a13a55e242
Introduced LeakDetector.id() to be consistent in logging the resource ID.
2015-03-07 22:32:59 +01:00
dd88d393e4
456521 - ShutdownHandler should shut down more gracefully
...
Also-by: Johannes Brodwall <johannes@brodwall.com>
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
2015-03-06 15:47:51 -07:00
2ceaf6e65c
460769 - ClientUpgradeRequest sends cookies in the wrong format
...
+ Simplifying HTTP Client "Cookie:" header based on rfcs
2015-03-06 15:19:41 -07:00
9edd7c4b30
461070 Handle setReadListener on request with no content
2015-03-07 08:44:51 +11:00
74ee154bb1
Cleaning up LeakTrackingByteBufferPool / LeakDetector
...
+ Making use of LeakTrackingByteBufferPool more consistent
+ Using MappedByteBufferPool.Tagged where appropriate in test cases
+ Adding leak count tracking to LeakDetector
+ Adding leak count tracking to LeakTrackingByteBufferPool
+ Renaming websocket LeakTrackingBufferPool to
LeakTrackingBufferPoolRule to reflect junit @Rule usage
+ Making websocket LeakTrackingBufferPoolRule always use
MappedByteBufferPool.Tagged
+ Fixed various grammar concerns
2015-03-06 14:27:58 -07:00
ac08e4ded7
461623 - BufferUtil.writeTo does not update position consistently
...
+ Making position update properly
+ Marking internal method as private
2015-03-06 14:20:59 -07:00
a6a71e4705
458140 Added DispatcherType support to RewriteHandler
2015-03-06 16:29:54 +11:00
6ed01cf15b
removed ambiguous logging
2015-03-06 16:29:25 +11:00
44bb8cefa5
461133 - allow stop port to reuse address
2015-03-06 12:23:58 +11:00
2c3a4869ab
Tagged buffers from MappedByteBufferPool
2015-03-06 11:47:56 +11:00
a07ad75a6c
Adding license header
2015-03-05 12:55:58 -07:00
d582fdbd51
Adding code to simulate multiple (browser) clients against the proxy
2015-03-05 12:49:25 -07:00
32de92f7d6
Updating Leak Tracking/Detecting behavior.
...
+ NOISY logging is now configurable at JVM startup
+ Improved javadoc in LeakDetector
2015-03-05 11:06:15 -07:00
4396987ad5
Making LeakTracking more noisy
2015-03-05 11:06:15 -07:00
5d29e6f351
Making report of release but not acquired report stack
2015-03-05 11:06:15 -07:00
3b8bf8d1e0
Made test more stable.
2015-03-05 16:55:04 +01:00
a64368df5d
Factored into own method code executed after creation of a new
...
connection.
2015-03-05 16:54:44 +01:00
1a5346ec4f
461499 - ConnectionPool may leak connections.
...
Now using SpinLock to guard concurrent access to the connection
queues (idle and active) so that operations on them are atomic.
2015-03-05 15:42:03 +01:00
2169ea6494
further cleanups of HttpExchange
2015-03-05 19:50:30 +11:00
e8b6902b16
SpinLock cleanup of HttpExchange
2015-03-05 18:32:17 +11:00
9b307bc0ed
added leak detecting bufferpool to ssl tests
2015-03-05 16:55:10 +11:00
2619edff5a
461452 Double release of buffer by HttpReceiverOverHTTP
...
Clear buffer on exception to avoid release guard
2015-03-05 16:22:07 +11:00
042de2ec9f
461452 Double release of buffer by HttpReceiverOverHTTP
...
Updated HttpParse to always return from parseNext when messageComplete is called. This allows it to notice reentrant state changes
and removes the need for HttpReceiverOverHTTP.methodComplete to return true, thus avoiding the race.
2015-03-05 15:26:19 +11:00
418a60bbd0
461452 Double release of buffer by HttpReceiverOverHTTP
...
Updated comments to better describe the race to double release the buffer
2015-03-05 12:15:31 +11:00
8cbab09527
461452 Double release of buffer by HttpReceiverOverHTTP
...
This commit is just a tidy up of the code to reduce the size of the race causing this problem. It is not a fix.
2015-03-05 11:05:35 +11:00
40ad8dc608
Notifying the "failure" event for request and response when the failure
...
is detected during the processing of another event.
2015-03-04 01:01:09 +01:00
da80498c56
Improved logging.
2015-02-26 23:38:27 +01:00
bbc2d0c29d
Simplified abort logic.
2015-02-26 23:26:24 +01:00
31a0b36f3c
Improved logging and toString() implementations.
2015-02-26 23:26:24 +01:00
008b03ac78
460905 - Make sure TimeoutCompleteListener is cancelled if the request cannot be sent.
2015-02-26 12:25:42 +01:00
d5a6ad2345
460905 - Make sure TimeoutCompleteListener is cancelled if the request cannot be sent.
2015-02-26 10:56:19 +01:00
7b25674aca
Adding reference to patched versions of jetty-http jars
2015-02-25 16:10:06 -07:00
8402d8c100
Adding 9.3 to advisory
2015-02-25 07:46:55 -07:00
8efa8de39d
Update to advisory
2015-02-25 07:34:01 -07:00
cfca172dd6
Adding credits for discovery
2015-02-24 17:17:50 -07:00
1e502d4922
Adding advisory text
2015-02-24 14:43:59 -07:00
1659d83978
[maven-release-plugin] prepare for next development iteration
2015-02-24 10:38:33 -06:00
dc0c3f4c52
[maven-release-plugin] prepare release jetty-9.2.9.v20150224
2015-02-24 10:38:28 -06:00
12031d77af
set for release
2015-02-24 10:06:46 -06:00
33b5cc80bb
set for release
2015-02-24 10:06:13 -06:00
2141cc2dc3
459273 - Redundant license notices
...
+ Only showing license acknowledgement request screen
if [FILES] section contains content that does not (yet) exist
2015-02-23 16:49:07 -07:00
3e7b5f0fa9
460642 - HttpParser error 400 can expose previous buffer contents in HTTP status reason message
...
+ Simplifying behavior
2015-02-23 14:57:22 -07:00
4df5647f6d
460642 - HttpParser error 400 can expose previous buffer contents in HTTP status reason message
2015-02-23 14:44:53 -07:00
b69b8dfe50
460443 - Race condition releasing the response buffer.
...
Fixed by making sure that the response buffer is released at the
appropriate times.
2015-02-20 16:00:42 +01:00
ddde0db339
Added test to check that a 100 Continue response and a normal response,
...
when read in a single read by the client, are processed correctly.
2015-02-20 16:00:42 +01:00
98a7ad3cf7
460372 if web.xml does not contain jspc maven plugin insertionMarker behavior is wrong
2015-02-20 10:24:56 +11:00
5ecc9b8a36
Adding some tests to prove out inherited lifecycle listener behavior
2015-02-19 14:43:24 -07:00
0bbe17186e
460371 - AsyncMiddleManServlet.GZipContentTransformer fails if last transform has no output
...
+ Making fix simpler, as prior complexity was introduced to satisfy
different test case dealing with discarded content.
That test case was improperly testing for no decompressed content.
Once that assertion was corrected, the ultimate fix is easier.
2015-02-19 14:07:43 -07:00
0242c28bd9
460371 - AsyncMiddleManServlet.GZipContentTransformer fails if last transform has no output
...
+ Ensure that content that has been started to be gzipped is .close()
properly when the transform finished == true.
If no content has been gzipped, the .close() and .reset() are not
called, ever.
2015-02-19 12:17:17 -07:00
a0967a935f
Setting 9.2.9-SNAPSHOT
2015-02-19 10:21:13 -07:00
Simone Bordet
Joakim Erdfelt
Greg Wilkins
Jesse McConnell
Jan Bartel