Alexander Motin

@Haravikk Scrub checkpoint does not include "which sector(s) were most recently scrubbed for each disk". It includes pool-wide bookmark (objset, object, level and offset) it scrubbed up to. It came...

Scrub restart behavior depends on the amount of memory. The more memory system has, the more it can accumulate block pointers for scan, the more time it will take to...

As Brian noted above, z_is_stale need to be garbage collected as no longer user. Otherwise it looks OK to me, though somewhere from stratosphere level, as I don't know much...

You should not wish 256KB ashift. It would be too space inefficient in most cases. Just recently my https://github.com/openzfs/zfs/pull/13798 was merged to master to improve this area. I think it...

@tonyhutter From that perspective I think it would be reasonable for ZED to just not make too quick extra movements like kicking in spares for few minutes after disk disappeared...

Neither of RAIDZ variants are good for 4KB writes, especially if you have ashift=12, independently of "Hot parity disks" "feature". You could consider dedicated special vdev for dedup data. Or...

Are you sure that the 2.2GB are in ARC, not dbuf cache? dbuf cache unlike ARC is not compressed and micro zap structures used for small directories use 128KB blocks....

As I see arc_dnode_size_limit limit is respected only during ARC data eviction phase. If your system has enough RAM, it does not apply.

I agree that quota check does not make much sense for ZVOLs with reservation, where you should not be able to create ZVOL above the parent dataset quota or snapshot...

@malventano Do you have some reason to think it is related other than the set parameter? BTW 884364ea85423fdd431f6269c42228ab74c4846d recently (in 2.1.6-staging now) further improved the prefetcher logic to use bigger...