Hexiaoqiao
Hexiaoqiao
@cxzl25 Thanks involve me here. I don't get this improvement. It seems that `System.nanoTime()` in single thread then compare. Do you meet any issue here? Thanks.
Sorry for the late response. But I did not get why System.nanoTime() may return a negative number. Do you mind to offer any jdk information? Thanks. > System.nanoTime()may return a...
OK, Got it, Thanks. LGTM. +1 from me.
Committed to trunk. Thanks @ZanderXu for your contributions!
@ZanderXu Thanks for your contribution. IIRC, there is also some other precondition checks before initialized, Would you mind to give another check? such as `org.apache.hadoop.hdfs.server.datanode.DataNode#initReplicaRecovery` or `replicaHandler = datanode.data.createRbw(storageType, storageId,...
Committed to trunk. Thanks @ZanderXu for your contribution, Thanks @ayushtkn for your reviews.
@ZanderXu Thanks for report and contribution. Sorry I don't get what scenario lead this issue. Would you like to offer more information such as version deployed and how to reproduce....
Thanks for starting this proposal. I think there are still many issues for data transfer for pipeline recovery from my practice, which includes both basic function and performance. IIRC, there...
> Fortunately, at present, as long as failed exception throw to client, the client defaults to thinking that the new dn is abnormal, and will exclude it and retry transfer....
Sorry for not very clear comment. I know not it's round-robin way to pick the source node, and at third round it will pick the original node again (no matter...