Public bug reported: Livelock between ZFS evict and writeback threads
[Impact] ZIO pipeline stalls, causing ZFS workloads to hang indefinitely [Description] For certain ZFS workloads, we start seeing hung task timeouts in the kernel logs due to zil_commit() stalling. This is due to zfs_zget() not detecting whether a znode has been marked for deletion before attempting to access it, causing a constant "retry loop" in zfs_get_data() if that znode has been unlinked already. An example of the stack traces follows: [72742.051703] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [72742.070429] mysqld D 0 5713 2881 0x00000320 [72742.073220] Call Trace: [72742.075305] __schedule+0x24e/0x880 [72742.090436] schedule+0x2c/0x80 [72742.090438] schedule_preempt_disabled+0xe/0x10 [72742.090441] __mutex_lock.isra.5+0x276/0x4e0 [72742.090547] ? dmu_tx_destroy+0x105/0x130 [zfs] [72742.090555] __mutex_lock_slowpath+0x13/0x20 [72742.115374] ? __mutex_lock_slowpath+0x13/0x20 [72742.132266] mutex_lock+0x2f/0x40 [72742.134207] zil_commit_impl+0x1b0/0x1b30 [zfs] [72742.150428] ? spl_kmem_alloc+0x115/0x180 [spl] [72742.152622] ? mutex_lock+0x12/0x40 [72742.154819] ? zfs_refcount_add_many+0x9a/0x100 [zfs] [72742.171450] zil_commit+0xde/0x150 [zfs] [72742.173687] zfs_fsync+0x77/0xe0 [zfs] [72742.175044] zpl_fsync+0x80/0x110 [zfs] [72742.191690] vfs_fsync_range+0x51/0xb0 [72742.193876] do_fsync+0x3d/0x70 [72742.195126] SyS_fsync+0x10/0x20 [72742.211059] do_syscall_64+0x73/0x130 [72742.214078] entry_SYSCALL_64_after_hwframe+0x3d/0xa2 It's possible to hit this issue due to a race between the ZFS evict and writeback threads. If the z_iput task is trying to evict a znode that's currently sitting in the writeback thread, both will "livelock" each other and stall the ZIO pipeline, causing other ZFS operations (such as zil_commit) to hang indefinitely. This has been documented and fixed upstream in PR#9583 [0]. We need to pull two fixes from upstream: the first one fixes the zfs_zget() issue in the writeback thread, while the second fixes a regression on O_TMPFILE descriptors caused by the first one. Upstream patches: - Break out of zfs_zget early if unlinked znode (41e1aa2a06f8) - Check for unlinked znodes after igrab() (0c46813805f4) [Test Case] Being a race condition, this issue has been hard to reproduce consistently. The racing window between evict() and the ZFS writeback thread is quite strict, but users have reported this to show up after some hours of running LXD-containerized mySQL workloads. [Regression Potential] These patches have been tested both in the ZFS test suite and in production environments, so the potential for further regressions should be low. Additional regressions would likely cause issues with the ZFS writeback/commit and IO pipeline, so they should be spotted fairly quickly. [0] https://github.com/zfsonlinux/zfs/pull/9583 [1] https://github.com/zfsonlinux/zfs/commit/41e1aa2a06f8 [2] https://github.com/zfsonlinux/zfs/commit/0c46813805f4 ** Affects: zfs-linux (Ubuntu) Importance: Undecided Assignee: Heitor Alves de Siqueira (halves) Status: Confirmed ** Affects: zfs-linux (Debian) Importance: Unknown Status: Unknown ** Tags: sts ** Bug watch added: Debian Bug tracker #946610 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=946610 ** Also affects: zfs-linux (Debian) via https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=946610 Importance: Unknown Status: Unknown -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1856084 Title: Livelock between ZFS evict and writeback threads To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/1856084/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs