Alexander Motin

> It also leads to some confusing asymmetries, eg the DVA throttle is held in the logical, and then released in pieces in the children. It might not look great,...

@Crest All blocks in ZFS are full blocks of record size, unless it is a file of only one block, in which case it might be any multiple of 1

@robn I think once user requested condense, we should do it as fast as possible, since user likely waits for it to reboot, export, whatever. This operation makes no sense...

@tonyhutter Do you see anything wrong about that patch other than it makes something else to misbehave? If you have some reproduction, then we should try to collect maximum details...

I'm sorry, Tony, but how do you expect to see a correlation between user file writes and ZFS disk writes, considering the second are issued asynchronously by sync threads? If...

Ok. Thanks. Just without explicit txg commits and with delays between files and disk writes how do we know it is not a coincedence? I'll try to play with it,...

@scratchings Have you checked to what replication process 594865 triggering the panic belongs? Is it remote raw/encrypted or local unencrypted one? The error 5 returned by `dmu_bonus_hold_by_dnode()` is EIO. But...

> it seems that maybe it's not that harmless marking as evictable, but perhaps it's also enabling some extra unexpected eviction, at least for znodes? @snajpa I don't believe KMC_RECLAIMABLE...

Appears even CI itself uses some. We need to update `tunables.cfg`.

I've just fixed couple things (systat and top) in FreeBSD base still using old names: https://cgit.freebsd.org/src/commit/?id=8aad1e6148d3389df100bb0391e2d3a909f26ecf .