[Bug 203877] NFS threads get blocked when writing to ZFS dataset that has reached quota

bugzilla-noreply at freebsd.org bugzilla-noreply at freebsd.org
Mon Oct 19 18:43:41 UTC 2015


https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=203877

            Bug ID: 203877
           Summary: NFS threads get blocked when writing to ZFS dataset
                    that has reached quota
           Product: Base System
           Version: 10.1-RELEASE
          Hardware: amd64
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: kern
          Assignee: freebsd-bugs at FreeBSD.org
          Reporter: wollman at FreeBSD.org

Figured it was about time to actually get this into the bug database, since
I've made no progress figuring it out.

We've seen this bug for a long time, since at least 9.2.  Given a set of
clients (in a cluster) that write logs to a file or a small set of files in NFS
(v3), ignoring write errors, and a FreeBSD NFS server using ZFS as a backing
store, with the particular ZFS dataset being below quota when the files are
initially opened for write, but reaching its quota while the clients are
actively writing, eventually something inside ZFS slows to a crawl
(synchronized to txg openings, perhaps?).  When this happens, all of the NFS
service threads eventually get tied up deep in ZFS and no new requests can be
processed from any client, even the non-misbehaving ones.  FHA makes this
happen faster, but it will still happen even with FHA disabled.

-- 
You are receiving this mail because:
You are the assignee for the bug.


More information about the freebsd-bugs mailing list