The possibility for inconsistent IO was already fixed under [1] a year ago, but we left the "rbd_op_threads" hard-coded to 1 due to other unrelated race conditions as you are discovering. Fixing those race conditions is being tracked under a different ticket [2]. [1] http://tracker.ceph.com/issues/15034 [2] http://tracker.ceph.com/issues/17379 On Thu, Oct 12, 2017 at 11:26 PM, 郝赏 <hs170779@xxxxxxx> wrote: > now i have also encountered the same problem as #15034 with 5 > rbd_op_threads, it caused some assert failures，why does this happen ？ and， > how can we avoid this problem by guaranteeing that the data written in the > client cache is written in the same order as in the journal log？ > > > 【网易自营】好吃到爆！鲜香弹滑加热即食，经典13香/麻辣小龙虾仅75元3斤>> -- Jason -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
