way to attempt a fix? Thanks.

I understand the both scrubbing and balancing of a BRTFS volume is useful for maintaining its health. Which one is more useful though? If I cannot afford to run both types of checks regularly, should I prefer the one over the other?

Scrubbing a BTRFS volume seems to be an expensive operation. How often should I be running it on a moderately busy system? Dayly? Weekly? Monthly?

I need to scrub a couple of very large HDDs. However, I can't do that from a desktop. I need to do it on the move, from a laptop. A single pass of scrub(1) on a HDD would take more than a day, but there is no way I can leave my laptop stationary for that long. scrub(1) itself doesn't support any kind of offset command line parameter. Is there a way to do what scrub(1) does (writing random bytes), but in a way that can be resumed? Basically the command would need to print out the offset when I interrupt it, and it needs to accept an offset parameter for resuming.