← Back to team overview

openjdk team mailing list archive

[Bug 1512760] Re: Java applications leaks shmem chunks

 

Launchpad has imported 6 comments from the remote bug at
http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=2568.

If you reply to an imported comment from within Launchpad, your comment
will be sent to the remote bug automatically. Read more about
Launchpad's inter-bugtracker facilities at
https://help.launchpad.net/InterBugTracking.

------------------------------------------------------------------------
On 2015-07-30T12:56:42+00:00 Baiduzhyi-devel wrote:

Created attachment 1396
Patch to fix the bad merge.

Finally found the cause of the issue. Looks like a bad merge from 2
parents.

Revision 7726:4b26f93b23ba is the first to cause the crash.

$ hg log -r 7726:4b26f93b23ba
changeset:   7726:4b26f93b23ba
parent:      7506:6aaf00000bd5
parent:      7696:8dc56d0f3e86
user:        goetz
date:        Mon Jul 07 12:45:48 2014 +0200
summary:     Merge


Both parents, 7506:6aaf00000bd5 and 7696:8dc56d0f3e86, are working fine.

Upstream repo is different than icedtea7-forest, and contains working
code.

Please see the patch for the issue in attachment.

Reply at:
https://bugs.launchpad.net/ubuntu/+source/openjdk-7/+bug/1512760/comments/0

------------------------------------------------------------------------
On 2015-07-30T14:22:25+00:00 Andrew John Hughes wrote:

The changes from:

changeset:   7162:cd2d8f23e93c
user:        alitvinov
date:        Thu Dec 19 15:21:08 2013 +0400
summary:     8025775: JNI warnings in TryXShmAttach

have got lost in this merge. Re-applying this should fix it.

Reply at:
https://bugs.launchpad.net/ubuntu/+source/openjdk-7/+bug/1512760/comments/1

------------------------------------------------------------------------
On 2015-07-30T16:53:45+00:00 Mercurial wrote:

details: http://icedtea.classpath.org//hg/icedtea7-forest/jdk?cmd=changeset;node=a2334007f1a8
author: andrew
date: Thu Jul 30 17:52:32 2015 +0100

	PR2568: openjdk causes a full desktop crash on RHEL 6 i586
	Summary: Re-apply "8025775: JNI warnings in TryXShmAttach"; some changes lost in bad merge changeset 4b26f93b23ba

Reply at:
https://bugs.launchpad.net/ubuntu/+source/openjdk-7/+bug/1512760/comments/2

------------------------------------------------------------------------
On 2015-09-09T01:39:56+00:00 Mercurial wrote:

details: http://icedtea.classpath.org//hg/icedtea7?cmd=changeset;node=761344512703
author: Andrew John Hughes <gnu_andrew@xxxxxxxxxxxxxx>
date: Tue Sep 08 23:46:03 2015 +0100

        Bump to icedtea-2.7.0pre01.

	Upstream changes:
	  - OPENJDK7-01: Allow OpenJDK to build on PaX-enabled kernels
	  - OPENJDK7-02: Fix build where PAX_COMMAND is not specified
	  - OPENJDK7-03: Only apply PaX-marking when needed by a running PaX kernel
	  - OPENJDK7-04: Revert introduction of lambda expression in sun.lwawt.macosx.LWCToolkit
	  - OPENJDK7-05: Fix mistake in 8075374 backport
	  - PR2487: Make jdk8 mode the default for jdk.tls.ephemeralDHKeySize
	  - PR2501: libjavasctp.so doesn't need to link against libdl when linking against libsctp
	  - PR2502: Remove -fno-tree-vectorize workaround now http://gcc.gnu.org/PR63341 is fixed
	  - PR2503: Add existence check for all optional dependencies in jdk_generic_profile.sh
	  - PR2521: Systems with a GLib without libgio segfault when obtaining proxy information
	  - PR2568: openjdk causes a full desktop crash on RHEL 6 i586
	  - S7105883, PR2560: JDWP: agent crash if there exists a ThreadGroup with null name
	  - S7124253: [macosx] Flavor change notification not coming
	  - S8007219: [macosx] Frame size reverts meaning of maximized attribute if frame size close to display
	  - S8013581: [macosx] Key Bindings break with awt GraphicsEnvironment setFullScreenWindow
	  - S8014464: Update jcheck for OpenJDK 7
	  - S8016105, PR2560: Add complementary RETURN_NULL allocation macros in allocation.hpp
	  - S8020210: [macosx] JVM crashes in CWrapper$NSWindow.screen(long)
	  - S8021897, PR2560: EXCEPTION_ACCESS_VIOLATION on debugging String.contentEquals()
	  - S8022683, PR2560: JNI GetStringUTFChars should return NULL on allocation failure not abort the VM
	  - S8025922, PR2560: JNI access to Strings need to check if the value field is non-null
	  - S8027058: sun/management/jmxremote/bootstrap/RmiBootstrapTest.sh Failed to initialize connector
	  - S8027561: [macosx] Cleanup "may not respond to selector" warnings in native code
	  - S8028617: Dvorak keyboard mapping not honored when ctrl key pressed
	  - S8029868: Fix KSS issues in sun.lwawt.macosx
	  - S8042205: javax/management/monitor/*: some tests didn't  get all the notifications
	  - S8043201: Deprecate RC4 in SunJSSE provider
	  - S8043202: Prohibit RC4 cipher suites
	  - S8046817: JDK 8 schemagen tool does not generate xsd files for enum types
	  - S8048194: GSSContext.acceptSecContext fails when a supported mech is not initiator preferred
	  - S8048214: Linker error when compiling G1SATBCardTableModRefBS after include order changes
	  - S8064546: CipherInputStream throws BadPaddingException if stream is not fully read
	  - S8065764: javax/management/monitor/CounterMonitorTest.java hangs
	  - S8066952: [TEST-BUG] javax/management/monitor/CounterMonitorTest.java hangs
	  - S8067694: Improved certification checking
	  - S8068674: Increment minor version of HSx for 7u85 and initialize the build number
	  - S8071668: [macosx] Clipboard does not work with 3rd parties Clipboard Managers
	  - S8071715: Tune font layout engine
	  - S8071731: Better scaling for C1
	  - S8072490: Better font morphing redux
	  - S8072887: Better font handling improvements
	  - S8073334: Improved font substitutions
	  - S8073357: schema1.xsd has wrong content. Sequence of the enum values has been changed
	  - S8073385: Bad error message on parsing illegal character in XML attribute
	  - S8073773: Presume path preparedness
	  - S8073894: Getting to the root of certificate chains
	  - S8074098: 2D_Font/Bug8067699 test fails with SIGBUS crash on Solaris Sparc
	  - S8074297: substring in XSLT returns wrong character if string contains supplementary chars
	  - S8074312: Enable hotspot builds on 4.x Linux kernels
	  - S8074330: Set font anchors more solidly
	  - S8074335: Substitute for substitution formats
	  - S8074865: General crypto resilience changes
	  - S8074871: Adjust device table handling
	  - S8075374: Responding to OCSP responses
	  - S8075378: JNDI DnsClient Exception Handling
	  - S8075575: com/sun/security/auth/login/ConfigFile/InconsistentError.java failed in certain env.
	  - S8075576: com/sun/security/auth/module/KeyStoreLoginModule/OptionTest.java failed in certain env.
	  - S8075667: (tz) Support tzdata2015b
	  - S8075738: Better multi-JVM sharing
	  - S8075833: Straighter Elliptic Curves
	  - S8075838: Method for typing MethodTypes
	  - S8075853: Proxy for MBean proxies
	  - S8076290: JCK test api/xsl/conf/string/string17 starts failing after JDK-8074297
	  - S8076328: Enforce key exchange constraints
	  - S8076376: Enhance IIOP operations
	  - S8076397: Better MBean connections
	  - S8076401: Serialize OIS data
	  - S8076405: Improve serial serialization
	  - S8076409: Reinforce RMI framework
	  - S8077520: Morph tables into improved form
	  - S8077685: (tz) Support tzdata2015d
	  - S8078348: sun/security/pkcs11/sslecc/ClientJSSEServerJSSE.java fails with BindException
	  - S8078439: SPNEGO auth fails if client proposes MS krb5 OID
	  - S8078529: Increment the build value to b02 for hs24.85 in 8u85
	  - S8078562: Add modified dates
	  - S8078666: JVM fastdebug build compiled with GCC 5 asserts with "widen increases"
	  - S8080318: jdk8u51 l10n resource file translation update
	  - S8081386: Test sun/management/jmxremote/bootstrap/RmiSslBootstrapTest.sh test has RC4 dependencies
	  - S8081622: Increment the build value to b03 for hs24.85 in 8u51
	  - S8081775: two lib/testlibrary tests are failing with "Error. failed to clean up files after test" with jtreg 4.1 b12

        2015-09-08  Andrew John Hughes  <gnu.andrew@xxxxxxxxxxxxxx>

		* Makefile.am:
		(JDK_UPDATE_VERSION): Set to 85.
		(BUILD_VERSION): Set to b01.
		(CORBA_CHANGESET): Update to icedtea-2.7.0pre01.
		(JAXP_CHANGESET): Likewise.
		(JAXWS_CHANGESET): Likewise.
		(JDK_CHANGESET): Likewise.
		(LANGTOOLS_CHANGESET): Likewise.
		(OPENJDK_CHANGESET): Likewise.
		(CORBA_SHA256SUM): Likewise.
		(JAXP_SHA256SUM): Likewise.
		(JAXWS_SHA256SUM): Likewise.
		(JDK_SHA256SUM): Likewise.
		(LANGTOOLS_SHA256SUM): Likewise.
		(OPENJDK_SHA256SUM): Likewise.
		* NEWS: Updated.
		* configure.ac: Bump to 2.7.0pre01.
		* hotspot.map.in: Update to icedtea-2.7.0pre01.
		* patches/boot/ecj-diamond.patch:
		Regenerate and fix all current cases,
		including in MacOS X and Windows Java code.
		* patches/boot/ecj-multicatch.patch:
		Update cases in javax.crypto.CipherInputStream and
		sun.security.ssl.ECDHCrypt.
		* patches/boot/ecj-trywithresources.patch:
		Regenerated.

Reply at:
https://bugs.launchpad.net/ubuntu/+source/openjdk-7/+bug/1512760/comments/3

------------------------------------------------------------------------
On 2015-10-19T08:28:39+00:00 Mercurial wrote:

details: http://icedtea.classpath.org//hg/release/icedtea7-forest-2.6/jdk?cmd=changeset;node=e92ba87e90a2
author: andrew
date: Thu Jul 30 17:52:32 2015 +0100

	PR2568: openjdk causes a full desktop crash on RHEL 6 i586
	Summary: Re-apply "8025775: JNI warnings in TryXShmAttach"; some changes lost in bad merge changeset 4b26f93b23ba

Reply at:
https://bugs.launchpad.net/ubuntu/+source/openjdk-7/+bug/1512760/comments/4

------------------------------------------------------------------------
On 2015-10-20T03:47:11+00:00 Mercurial wrote:

details: http://icedtea.classpath.org//hg/release/icedtea7-2.6?cmd=changeset;node=723ef630c332
author: Andrew John Hughes <gnu_andrew@xxxxxxxxxxxxxx>
date: Tue Oct 20 04:30:23 2015 +0100

        Bump to icedtea-2.6.2pre02.

	Upstream changes:
	  - Bump to icedtea-2.6.2pre01
	  - Bump to icedtea-2.6.2pre02
	  - PR2512: Reset success following calls in LayoutManager.cpp
	  - PR2568: openjdk causes a full desktop crash on RHEL 6 i586
	  - S6880559, PR2674: Enable PKCS11 64-bit windows builds
	  - S6904403, PR2674: assert(f == k-&gt;has_finalizer(),&quot;inconsistent has_finalizer&quot;) with debug VM
	  - S7011441, PR2674: jndi/ldap/Connection.java needs to avoid spurious wakeup
	  - S7059542, PR2674: JNDI name operations should be locale independent
	  - S7105461, PR2571: Large JTables are not rendered correctly with Xrender pipeline
	  - S7105883, PR2560: JDWP: agent crash if there exists a ThreadGroup with null name
	  - S7107611, PR2674: sun.security.pkcs11.SessionManager is scalability blocker
	  - S7127066, PR2674: Class verifier accepts an invalid class file
	  - S7150092, PR2674: NTLM authentication fail if user specified a different realm
	  - S7150134, PR2571: JCK api/java_awt/Graphics/index.html#DrawLine fails with OOM for jdk8 with XRender pipeline
	  - S7152582, PR2674: PKCS11 tests should use the NSS libraries available in the OS
	  - S7156085, PR2674: ArrayIndexOutOfBoundsException throws in UTF8Reader of SAXParser
	  - S7177045, PR2674: Rework the TestProviderLeak.java regression test, it is too fragile to low memory errors.
	  - S7190945, PR2674: pkcs11 problem loading NSS libs on Ubuntu
	  - S8005226, PR2674: java/rmi/transport/pinClientSocketFactory/PinClientSocketFactory.java fails intermittently
	  - S8009438, PR2674: sun/security/pkcs11/Secmod tests failing on Ubuntu 12.04
	  - S8011709, PR2509: [parfait] False positive: memory leak in jdk/src/share/native/sun/font/layout/CanonShaping.cpp
	  - S8012971, PR2674: PKCS11Test hiding exception failures
	  - S8016105, PR2560: Add complementary RETURN_NULL allocation macros in allocation.hpp
	  - S8020424, PR2674: The NSS version should be detected before running crypto tests
	  - S8020443, PR2674: Frame is not created on the specified GraphicsDevice with two monitors
	  - S8021897, PR2560: EXCEPTION_ACCESS_VIOLATION on debugging String.contentEquals()
	  - S8022683, PR2560: JNI GetStringUTFChars should return NULL on allocation failure not abort the VM
	  - S8023052, PR2509: JVM crash in native layout
	  - S8025922, PR2560: JNI access to Strings need to check if the value field is non-null
	  - S8027058: sun/management/jmxremote/bootstrap/RmiBootstrapTest.sh Failed to initialize connector
	  - S8027624, PR2674: com/sun/crypto/provider/KeyFactory/TestProviderLeak.java unstable again
	  - S8033069, PR2674: mouse wheel scroll closes combobox popup
	  - S8035150, PR2674: ShouldNotReachHere() in ConstantPool::copy_entry_to
	  - S8039212, PR2674: SecretKeyBasic.sh needs to avoid NSS libnss3 and libsoftokn3 version mismatches
	  - S8042855, PR2509: [parfait] Potential null pointer dereference in IndicLayoutEngine.cpp
	  - S8044364, PR2674: runtime/RedefineFinalizer test fails on windows
	  - S8048194: GSSContext.acceptSecContext fails when a supported mech is not initiator preferred
	  - S8048353: jstack -l crashes VM when a Java mirror for a primitive type is locked
	  - S8050123, PR2674: Incorrect property name documented in CORBA InputStream API
	  - S8056122, PR1896: Upgrade JDK to use LittleCMS 2.6
	  - S8056124, PR2674: Hotspot should use PICL interface to get cacheline size on SPARC
	  - S8057934, PR1896: Upgrade to LittleCMS 2.6 breaks AIX build
	  - S8059200, PR2674: Promoted JDK9 b31 for Solaris-amd64 fails (Error: dl failure on line 744, no picl library) on Solaris 11.1
	  - S8059588, PR2674: deadlock in java/io/PrintStream when verbose java.security.debug flags are set
	  - S8062518, PR2674: AIOBE occurs when accessing to document function in extended function in JAXP
	  - S8062591, PR2674: SPARC PICL causes significantly longer startup times
	  - S8072863, PR2674: Replace fatal() with vm_exit_during_initialization() when an incorrect class is found on the bootclasspath
	  - S8073453, PR2674: Focus doesn't move when pressing Shift + Tab keys
	  - S8073773: Presume path preparedness
	  - S8073894: Getting to the root of certificate chains
	  - S8074098: 2D_Font/Bug8067699 test fails with SIGBUS crash on Solaris Sparc
	  - S8074297: substring in XSLT returns wrong character if string contains supplementary chars
	  - S8074330: Set font anchors more solidly
	  - S8074335: Substitute for substitution formats
	  - S8074350, PR2674: Support ISO 4217 "Current funds codes" table (A.2)
	  - S8074865: General crypto resilience changes
	  - S8074869, PR2674: C2 code generator can replace -0.0f with +0.0f on Linux
	  - S8074871: Adjust device table handling
	  - S8075374: Responding to OCSP responses
	  - S8075378: JNDI DnsClient Exception Handling
	  - S8075575: com/sun/security/auth/login/ConfigFile/InconsistentError.java failed in certain env.
	  - S8075576: com/sun/security/auth/module/KeyStoreLoginModule/OptionTest.java failed in certain env.
	  - S8075609, PR2674: java.lang.IllegalArgumentException: aContainer is not a focus cycle root of aComponent
	  - S8075667: (tz) Support tzdata2015b
	  - S8075738: Better multi-JVM sharing
	  - S8075773, PR2674: jps running as root fails after the fix of JDK-8050807
	  - S8075833: Straighter Elliptic Curves
	  - S8075838: Method for typing MethodTypes
	  - S8075853: Proxy for MBean proxies
	  - S8076040, PR2674: Test com/sun/crypto/provider/KeyFactory/TestProviderLeak.java fails with -XX:+UseG1GC
	  - S8076290: JCK test api/xsl/conf/string/string17 starts failing after JDK-8074297
	  - S8076328: Enforce key exchange constraints
	  - S8076397: Better MBean connections
	  - S8076409: Reinforce RMI framework
	  - S8076455, PR2674: IME Composition Window is displayed on incorrect position
	  - S8076968, PR2674: PICL based initialization of L2 cache line size on some SPARC systems is incorrect
	  - S8077102, PR2674: dns_lookup_realm should be false by default
	  - S8077409, PR2674: Drawing deviates when validate() is invoked on java.awt.ScrollPane
	  - S8077520: Morph tables into improved form
	  - S8077685: (tz) Support tzdata2015d
	  - S8078113, PR2674: 8011102 changes may cause incorrect results
	  - S8078331, PR1896: Upgrade JDK to use LittleCMS 2.7
	  - S8078348: sun/security/pkcs11/sslecc/ClientJSSEServerJSSE.java fails with BindException
	  - S8078439: SPNEGO auth fails if client proposes MS krb5 OID
	  - S8078529: Increment the build value to b02 for hs24.85 in 7u85
	  - S8078562: Add modified dates
	  - S8080012, PR2674: JVM times out with vdbench on SPARC M7-16
	  - S8080318: jdk7u85 l10n resource file translation update
	  - S8081386: Test sun/management/jmxremote/bootstrap/RmiSslBootstrapTest.sh test has RC4 dependencies
	  - S8081392, PR2674: getNodeValue should return 'null' value for Element nodes
	  - S8081470, PR2674: com/sun/jdi tests are failing with "Error. failed to clean up files after test" with jtreg 4.1 b12
	  - S8081622: Increment the build value to b03 for hs24.85 in 7u85
	  - S8081756, PR1896: Mastering Matrix Manipulations
	  - S8081775: two lib/testlibrary tests are failing with "Error. failed to clean up files after test" with jtreg 4.1 b12
	  - S8130297, PR2674: com/sun/crypto/provider/KeyFactory/TestProviderLeak.java still failing after JDK-8076040
	  - S8133636, PR2674: [TEST_BUG] Import/add tests for the problem seen in 8076110
	  - S8133966: Allow OpenJDK to build on PaX-enabled kernels
	  - S8133967: Fix build where PAX_COMMAND is not specified
	  - S8133968: Revert 8014464 on OpenJDK 7
	  - S8133970: Only apply PaX-marking when needed by a running PaX kernel
	  - S8133990: Revert introduction of lambda expression in sun.lwawt.macosx.LWCToolkit
	  - S8133991: Fix mistake in 8075374 backport
	  - S8133993: [TEST_BUG] Make CipherInputStreamExceptions compile on OpenJDK 7
	  - S8134248: Fix recently backported tests to work with OpenJDK 7u
	  - S8134610: Mac OS X build fails after July 2015 CPU
	  - S8134618: test/javax/xml/jaxp/transform/8062923/XslSubstringTest.java has bad license header

        2015-10-19  Andrew John Hughes  <gnu.andrew@xxxxxxxxxxxxxx>

		* Makefile.am:
		(CORBA_CHANGESET): Update to icedtea-2.6.2pre02.
		(JAXP_CHANGESET): Likewise.
		(JAXWS_CHANGESET): Likewise.
		(JDK_CHANGESET): Likewise.
		(LANGTOOLS_CHANGESET): Likewise.
		(OPENJDK_CHANGESET): Likewise.
		(CORBA_SHA256SUM): Likewise.
		(JAXP_SHA256SUM): Likewise.
		(JAXWS_SHA256SUM): Likewise.
		(JDK_SHA256SUM): Likewise.
		(LANGTOOLS_SHA256SUM): Likewise.
		(OPENJDK_SHA256SUM): Likewise.
		* NEWS: Updated.
		* configure.ac: Bump to 2.6.2pre02.
		* hotspot.map.in: Update to icedtea-2.6.2pre02.
		* patches/boot/ecj-stringswitch.patch:
		Add new case in sun.security.krb.Config.
		* patches/boot/ecj-trywithresources.patch:
		Regenerated due to changes in java.util.Currency.
		* patches/boot/ecj-underscored_literals.patch:
		Add new case in com.sun.jndi.ldap.Connection.

Reply at:
https://bugs.launchpad.net/ubuntu/+source/openjdk-7/+bug/1512760/comments/5


** Changed in: icedtea
       Status: Unknown => Fix Released

** Changed in: icedtea
   Importance: Unknown => Medium

-- 
You received this bug notification because you are a member of OpenJDK,
which is subscribed to openjdk-7 in Ubuntu.
https://bugs.launchpad.net/bugs/1512760

Title:
  Java applications leaks shmem chunks

Status in Iced Tea:
  Fix Released
Status in openjdk-7 package in Ubuntu:
  Confirmed

Bug description:
  After last update of openjdk few days ago system is going to be
  unsable, because any java application (JetBrains, IcedTea) leaks shmem
  chunks, which can be detected by "ipcs -m -p" command. This issues to
  unable to start any Qt application - vlc for example:

  QNativeImage: Unable to attach to shared memory segment. 
  (vlc:8541): Gdk-WARNING **: shmget failed: error 28

  Steps to reproduce:
  - run any java application
  - monitor shmem stats: ipcs -m -p | wc -l
  - try to run any Qt app 

  ~$ java -version
  java version "1.7.0_85"
  OpenJDK Runtime Environment (IcedTea 2.6.1) (7u85-2.6.1-5ubuntu0.15.04.1)
  OpenJDK 64-Bit Server VM (build 24.85-b03, mixed mode)

  ~$ lsb_release -a
  No LSB modules are available.
  Distributor ID: Ubuntu
  Description:    Ubuntu 15.04
  Release:        15.04
  Codename:       vivid

To manage notifications about this bug go to:
https://bugs.launchpad.net/icedtea/+bug/1512760/+subscriptions


References