[
https://issues.apache.org/jira/browse/HDFS-7748?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14372149#comment-14372149
]
Haohui Mai commented on HDFS-7748:
----------------------------------
{code}
- public PipelineAck(long seqno, int[] replies,
- long downstreamAckTimeNanos) {
- ArrayList<Integer> replyList = Lists.newArrayList();
- for (int r : replies) {
- replyList.add(r);
- }
- proto = PipelineAckProto.newBuilder()
+ public PipelineAck(long seqno, Status[] replies,
+ long downstreamAckTimeNanos,
+ int flag) {
+
+ PipelineAckProto.Builder protoBuilder = PipelineAckProto.newBuilder()
.setSeqno(seqno)
- .addAllReply(replyList)
- .setDownstreamAckTimeNanos(downstreamAckTimeNanos)
- .build();
+ .addAllStatus(Arrays.asList(replies))
+ .setDownstreamAckTimeNanos(downstreamAckTimeNanos);
+
+ if (flag != Flags.ECN_DISABLED_VALUE) {
+ protoBuilder.setFlags(0,flag);
+ }
{code}
Note that the ECN flag needs to be preserved through the whole pipeline --
which means that the array should contain the ECN status for all nodes that are
involved in the pipeline.
> Separate ECN flags from the Status in the DataTransferPipelineAck
> -----------------------------------------------------------------
>
> Key: HDFS-7748
> URL: https://issues.apache.org/jira/browse/HDFS-7748
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Haohui Mai
> Assignee: Anu Engineer
> Priority: Blocker
> Attachments: hdfs-7748.001.patch, hdfs-7748.002.patch,
> hdfs-7748.003.patch, hdfs-7748.004.patch
>
>
> Prior to the discussions on HDFS-7270, the old clients might fail to talk to
> the newer server when ECN is turned on. This jira proposes to separate the
> ECN flags in a separate protobuf field to make the ack compatible on both
> versions.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
