Before merging this you there are a couple recent bug fixes related to large dnodes you should be aware of and may want to include in this PR.
- https://github.com/zfsonlinux/zfs/pull/6564 - There was an unfortunate interaction with OpenZFS 8199, the multi-threaded `dmu_object_alloc()`, which could lead to corruption of the dnode object when many threads are allocating concurrently. Normal workloads initially failed to uncover it but this happens to be a common Lustre workload. - https://github.com/zfsonlinux/zfs/pull/6396 - Some additional send stream validity checking needed to be added. - https://github.com/zfsonlinux/zfs/pull/6576 - Receiving an incremental stream with dnode sizes which differ between the source and the target will fail. The issue is understand and a fix is being worked on. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/409#issuecomment-327868313 ------------------------------------------ openzfs-developer Archives: https://openzfs.topicbox.com/groups/developer/discussions/Tfefe6318017dd4c7-Mf9c0f317f09c820ed172a391 Powered by Topicbox: https://topicbox.com
