8276798: HttpURLConnection sends invalid HTTP request

[jaikiran]

Can I please get a review of this change which proposes to fix https://bugs.openjdk.java.net/browse/JDK-8276798?

sun.net.www.protocol.http.HttpURLConnection has a (private) writeRequests method. This method is responsible for creating the standard HTTP request headers (include the request line) and then writing it out to the OutputStream which communicates with the HTTP server. While writing out these request headers, if there's an IOException, then HttpURLConnection marks a failedOnce flag to true and attempts to write these again afresh (after creating new connection to the server). This re-attempt is done just once.

As noted in the linked JBS issue, specifically this comment https://bugs.openjdk.java.net/browse/JDK-8276798?focusedCommentId=14500074&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14500074, there's a specific case where creating and writing out these request headers ends up skipping the request line, which causes the server to respond back with a "Bad Request" response code.

The commit in this PR removes the use of failedOnce flag that was being used to decide whether or not to write the request line. The existing code doesn't have any specific comments on why this check was there in first place, nor does the commit history show anything about this check. However, reading through that code, my guess is that, it was there to avoid writing the request line twice when the same requests object gets reused during the re-attempt. I think a better check would be the see if the requests already has the request line and if not add it afresh.
While in this code, I also removed the check where the failedOnce flag was being used to check if the Connection: Keep-Alive/Proxy-Connection: Keep-alive header needs to be set. This part of the code already has a call to setIfNotSet, so I don't think we need the failedOnce check here.

tier1, tier2 and tier3 tests have passed without issues. However, given the nature of this code, I'm not too confident that we have tests which can catch this issue (and adding one isn't easy), so I would like inputs on whether this change is good enough or whether it has the potential to cause any non-obvious regressions.

---

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-8276798: HttpURLConnection sends invalid HTTP request

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 pull/9038/head:pull/9038
$ git checkout pull/9038

Update a local copy of the PR:
$ git checkout pull/9038
$ git pull https://git.openjdk.org/jdk pull/9038/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 9038

View PR using the GUI difftool:
$ git pr show -t 9038

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/9038.diff

[bridgekeeper]

👋 Welcome back jpai! 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]

@jaikiran 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]

Webrevs

• 00: Full (792947cd)

[jaikiran]

tier1, tier2 and tier3 tests have passed without issues.

Additionally, the reproducer attached to the JBS issue has been run locally. The test fails regularly without this change and hasn't failed even once after this change.

[dfuch]

This looks reasonable to me but I would like to get a second opinion. @Michael-Mc-Mahon ?

[openjdk]

@jaikiran 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:

8276798: HttpURLConnection sends invalid HTTP request

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 241 new commits pushed to the master branch:

• 270cf67: 8288752: Split thread implementation files
• d51f4f4: 8287011: Improve container information
• affbd72: 8288744: Remove tools/jlink/plugins/CompressorPluginTest.java from problemlist
• 2bf5c9a: Merge
• 70008da: 8287971: Throw exception for missing values in .jpackage.xml
• d7b43af: 8288761: SegmentAllocator:allocate(long bytesSize) not throwing IAEx when bytesSize < 0
• 834d92d: 8288754: GCC 12 fails to build zReferenceProcessor.cpp
• 198cec9: 8286103: VThreadMonitorTest fails "assert(!current->cont_fastpath() || (current->cont_fastpath_thread_state() && !interpreted_native_or_deoptimized_on_stack(current))) failed"
• 97200a7: 8278053: serviceability/jvmti/vthread/ContStackDepthTest/ContStackDepthTest.java failing in loom repo with Xcomp
• 31d981e: 8288532: additional review changes for JDK-8286830
• ... and 231 more: https://git.openjdk.org/jdk/compare/24fe8ad74cc481d18bed6896ca54a8d91c651d4a...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]

LGTM. Probably need to add noreg-hard label to bug.

[jaikiran]

Probably need to add noreg-hard label to bug.

Done.
Thank you Daniel and Michael for the reviews.

[jaikiran]

/integrate

[openjdk]

Going to push as commit 50c37f5.
Since your change was applied there have been 241 commits pushed to the master branch:

• 270cf67: 8288752: Split thread implementation files
• d51f4f4: 8287011: Improve container information
• affbd72: 8288744: Remove tools/jlink/plugins/CompressorPluginTest.java from problemlist
• 2bf5c9a: Merge
• 70008da: 8287971: Throw exception for missing values in .jpackage.xml
• d7b43af: 8288761: SegmentAllocator:allocate(long bytesSize) not throwing IAEx when bytesSize < 0
• 834d92d: 8288754: GCC 12 fails to build zReferenceProcessor.cpp
• 198cec9: 8286103: VThreadMonitorTest fails "assert(!current->cont_fastpath() || (current->cont_fastpath_thread_state() && !interpreted_native_or_deoptimized_on_stack(current))) failed"
• 97200a7: 8278053: serviceability/jvmti/vthread/ContStackDepthTest/ContStackDepthTest.java failing in loom repo with Xcomp
• 31d981e: 8288532: additional review changes for JDK-8286830
• ... and 231 more: https://git.openjdk.org/jdk/compare/24fe8ad74cc481d18bed6896ca54a8d91c651d4a...master

Your commit was automatically rebased without conflicts.

[openjdk]

@jaikiran Pushed as commit 50c37f5.

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