[
https://issues.apache.org/jira/browse/HADOOP-18146?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17641281#comment-17641281
]
ASF GitHub Bot commented on HADOOP-18146:
-----------------------------------------
steveloughran commented on code in PR #4039:
URL: https://github.com/apache/hadoop/pull/4039#discussion_r1035926910
##########
hadoop-tools/hadoop-azure/src/main/java/org/apache/hadoop/fs/azurebfs/services/AbfsHttpOperation.java:
##########
@@ -314,18 +314,21 @@ public void sendRequest(byte[] buffer, int offset, int
length) throws IOExceptio
if (this.isTraceEnabled) {
startTime = System.nanoTime();
}
- try (OutputStream outputStream = this.connection.getOutputStream()) {
- // update bytes sent before they are sent so we may observe
- // attempted sends as well as successful sends via the
- // accompanying statusCode
- this.bytesSent = length;
+ OutputStream outputStream;
+ try {
+ try {
+ outputStream = this.connection.getOutputStream();
+ } catch (IOException e) {
+ // If getOutputStream fails with an exception due to 100-continue
Review Comment:
i still dont like this.
there are a lot of reasons for failure, given this is the call which tries
to set up the entire http request. if it is only expect 100 errors then
1. handler needs to identify those and throw everything else
2. you need a strategy for downgrading
as it is, it looks to me that if you have any problem talking to the far end
then the method will silently swallow it, while updating the bytes sent counter
> ABFS: Add changes for expect hundred continue header with append requests
> -------------------------------------------------------------------------
>
> Key: HADOOP-18146
> URL: https://issues.apache.org/jira/browse/HADOOP-18146
> Project: Hadoop Common
> Issue Type: Sub-task
> Affects Versions: 3.3.1
> Reporter: Anmol Asrani
> Assignee: Anmol Asrani
> Priority: Major
> Labels: pull-request-available
> Time Spent: 2h 10m
> Remaining Estimate: 0h
>
> Heavy load from a Hadoop cluster lead to high resource utilization at FE
> nodes. Investigations from the server side indicate payload buffering at
> Http.Sys as the cause. Payload of requests that eventually fail due to
> throttling limits are also getting buffered, as its triggered before FE could
> start request processing.
> Approach: Client sends Append Http request with Expect header, but holds back
> on payload transmission until server replies back with HTTP 100. We add this
> header for all append requests so as to reduce.
> We made several workload runs with and without hundred continue enabled and
> the overall observation is that :-
> # The ratio of TCP SYN packet count with and without expect hundred continue
> enabled is 0.32 : 3 on average.
> # The ingress into the machine at TCP level is almost 3 times lesser with
> hundred continue enabled which implies a lot of bandwidth save.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org
