I've run hast synchronization from primary to secondary. Within some time of synchronization run zpool status zroot. zpool shows status of zroot and after some seconds hast syncronization stops. In that some seconds only 6-10Mb writes to primary device. I've see that in dirty section from "hastctl list". Any write operations to zfs pool stops too. I can't mount zfs filesystem datasets anymore. Only ACPI button poweroff or Ctrl+Alt+Delete works with some small timeouts. Fix: vfs.zfs.vdev.bio_delete_disable=1 vfs.zfs.vdev.bio_flush_disable=1 fix my problem. May be only one of them or vfs.zfs.trim.enabled=0. Didn't check, sorry. How-To-Repeat: - create hast device on primary - create ZFS pool on primary hast device - create hast device on secondary - run synchronization (memory type) from primary to secondary - make some write operations on primary ZFS pool: create datasets, mount datasets, write some files.
Responsible Changed From-To: freebsd-bugs->freebsd-fs Over to maintainer(s).
I have failed to reproduce the issue. So could you give more details? It would be nice to see your configs and logs from both primary and secondary, and the sequence of commands that led to the hang. Also, it would be very nice to find out exactly what option (or combination) caused the issue. -- Mikolaj Golub
This is a multi-part message in MIME format.
For bugs matching the following criteria: Status: In Progress Changed: (is less than) 2014-06-01 Reset to default assignee and clear in-progress tags. Mail being skipped