RFR: 8326498: java.net.http.HttpClient connection leak using http/2 [v3]

[Jaikiran Pai]

On Wed, 12 Nov 2025 09:08:43 GMT, Volkan Yazici <vyazici at openjdk.org> wrote:

>> Jaikiran Pai has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains 12 additional commits since the last revision:
>> 
>>  - rename isErroneousClose() to isAbnormalClose()
>>  - stackless instance for idle timed out NoError
>>  - Volkan's suggestion - log message improvement
>>  - merge latest from master branch
>>  - mark jdk.internal.net.http.Http2Connection as Closable
>>  - reduce number of concurrent requests
>>  - cleanup
>>  - update test
>>  - Return false from isOpen() if underlying channel is not open
>>  - Daniel's review suggestion - stop the scheduler when TubeSubscriber errored or completed
>>  - ... and 2 more: https://git.openjdk.org/jdk/compare/3b9f58c2...1c3f73ba
>
> src/java.net.http/share/classes/jdk/internal/net/http/Http2Connection.java line 234:
> 
>> 232:                 if (cancelled) {
>> 233:                     if (debug.on()) {
>> 234:                         debug.log("Not initiating idle connection close");
> 
> Suggestion:
> 
>                         debug.log("Connection is already cancelled, skipping idle connection close");

I updated the PR with a slightly different message - it's not the connection which is cancelled but it is the idle timeout event which is cancelled.

> src/java.net.http/share/classes/jdk/internal/net/http/Http2TerminationCause.java line 144:
> 
>> 142:      */
>> 143:     public static Http2TerminationCause idleTimedOut() {
>> 144:         return new NoError("HTTP/2 connection idle timed out", "idle timed out");
> 
> Any particular reason this is not cached in `NoError.IDLE_TIMED_OUT` in a similar manner to `NoError.INSTANCE`? I could not see a place where its stack trace would be of value.

You are right, the idle timed out instance could also do away with the stacktrace. I have updated the PR with this change.

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/28233#discussion_r2533799723
PR Review Comment: https://git.openjdk.org/jdk/pull/28233#discussion_r2533797755