7026262: HttpServer: improve handling of finished HTTP exchanges #13157

djelinski · 2023-03-23T11:33:32Z

This patch fixes the following issues with HttpHandler exception handling:

The connection is no longer closed if an exception is thrown after the response is completed. As soon as the response is completed, the connection is added back to the pool, ready for reuse. Closing the connection would race with the subsequent handler.
The stream returned by getResponseBody is now usable even if the response has zero-length body. Writing any data to the stream will still fail, but zero-length writes and closing the stream will now succeed.
ServerImpl.Exchange.reject now sends a Connection: close header in addition to closing the connection

The test B8293562 had to be adjusted to avoid Connection: close.
Additionally, while I was looking for a good test to copy from, I found and fixed a bug in test B5045306.

The new tests are passing with this patch, failing without it. Tier 1-3 clean.

Progress

Change must be properly reviewed (1 review required, with at least 1 Reviewer)
Change must not contain extraneous whitespace
Commit message must refer to an issue

Issue

JDK-7026262: HttpServer: improve handling of finished HTTP exchanges

Reviewers

Daniel Fuchs (@dfuch - Reviewer)
Michael McMahon (@Michael-Mc-Mahon - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/13157/head:pull/13157
$ git checkout pull/13157

Update a local copy of the PR:
$ git checkout pull/13157
$ git pull https://git.openjdk.org/jdk.git pull/13157/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 13157

View PR using the GUI difftool:
$ git pr show -t 13157

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/13157.diff

bridgekeeper · 2023-03-23T11:35:19Z

👋 Welcome back djelinski! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2023-03-23T11:37:51Z

@djelinski The following label will be automatically applied to this pull request:

net

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.

mlbridge · 2023-03-23T15:58:50Z

Webrevs

01: Full - Incremental (6a7220a1)
00: Full (35731882)

src/jdk.httpserver/share/classes/sun/net/httpserver/ServerImpl.java

test/jdk/sun/net/www/http/KeepAliveCache/B5045306.java

dfuch · 2023-03-23T18:34:37Z

src/jdk.httpserver/share/classes/sun/net/httpserver/FixedLengthOutputStream.java

+        if (zerolength) {
+            // WriteFinishedEvent was sent already; nothing to do
+            return;
+        }


Wouldn't it be more correct to have sendResponseHeader close the output stream, and let the output stream close generate the write event, rather than having sendResponseHeader generate the event and not doing anything here? Or would there be an issue with the case where the stream is chunked output stream?

sendResponseStream could close the stream when called with length = -1; the output stream is always fixed length in that case. I'd need to modify this class to accept zero-length writes even when closed, but I think there's a precedent for that already. Will look into it.

That was just a suggestion - I was a bit bothered with the hidden interaction between FixedLengthOutputStream and sendResponseHeaders here, where FixedLengthOutputStream has to trust that sendResponseHeaders has sent the event. It looked strange. Maybe there are other possibilities - like e.g a ZeroLengthOutputStream :-)

I changed sendResponseHeaders to close the exchange, and I like the result much better.
ExceptionKeepAlive test no longer needs to explicitly close the request body before sending the response - exchange's close takes care of that.
Let me know what you think.

I wonder should we move the code that sends the write finished event in this case to here? I agree this code looks pretty strange. Never mind. I see this has changed in the new revision.

dfuch · 2023-03-23T18:39:22Z

src/jdk.httpserver/share/classes/sun/net/httpserver/FixedLengthOutputStream.java

-        eof = (remaining == 0);
-        if (eof) {
-            throw new StreamClosedException();
-        }


Maybe before checking closed we should just return if len == 0? (and throw if < 0 BTW)
Possibly consider doing the same for ChunkedOutputStream.
Also ChunkedOutputStream throws StreamClosedException() when closed is true. Not sure why we're throwing plain IO here. Or rather - not sure why we have StreamClosedException() in the first place, it doesn't seem to be handled specially anywhere, and it can reach user code (code calling write()), and it's not a public exception - so maybe it should be removed altogether and replaced with IOException("stream closed") everywhere? But getting rid of StreamClosedException may be a task for another PR.

+1. Filed JDK-8304873 to remove StreamClosedException.

dfuch

That looks much cleaner. It would be good to get @Michael-Mc-Mahon review too. But if all tests are passing and stable, it sounds good to me.

openjdk · 2023-03-24T11:08:12Z

@djelinski This change now passes all automated pre-integration checks.

ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details.

After integration, the commit message for the final commit will be:

7026262: HttpServer: improve handling of finished HTTP exchanges

Reviewed-by: dfuchs, michaelm

You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed.

At the time when this comment was updated there had been 33 new commits pushed to the master branch:

d8ba227: 8304069: ClassFileParser has ad-hoc hashtables
9a8a60f: 8304833: (fc) Remove dead code in sun.nio.ch.FileChannelImpl::implCloseChannel
f96aee7: 8291154: Create a non static nested class without enclosing class throws VerifyError
4ec720d: 8297977: vmTestbase/nsk/stress/except/except012.java fails with unexpected Exception
13dd19a: 8304802: After JDK-8297639 the flag G1UsePreventiveGC needs to be added to the obsoletion table
d61de14: 8303508: Vector.lane() gets wrong value on x86
941a7ac: 8304301: Remove the global option SuperWordMaxVectorSize
ac6af6a: 7176515: ExceptionInInitializerError for an enum with multiple switch statements
dd23ee9: 8303917: Update ISO 639 language codes table
6f67abd: 8304557: java/util/concurrent/CompletableFuture/CompletableFutureOrTimeoutExceptionallyTest.java times out
... and 23 more: https://git.openjdk.org/jdk/compare/ddf1e34c1a0815e8677212f1a7860ca7cf9fc2c9...master

As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this automatic rebasing, please check the documentation for the /integrate command for further details.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

Michael-Mc-Mahon

Looks fine to me

Michael-Mc-Mahon · 2023-03-24T15:23:39Z

src/jdk.httpserver/share/classes/sun/net/httpserver/ChunkedOutputStream.java

+        Objects.checkFromIndexSize(off, len, b.length);
+        if (len == 0) {
+            return;
+        }
        if (closed) {


Initially I didnt think the above was a great idea as OutputStream.write is spec'd to throw IOE if the stream is closed, but it appears that some commonly used OutputStream classes are tolerant of zero length writes

djelinski · 2023-03-27T05:20:40Z

/integrate

openjdk · 2023-03-27T05:22:59Z

Going to push as commit a5ffa07.
Since your change was applied there have been 40 commits pushed to the master branch:

65e01da: 8304013: Add a fast, non-manual alternative to test/jdk/java/util/zip/ZipFile/TestTooManyEntries
38e1714: 8304258: x86: Improve the code generation of VectorRearrange with int and float
765a942: 8304136: Match allocation and free in sspi.cpp
3f59b75: 8304898: Fix Copyright Headers for JLink Source Files
501b606: 8298725: Add BitMap support for reverse iteration
9764948: 8273986: JEditorPane HTML Demo - Accessibility issues
5727610: 8304353: Add lib-test tier1 testing in GHA
d8ba227: 8304069: ClassFileParser has ad-hoc hashtables
9a8a60f: 8304833: (fc) Remove dead code in sun.nio.ch.FileChannelImpl::implCloseChannel
f96aee7: 8291154: Create a non static nested class without enclosing class throws VerifyError
... and 30 more: https://git.openjdk.org/jdk/compare/ddf1e34c1a0815e8677212f1a7860ca7cf9fc2c9...master

Your commit was automatically rebased without conflicts.

openjdk · 2023-03-27T05:23:16Z

@djelinski Pushed as commit a5ffa07.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.

djelinski added 3 commits March 22, 2023 21:05

OutputStream should be closeable even when response is zero-length

ae97817

Fix test

16cf8e2

Do not close connection

6e2d75e

openjdk bot changed the title ~~7026262~~ 7026262: HttpServer: improve handling of finished HTTP exchanges Mar 23, 2023

openjdk bot added the net net-dev@openjdk.org label Mar 23, 2023

Fix broken test

3573188

djelinski marked this pull request as ready for review March 23, 2023 15:50

openjdk bot added the rfr Pull request is ready for review label Mar 23, 2023

dfuch reviewed Mar 23, 2023

View reviewed changes

djelinski added 3 commits March 24, 2023 08:31

Remove redundant closeConnection

84bebb3

Close exchange on zero-length response

3206e4a

Explicit close no longer needed

6a7220a

dfuch approved these changes Mar 24, 2023

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Mar 24, 2023

Michael-Mc-Mahon approved these changes Mar 24, 2023

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Mar 27, 2023

openjdk bot closed this Mar 27, 2023

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Mar 27, 2023

djelinski deleted the server-exceptions branch March 27, 2023 05:51

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

7026262: HttpServer: improve handling of finished HTTP exchanges #13157

7026262: HttpServer: improve handling of finished HTTP exchanges #13157

djelinski commented Mar 23, 2023 •

edited by openjdk bot

Loading

bridgekeeper bot commented Mar 23, 2023

openjdk bot commented Mar 23, 2023

mlbridge bot commented Mar 23, 2023 •

edited

Loading

dfuch Mar 23, 2023

djelinski Mar 24, 2023

dfuch Mar 24, 2023

djelinski Mar 24, 2023

Michael-Mc-Mahon Mar 24, 2023 •

edited

Loading

dfuch Mar 23, 2023

djelinski Mar 24, 2023

dfuch left a comment

openjdk bot commented Mar 24, 2023 •

edited

Loading

Michael-Mc-Mahon left a comment

Michael-Mc-Mahon Mar 24, 2023

djelinski commented Mar 27, 2023

openjdk bot commented Mar 27, 2023

openjdk bot commented Mar 27, 2023

7026262: HttpServer: improve handling of finished HTTP exchanges #13157

7026262: HttpServer: improve handling of finished HTTP exchanges #13157

Conversation

djelinski commented Mar 23, 2023 • edited by openjdk bot Loading

Progress

Issue

Reviewers

Reviewing

bridgekeeper bot commented Mar 23, 2023

openjdk bot commented Mar 23, 2023

mlbridge bot commented Mar 23, 2023 • edited Loading

Webrevs

dfuch Mar 23, 2023

Choose a reason for hiding this comment

djelinski Mar 24, 2023

Choose a reason for hiding this comment

dfuch Mar 24, 2023

Choose a reason for hiding this comment

djelinski Mar 24, 2023

Choose a reason for hiding this comment

Michael-Mc-Mahon Mar 24, 2023 • edited Loading

Choose a reason for hiding this comment

dfuch Mar 23, 2023

Choose a reason for hiding this comment

djelinski Mar 24, 2023

Choose a reason for hiding this comment

dfuch left a comment

Choose a reason for hiding this comment

openjdk bot commented Mar 24, 2023 • edited Loading

Michael-Mc-Mahon left a comment

Choose a reason for hiding this comment

Michael-Mc-Mahon Mar 24, 2023

Choose a reason for hiding this comment

djelinski commented Mar 27, 2023

openjdk bot commented Mar 27, 2023

openjdk bot commented Mar 27, 2023

djelinski commented Mar 23, 2023 •

edited by openjdk bot

Loading

mlbridge bot commented Mar 23, 2023 •

edited

Loading

Michael-Mc-Mahon Mar 24, 2023 •

edited

Loading

openjdk bot commented Mar 24, 2023 •

edited

Loading