Allan Jude

Is this a known test failure? ``` Test: /usr/share/zfs/zfs-tests/tests/functional/cli_root/zpool_create/zpool_create_features_009_pos (run as root) [00:23] [FAIL] 23:57:21.47 ASSERTION: verify '-o compatibility' in MOS object and cache file 23:57:21.69 SUCCESS: zpool create -f...

> > > @allanjude it's a recently added test case, so we have less mileage on it, but I don't recall seeing it fail in the CI before. See [88a4833](https://github.com/openzfs/zfs/commit/88a4833039b4a3f08139c5b69a2300424fddfd0f)....

> Please say this PR isn't abandoned. ZFS locking down all zpools on an entire system due to a single pool being suspended should be the most critical problem to...

> This does seem better than the mess I was about to make of INGLOBALZONE. > > I do not like the idea that a jail can mount snapshots of...

KlaraSystems.com has done some work similar to this in the past, around dealing with DDT ZAPs that are too large, or sparsely populated like this. If you are interested in...

If you look at the arcstats, what is the size of the 'other' portion of the ARC? It may be that you have a very large amount of dirty data,...

The stat I was looking for was the `anon_size` from `/proc/spl/kstat/zfs/arcstats`, which will be the amount of dirty data. It will change as writes come in, and are flushed one.

`zpl_show_devname()` acquires the teardown lock for the mountpoint, is something else also holding that lock when you're trying to `cat /proc/mounts`?

Right, the question is what other process is holding that lock, preventing cat from being able to get it.

> 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...