Possible VM deadlock due to FileSystems.getDefault and System.loadLibrary interplay
Vitaly Davidovich
vitalyd at gmail.com
Thu Jan 4 02:26:18 UTC 2018
On Wed, Jan 3, 2018 at 8:36 PM David Holmes <david.holmes at oracle.com> wrote:
> On 4/01/2018 3:13 AM, Vitaly Davidovich wrote:
> > On Wed, Jan 3, 2018 at 12:00 PM, Alan Bateman <Alan.Bateman at oracle.com>
> > wrote:
> >
> >> The stack trace looks like JDK 8 or older. The initialization has
> changed
> >> quite a bit in JDK 9+ and would be interesting to see if you can
> duplicate
> >> it there.
> >>
> > Yes - it's 8u152 as I mentioned in the follow-up email (sorry, should've
> > mentioned that initially).
> >
> > Trying this on JDK9 is a bit tricky as the code (and its dependencies) in
> > question isn't easily runnable on 9 without quite a bit of work.
> >
> > Do you agree with the deadlock assessment though? That's what it looks
> like
> > to me but wanted to make sure I'm not missing anything.
>
> You're not missing anything. It's a class initialization related
> "deadlock".
>
> Thread A has called FileSystems.getDefault() which is doing <clinit> of
> DefaultFileSystemHolder, which in turn leads to Runtime.loadLibrary0
> which needs to lock the Runtime instance.
>
> Thread B is already doing a loadLibrary and holds the Runtime lock, the
> loadLibrary code then needs to do FileSystems.getDefault() which has to
> load and initialize DefaultFileSystemHolder, but that initialization is
> already in progress so internally the thread does a wait().
>
> So Thread B is waiting for Thread A to finish initialization, but holds
> the monitor lock that Thread A needs to finish the initialization.
> Deadlock.
>
> AFAICS this will still be possible in 9/10/11
That’s what I thought - thanks for confirming David. I’m slightly
surprised this hasn’t been reported before, although perhaps I’ve failed in
finding the report.
Does this deserve a bug report? I know Hotspot and some core classes in the
JDK do an elaborate dance to avoid classloading issues, but this is “user”
code. This has been worked around on our end by forcing the default FS to
load before this code runs. Perhaps something similar can be done as the
JDK bootstraps.
>
>
> Cheers,
> David
>
> > Thanks Alan
> >
> >>
> >> -Alan
> >>
> >>
> >> On 03/01/2018 16:56, Vitaly Davidovich wrote:
> >>
> >>> Hi all,
> >>>
> >>> Consider the following stacktraces of the main app thread and a
> background
> >>> thread:
> >>>
> >>> "main" #1 prio=5 os_prio=0 tid=0x0000000001bfd000 nid=0x1958 waiting
> for
> >>> monitor entry [0x00002b98ceba3000]
> >>>
> >>> java.lang.Thread.State: BLOCKED (on object monitor)
> >>>
> >>> at java.lang.Runtime.loadLibrary0(Runtime.java:862)
> >>>
> >>> - waiting to lock <0x00000007bf834b20> (a
> >>> java.lang.Runtime)
> >>>
> >>> at java.lang.System.loadLibrary(System.java:1122)
> >>>
> >>> at sun.nio.fs.UnixNativeDispatcher$1.run(
> >>> UnixNativeDispatcher.java:573)
> >>>
> >>> at sun.nio.fs.UnixNativeDispatcher$1.run(
> >>> UnixNativeDispatcher.java:571)
> >>>
> >>> at java.security.AccessController.doPrivileged(Native
> >>> Method)
> >>>
> >>> at sun.nio.fs.UnixNativeDispatcher.<clinit>(
> >>> UnixNativeDispatcher.java:571)
> >>>
> >>> at sun.nio.fs.UnixFileSystem.<ini
> >>> t>(UnixFileSystem.java:67)
> >>>
> >>> at sun.nio.fs.LinuxFileSystem.<
> >>> init>(LinuxFileSystem.java:39)
> >>>
> >>> at sun.nio.fs.LinuxFileSystemProvider.newFileSystem(
> >>> LinuxFileSystemProvider.java:46)
> >>>
> >>> at sun.nio.fs.LinuxFileSystemProvider.newFileSystem(
> >>> LinuxFileSystemProvider.java:39)
> >>>
> >>> at sun.nio.fs.UnixFileSystemProvider.<init>(
> >>> UnixFileSystemProvider.java:56)
> >>>
> >>> at sun.nio.fs.LinuxFileSystemProvider.<init>
> >>> (LinuxFileSystemProvider.java:41)
> >>>
> >>> at
> >>> sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native
> >>> Method)
> >>>
> >>> at sun.reflect.NativeConstructorA
> >>> ccessorImpl.newInstance(
> >>> NativeConstructorAccessorImpl.java:62)
> >>>
> >>> at sun.reflect.DelegatingConstructorAccessorI
> >>> mpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
> >>>
> >>> at java.lang.reflect.Constructor.
> >>> newInstance(Constructor.java:423)
> >>>
> >>> at java.lang.Class.newInstance(Class.java:442)
> >>>
> >>> at
> sun.nio.fs.DefaultFileSystemProvider.createProvider(
> >>> DefaultFileSystemProvider.java:48)
> >>>
> >>> at sun.nio.fs.DefaultFileSystemProvider.create(
> >>> DefaultFileSystemProvider.java:63)
> >>>
> >>> at java.nio.file.FileSystems$DefaultFileSystemHolder.
> >>> getDefaultProvider(FileSystems.java:108)
> >>>
> >>> at java.nio.file.FileSystems$DefaultFileSystemHolder.
> >>> access$000(FileSystems.java:89)
> >>>
> >>> at java.nio.file.FileSystems$Defa
> >>> ultFileSystemHolder$1.run(
> >>> FileSystems.java:98)
> >>>
> >>> at java.nio.file.FileSystems$Defa
> >>> ultFileSystemHolder$1.run(
> >>> FileSystems.java:96)
> >>>
> >>> at java.security.AccessController.doPrivileged(Native
> >>> Method)
> >>>
> >>> at java.nio.file.FileSystems$DefaultFileSystemHolder.
> >>> defaultFileSystem(FileSystems.java:96)
> >>>
> >>> at
> java.nio.file.FileSystems$DefaultFileSystemHolder.<
> >>> clinit>(FileSystems.java:90)
> >>>
> >>> at java.nio.file.FileSystems.getD
> >>> efault(FileSystems.java:
> >>> 176)
> >>>
> >>> at java.nio.file.Paths.get(Paths.java:84)
> >>>
> >>> at <some_package>.ServiceLoader.
> >>> setup(ServiceLoader.java:446)
> >>>
> >>> at <some_package>.MetadataAdminLoader.main(
> >>> MetadataAdminLoader.java:52)
> >>>
> >>>
> >>>
> >>>
> >>>
> >>> "shardDbExecutor-0-pool-0" #33 prio=5 os_prio=0 tid=0x0000000001ebf000
> >>> nid=0x1d01 in Object.wait() [0x00002b991f941000]
> >>>
> >>> java.lang.Thread.State: RUNNABLE
> >>>
> >>> at java.nio.file.FileSystems.getD
> >>> efault(FileSystems.java:
> >>> 176)
> >>>
> >>> at java.nio.file.Paths.get(Paths.java:138)
> >>>
> >>> at sun.misc.Launcher$ExtClassLoader.findLibrary(
> >>> Launcher.java:224)
> >>>
> >>> at java.lang.ClassLoader.loadLibr
> >>> ary(ClassLoader.java:1830)
> >>>
> >>> at java.lang.Runtime.loadLibrary0(Runtime.java:870)
> >>>
> >>> - locked <0x00000007bf834b20> (a java.lang.Runtime)
> >>>
> >>> at java.lang.System.loadLibrary(System.java:1122)
> >>>
> >>> at sun.security.ec.SunEC$1.run(SunEC.java:60)
> >>>
> >>> at sun.security.ec.SunEC$1.run(SunEC.java:58)
> >>>
> >>> at java.security.AccessController.doPrivileged(Native
> >>> Method)
> >>>
> >>> at sun.security.ec.SunEC.<clinit>(SunEC.java:58)
> >>>
> >>> at
> >>> sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native
> >>> Method)
> >>>
> >>> at sun.reflect.NativeConstructorA
> >>> ccessorImpl.newInstance(
> >>> NativeConstructorAccessorImpl.java:62)
> >>>
> >>> at sun.reflect.DelegatingConstructorAccessorI
> >>> mpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
> >>>
> >>> at java.lang.reflect.Constructor.
> >>> newInstance(Constructor.java:423)
> >>>
> >>> at java.lang.Class.newInstance(Class.java:442)
> >>>
> >>> at sun.security.jca.ProviderConfig$2.run(
> >>> ProviderConfig.java:221)
> >>>
> >>> at sun.security.jca.ProviderConfig$2.run(
> >>>
> >>> ProviderConfig.java:206)
> >>>
> >>> at java.security.AccessController.doPrivileged(Native
> >>> Method)
> >>>
> >>> at sun.security.jca.ProviderConfig.doLoadProvider(
> >>> ProviderConfig.java:206)
> >>>
> >>> at sun.security.jca.ProviderConfig.getProvider(
> >>> ProviderConfig.java:187)
> >>>
> >>> - locked <0x00000007bea03f48> (a sun.security.jca.
> >>> ProviderConfig)
> >>>
> >>> at sun.security.jca.ProviderList.
> >>> getProvider(ProviderList.java:233)
> >>>
> >>> at sun.security.jca.ProviderList.
> >>> getService(ProviderList.java:331)
> >>>
> >>> at sun.security.jca.GetInstance.
> >>> getInstance(GetInstance.java:157)
> >>>
> >>> at javax.net.ssl.SSLContext.getIn
> >>> stance(SSLContext.java:156)
> >>>
> >>> at com.microsoft.sqlserver.jdbc.
> >>> TDSChannel.enableSSL(IOBuffer.java:1658)
> >>>
> >>> at com.microsoft.sqlserver.jdbc.SQLServerConnection.
> >>> connectHelper(SQLServerConnection.java:1976)
> >>>
> >>> at com.microsoft.sqlserver.jdbc.S
> >>> QLServerConnection.login(
> >>> SQLServerConnection.java:1627)
> >>>
> >>> at com.microsoft.sqlserver.jdbc.SQLServerConnection.
> >>> connectInternal(SQLServerConnection.java:1458)
> >>>
> >>> at com.microsoft.sqlserver.jdbc.S
> >>> QLServerConnection.connect(
> >>> SQLServerConnection.java:772)
> >>>
> >>> at
> com.microsoft.sqlserver.jdbc.SQLServerDriver.connect(
> >>> SQLServerDriver.java:1168)
> >>>
> >>> at
> java.sql.DriverManager.getConnection(DriverManager.
> >>> java:664)
> >>>
> >>> at
> java.sql.DriverManager.getConnection(DriverManager.
> >>> java:208)
> >>>
> >>> at
> <some_package>.ConnectionPool.createGoodConnection(
> >>> ConnectionPool.java:565)
> >>>
> >>> at <some_package>.ConnectionPool.createNewConnection(
> >>> ConnectionPool.java:519)
> >>>
> >>> at <some_package>.ConnectionPool.
> >>> getConnection(ConnectionPool.java:407)
> >>>
> >>> at <some_package>.StatementCachingConnectionImpl
> >>> .setupConnection(StatementCachingConnectionImpl.java:94)
> >>>
> >>> at
> <some_package>.StatementCachingConnectionImpl.<init>(
> >>> StatementCachingConnectionImpl.java:54)
> >>>
> >>> at <some_package>.TestConnectionFactories$
> >>> FaultInjectingConnection.<init>(TestConnectionFactories.java:90)
> >>>
> >>> at <some_package>.TestConnectionF
> >>> actories.lambda$static$3(
> >>> TestConnectionFactories.java:33)
> >>>
> >>> at <some_package>.TestConnectionFactories$$
> >>> Lambda$12/1738236591.newConnection(Unknown Source)
> >>>
> >>> at <some_package>.SedaExecutor.
> >>> setupConnections(SedaExecutor.java:130)
> >>>
> >>> at
> <some_package>.SedaExecutor$SedaThreadFactory.lambda$
> >>> newThread$0(SedaExecutor.java:165)
> >>>
> >>> at
> >>> <some_package>.SedaExecutor$SedaThreadFactory$$Lambda$22/139
> >>> 2419022.run(Unknown
> >>> Source)
> >>>
> >>> at java.lang.Thread.run(Thread.java:748)
> >>>
> >>>
> >>> The application intermittently hangs on startup, and the above
> callstacks
> >>> are present. It seems like there's a deadlock due to concurrent class
> >>> loading and default FileSystem initialization (which itself triggers
> class
> >>> loading here).
> >>>
> >>> I wasn't able to find any existing JBS entries reporting anything
> similar.
> >>> Am I seeing this right? Is this a known issue?
> >>>
> >>> Thanks
> >>>
> >>
> >>
>
--
Sent from my phone
More information about the core-libs-dev
mailing list